London School of Economics

Monographs on Social Anthropology

Managing Editor: Peter Loizos

The Monographs on Social Anthropology were

established in 1940 and aim to publish results of
modem anthropological research of primary interest
to specialists.
The continuation ofthe series was made possible by
a grant in aid from the Wenner-Gren Foundation for
Anthropological Research, and, more recently, by a
further grant from the Governors of the London
School of Economics and Political Science. Income
from sales is returned to a revolving fund to assist
further publications.
The Monographs are under the direction of an
Editorial Board associated with the Department of
Anthropology of the London School of Economics
and Political Science.
 London School of Economics
Monographs on Social Anthropology
No63

Andre Beteille

Society and Politics in India:

Essays in a Comparative
Perspeaive
First published 1991 by The Athlone Press Ltd

Published 2020 by Routledge

2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
605 Third Avenue, New York, NY 10017

Routledge is an imprint of the Taylor & Francis Group, an informa business

© André Béteille 1991

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.

Notice:
Product or corporate names may be trademarks or registered trademarks,
and are used only for identification and explanation without intent to
infringe.
British Library Cataloging in Publication Data
A catalogue record for this book is available
from the British Library
Library of Congress Cataloging in Publication Data
Béteille, André
Society and politics in India: essays in a comparative
perspective / André Béteille.
p. cm. -- (Monographs on social anthropology: no. 63)
Includes bibliographical references (p. ) and index.
ISBN 0-485-19563-1 (cloth)
1. India—Social conditions—1947-2. India—Politics and
government—1947-3. Equality—India. 4. Individualism—India.
5. Socialstructures—India.6. Social values.I. Title
II. Series.
HN683.5.B37 1991
306'.0954—dc20
Typeset by Bookman Ltd, Bristol

ISBN 13: 978-0-4851-9563-7 (hbk)

 Contents

Acknowledgements vi

Introduction 1
1 Race, caste and gender 15
2 Race, caste and ethnic identity 37
3 The concept of tribe with special reference to India 57
4 Caste and politics in Tamilnadu 79
5 The politics of 'non-antagonistic' strata 122
6 Networks in Indian social structure 140
7 The future of the Backward Classes: the competing
demands of status and power 150
8 Equality as a right and as a policy 192
9 Individualism and equality 215
10 Individual and·person as subjects for sociology 250

Notes and sources 276

Index 303
 Acknowledgements
I have incurred many debts in writing the essays brought
together in this volume. The first among these is to my students
and colleagues in the Delhi School of Economics where I have
taught since 1959. This collection may serve as a record and
an acknowledgement of that long association. More recently,
the Department of Anthropology of the London School of
Economics has been a kind of second intellectual home for
me, and I gratefully acknowledge the friendship and hospitality
I have received there. The essays were put together while I was
a Fellow at the Institute for Advanced Study in Berlin; I am
grateful to the Institute for its congenial atmosphere and its
excellent facilities.
Chapter 1: Race, caste and gender A first draft of this paper was
prepared to be read at a number of American universities -
Duke University, the University of Chicago and the University
of California at Santa Barbara - whose hospitality I enjoyed as
a Fulbright Distinguished Lecturer in May 1989; I would like
to thank the US Educational Foundation in India for providing
me with the opportunity to travel in the United States. The
paper was revised for publication at the Wissenschaftskolleg zu
Berlin where I was a Fellow in 1989-90; I would like to thank
the Kolleg for its generous hospitality, and in particular Peter
Burke, Robert Damton and Esther Goody for their stimulating
comments.
Chapter 2: Race, caste and ethnic identity I am grateful to my col-
leagues A. Sharma and S.C. Tiwari of the Department of Anthro-
pology and M.S.A. Rao of the Department ofSociology, University
of Delhi, for much help in the preparation of this chapter.

Chapter 3: The concept of tribe with special reference to India

This paper was prepared for a seminar in the Department
of Anthropology of the London School of Economics where
I was an Academic Visitor during the Summer term of 1986.
 Acknowledgements vu

I would like to thank all the members of the Department, and

in particular Maurice Bloch, Chris Fuller, Alfie Gell and Johnny
Parry for their friendship and their intellectual stimulation.
Chapter 5: The politics of 'non-antagonistic' strata An earlier
version of this paper was prepared in May 1966 for a seminar
at the Centre of Indian Studies at the Ecole Pratique des
Hautes Etudes. It appears here substantially revised and I am
grateful to the Jawaharlal Nehru Memorial Fund for providing
me with a Fellowship which gave me the time and leisure to
prepare it for publication.
Chapter 7: the Future of the Backward Classes I am grateful to
Professor M.N. Srinivas for having allowed me to make use
of his unpublished Tagore Lectures from which some of the
basic concepts employed here have been drawn. Although I
have not made specific acknowledgements at every point, my
debt to him will be recognized by anyone who is familiar with
Indian sociology.
Chapter 9: Individualism and equality The argument of this essay
was first presented to a small group at the University of Leiden
in April 1984 and shortly thereafter to similar groups at Utrecht
and Oxford. It was written up in roughly its present version
for a seminar at the London School of Economics in May
1985, and the paper was presented later in the same month
at Manchester and at Leeds. I am grateful to the members of
my various audiences for their many helpful comments.
Acknowledgements are gratefully made for permission to repub-
lish copyright material: Chapters 1·and 6 first appeared in Man,
in 1990 and 1964 respectively; Chapter 2 in International Social
Science Journal, 1971; Chapter 3 in European Journal ofSociology
[Archiv. europ. sociol] 1986; Chapter 4 in Castes Old and New,
1969, Asia Publishing House; Chapter 5 in Contributions to
Indian Sociology, 1969; Chapter 7 in Indian Journal of Public
Administration, 1965; Chapter 8 in London School ofEconomics
Quarterly, 1986; Chapter 9 in Cu"entAnthropology, 1986.
For M.N. Srinivas, on his seventy-fifth birthday
 Introduaion
The essays brought together in this collection were written or
published between 1964 and 1990. A quarter-century is not a
small span in the professional career of a single individual. In
this case it was also a period of many changes in the disciplines
of social anthropology and sociology, as well as in the social
and political environment to whose understanding the essays
are mainly devoted. I thought it might be of. interest to use
the work of a single individual to recall and recount some of
these changes. Hence I have included some early essays even
though the volume might have had an appearance_ of greater
unity if I had put in their place some more recent ones not
included here.
I would not like to claim for the collection any substantial
unity in terms of theory, argument and conclusion. Perhaps to
an even greater extent than in other such collections, this one
has neither a beginning nor an end. The work that it reports
upon was taken up in the middle of a growing discipline, and
it continues in the middle of the effort to understand social and
political processes in a state of flux. Such unity as the volume
has derives from the selection of particular problems, and my
responses to those problems as well as to other ways of looking
at them. I have deliberately avoided arranging the essays in a
chronological order since I believe that, although my answers
to particular questions might have changed in some cases, the
orientation to thein remains basically the same.
Society and politics are subjects of continuous and animated
discussion in contemporary India. Indians are eloquent speakers
and there is a great deal of writing in newspapers, magazines
and books about the past, present and future of Indian society
and politics; it should be stressed that sociologists and social
anthropologists, or even social theorists in the wider sense, are
not the only ones who think and write on these subjects. A
number of changes in the intellectual climate of India set in
2 Society and Politics in India
around 1977, leading more and more persons to question the
value of existing theoretical approaches to the Indian reality
and to search for alternative approaches to it. The anxieties of
Indian intellectuals about their framework of understanding and
explanation reflect to some extent their anxieties about society
and politics in contemporary India.
It is impossible for a student of society and politics to live
and work in India without responding at all to the changes in
the general climate of opinion taking place in the country; and
I have expressed myself periodically in articles in newspapers
and magazines. But the essays here are not presented as instant
political commentary or reflections on current affairs. I view
them rather as contributions to a discussion that is shaped
primarily by the demands of sociology and social anthropology
as academic disciplines, with occasional excursions into related
disciplinary domains such as political theory, law and history.
An important difference between what is offered as political
commentary and the essays presented here is the extensive use of
the comparative method, or at least ofcomparisons, in each one ·
of the latter. I have commented briefly on the use and abuse of
the comparative method in the essay on race, caste, and gender
in this volume, and more extensively in a recent essay devoted
to the subject but not included here. 1 Suffice it to say that
although there are, and will continue to be, disagreements on
the procedures to be used and the precautions to be taken in
making comparisons, there is a more fundamental question of
orientation. Here I derive my orientation from Durkheim who
said, 'Comparative sociology is not a special branch of sociology;
it is sociology itself'. 2 I do not believe that we should refrain from
making comparisons or seeking illumination from them until the
perfect method has been devised on whose soundness we will all
be in full agreement.
A second important difference between commentaries on
social and political processes and the approach adopted in these
essays lies in the more or less consistent use I have tried to make
of the concept of social structure. When I began my career as a
social anthropologist, the notion still had a prominent, not to say
a unique, place in social anthropology, particularly in the form
given to it by Radcliffe-Brown, Fortes, and also Evans-Pritchard
 lntroduaion 3

in his earlier writings. That concept had been greatly influenced

by Durkheim's notion ofsocial morphology. It has now gone out
of fashion, having been largely superseded by other concepts of
structure that derive more from Durkheim's views on collective
representations than those on social morphology. But I have
retained some attachment to the old idea, particularly as used in
Evans-Pritchard's first book on the Nuer and in African Political
Systm,s, partly out of inertia, but I hope not entirely out of a lack
· of good sense.
Social anthropology in the 1950s and 1960s was different in
many ways from what it is today. Despite its fascination for
the comparative method, the anthropology of Radcliffe-Brown
- and also of Fortes and Evans-Pritchard - was firmly rooted
in the study of primitive or pre-literate societies, or at least
of societies that differed markedly from those to which the
anthropologist belonged. In Britain and in Western countries
generally, it was accepted by and large that if a scholar studied
other societies and cultures, he was an anthropologist, whereas
if he studied his own or his own type of society, he was a soci-
ologist. This I regarded as a challenge and an opportunity, for
I wanted to be a sociologist, although my intellectual equipment
derived from the heritage of anthropologists such as Rivers,
Lowie, Radcliffe-Brown, Malinowski and Evans-Pritchard.
Thirty years ago one could work from the assumption of the
unity of sociology and social anthropology, if not as an estab-
lished fact, at least as a realizable possibility. The relationship
between the two disciplines now appears, at least to me, to be
more problematic, for they have both changed. Anthropology as
a whole is now concerned much more with the study of culture
than of social structure, with the 'thought-out' rather than the
'lived-in' order; and the most prominent anthropologist of the
1970s and 1980s has reiterated the commitment ofhis discipline
to 'the view from afar'.3 Perhaps I have worried a little more ·
than others in my position about the implications of this shift
in perspective for the identity of the anthropologist engaged in
the study of his own society.4

The unity of sociology and anthropology, the definition of social

anthropology as comparative sociology, and the primacy of the
4 Society and Politics in India
concept of social structure were all more or less articles of
faith in the new department of sociology that I joined in the
University of Delhi in 1959.5 These convictions were strongly
held and expressed by M.N. Srinivas, the first professor in the
department who had worked in Oxford with Radcliffe-Brown
and Evans-Pritchard both ofwhom he admired equally. 6 He had
already made a mark in the field of Indian studies, and his work
played a large part in shaping the field in the 1960s.
Until the 1950s Indian anthropology was largely identified
with tribal studies. In the four years that I spent as a student of
anthropology in the University of Calcutta, I read innumerable
monographs on tribes: the Todas, the Andaman Islanders, the
tribes of Chotanagpur and those of the north east. There were,
of course, notable exceptions, 7 but popular belief still viewed
anthropology as 'the study of oddments by eccentrics'. I was
determined to move away from the study of tribes and on to
areas considered·more central, although I have, through various
routes, returned to their study from time to time, as can be seen
from two of the essays in the present volume.
The new field that was opening up in Indian anthropology in
the 1950s may be described as 'village studies' as against 'tribal
studies'. Two important collections of essays were published
in 1955, India's Villages, edited by M.N. Srinivas, and Village
India, edited by McKim Marriott. 8 These and subsequent
studies explored the different domains of Indian society, such
as kinship, religion, economics and politics through the intensive
study ofsingle communities. I began with a case study of a village
in Tanjore district in south India where the focus of attention
was social stratification, till then not a subject of central concern
among anthropologists. 9
As the collection by Marriott in particular showed, an impor-
tant shift had come about in the orientation of American
anthropology which made a strong impact on Indian studies.
The moving spirit behind it was Robert Redfield whose work
opened the way for the anthropological study of civilizations
through detailed investigations of peasant societies and cultures.
Anthropologists could not only move out of tribal studies, but
they had found a way of exploring very large and very complex
systems through the kind of case study that they had made their
 lntroduaion 5

own. Village studies began to be made all over the world, and
these gave a new lease of life to anthropology in India.
Srinivas, who played a leading part in developing village
studies in India, was on the whole sceptical of the approach
of Redfield which he contrasted unfavourably with that of
Radcliffe-Brown. He felt that Radcliffe-Brown's concept of
structure was superior in clarity and rigour to Redfield's concept
of culture, and that view (or prejudice) made its way into my
work, as can be seen in several of the essays, particularly the
earlier ones. However, analytical rigour was not Srinivas' strong
suit; his strength lay rather in his sensitive imagination and his
unerring instinct for the ambiguities in a social situation, or what
he called its 'messiness'. My deep suspicion of the modem
'structuralist' approach, implicit or explicit in many ofthe essays
here, derives partly from this and partly from my own experience
of everyday life in the society in which I live and work.
Perhaps even more than the concept of social structure,
Srinivas valued fieldwork as the basis for understanding society.
Fieldwork is, of course, the trademark of the professional
anthropologist, and its value was underlined for me even
before I came to Delhi by my teachers of anthropology in
Calcutta. There its exemplar was N.K. Bose who was a man
of enormous intellectual vitality, largely self-taught, but with a
command over an astonishing range of subjects. Later, I was
able to see the difference between his approach to fieldwork - as
to many other matters - and that of Srinivas. Srinivas' approach
conformed more to the orthodoxy then prevalent in Britain,
particularly Oxford, which required the anthropologist to spend
twelve to eighteen months at a stretch in the intensive study of a
single community. Bose did not feel obliged to conform to any
orthodoxy, and developed an approach to fieldwork that may be
called 'extensive' rather than 'intensive'.1° He made brief but
repeated visits to several communities, and did this throughout
his career, collecting a large body of information from different
parts of the country. He felt that this was a better strategy for the
anthropologist engaged in the study ofhis own society which was
in many ways more easily accessible to him than to an outsider.
Bose and Srinivas were both superb fieldworkers, but I,
unfortunately, failed to become one. For Srinivas, fieldwork not
6 Society and Politics in India
only provided reliable information, it opened up a new view of
society, and that was its most significant contribution. Srinivas,
Bose, and others among my teachers were not only professional
anthropologists, they were heirs to an intellectual tradition
that reached far beyond Evans-Pritchard, Radcliffe-Brown,
Malinowski and Rivers. They were proud of that tradition,
but also reacted against it. The Brahminical tradition had many
achievements in mathematics, logic and grammar, and in several
branches of speculative thought; it had little to show for itselfin
history and geography, and in branches ofleaming that required
careful observation and exact description. Both Srinivas and
Bose felt that fieldwork was essential to correct the biases of
their own intellectual tradition.
It is in this light that we must view the great significance ofand
the initial enthusiasm for the distinction proposed by Srinivas
between the 'field-view' and the 'book-view' of Indian society.
There were book-views of every major institution: of caste, of
the joint family and of the village community. Srinivas attacked
these views as being out of date, out of touch with reality, and
mistaken. He argued for their replacement by accounts based
on fieldwork that would reveal the dynamic tensions in their
actual operation. As I have indicated, this was largely a reaction
against the Brahminical bias for the 'thought-out' as against the
'lived-in' order; but the reaction was fuelled by the antipathy,
derived from Radcliffe-Brown, towards 'conjectural history'.
I would like to underline the point made in the concluding
section of the essay on race, caste and gender that the field-view
entails more than the use of one's own fieldwork as the main
basis of analysis. There are in any case strict limits to the extent
to which that can be done in the study oflndian society. Indeed,
it should be possible, at least in principle, to adopt that view
in preference to the 'book-view' without oneself undertaking
fieldwork in the strict sense. The field-view, as I understand
it, is at bottom an orientation to the lived experiences of people,
with all their inner tensions and contradictions, that one seeks
to understand and interpret.
The basic orientation of the essays in this volume derives
from the field-view in the extended sense that I have given
to the term. Their focus of attention is the lived experience
 Introduaion 7
of Indians here and now: they are not views from afar. But,
I must insist, they are not instant political commentary either.
I have tried to apply certain methods and concepts developed
in social theory to re~ch a clearer understanding of experiences
that take shape in Indian society every day. And I have tried to
compare them with experiences in otlter societies to make that
understanding more complete.

The great wave of enthusiasm for the 'field-view' that had

animated anthropological studies in India in the first two
decades after independence began to decline towards the end
of the 1960s. The focus of attention shifted away from the
social and political processes of contemporary life to structures
of thought and representation. For me this world-wide change
in anthropological orientation was dramatized by the eclipse of
Radcliffe-Brown (and likewise Redfield), and the ascendancy of
Levi-Strauss. It gradually made its impact on the anthropology
of India. I was young enough to get some sense of the
significance of the change when it was coming, but did not
yield as readily to its temptation as did some others.
Radcliffe-Brown's concept of social structure, by which
Srinivas had set so much store, was now replaced by a new
and very different concept of structure. I I This new concept
drew its inspiration from linguistic analysis whereas the old one
had been based on a biological metaphor. The earlier idea of
social structure, which goes back to Durkheim's concept ofsocial
morphology, still retains some attraction for me, partly because of
the great importance in Indian society ofcollective identities, and
ofthe arrangements ofgroups that retain a degree ofconsistency
and constancy. At the same time, the shift away from morphology
was welcome as it brought the study of ideas and values, or of
collective representations, to the centre of attention.
The new trend in the anthropological study of India was
heralded by the work of Louis Dumont. Dumont had published
an outstanding monograph in French on the Pramalai K.allar,
but it did not make an immediate impact, partly because it
was not available in English until recently. I 2 But his book,
Homo hierarchious, which first appeared in 1966, became soon
available in English, and made an impact almost at once
8 Society and Politics in India
(I wrote a very early response to Dumont's approach in a
paper presented in Paris in 1966 which appears here as 'The
politics of "non-antagonistic strata"'). Even more important in
creating and sustaining an interest in Dumont's approach was
the journal, Contributions to Indian Sociology, launched by him (in
association with D.F. Pocock) in 1957.
It is through Dumont's work that the new concept of struc-
ture, as against the older one of 'social structure', made its
appearance in the social anthropology of India. Dumont had
been directly influenced by Levi-Strauss, as is evident from the
dedication of the monograph, Hierarchy and Marriage Alliance
in South Indian Kinship 13 to him and the acknowledgement
to Levi-Strauss for his idea of structure in the opening pages
of the monograph on the Pramalai Kallar, 14 both published in
1957. It took anthropologists of India, accustomed to the staple
fare of British and American anthropology (whether inspired by
Radcliffe-Brown or Redfield), a little time to realize that a major
change of orientation had been introduced with the new concept
of structure.
The early numbers of Contributions to Indian Sociology 15
defined the focus of study in terms of religion and kinship
rather than economics and politics. They also called for a
closer relationship between the sociology of India and Indology,
as based on the study of classical texts. Thus, the book-view
of society or the indological approach was reintroduced as an
integral part of the study of Indian society. However, unlike in
the past, this time it was informed by ethnographic fieldwork,
sometimes of a very high quality. But, despite the very high value
placed on fieldwork, it was from the structures of traditional
Indian thought as described in the classical texts that this work
appeared to take its orientation.
Like others before and after him, Dumont finds in the caste
system the defining features of Indian society, but views it first
and foremost as a system of ideas and values. Further, he argues
that the principal task of identifying the systemic properties of
caste had been performed two thousand and more years ago
by the authors of the Hindu Dharmashastras, and that we
should take their representation of the system as our point of
departure.
 lntroduaion 9

Yet on certain points we shall take the liberty of completing and

systematizing the indigenous and orthogenic theory of caste - not
without employing empirical aspects in a secondary capacity - by
postulating that men in society behave in a coherent and rational
manner, especially in such an important matter, and that it is
possible to recover the simple principle of their thought. 16
It would be a serious mistake to regard Dumont as simple-
minded, as might well appear from the quotation above. Homo
hierarchicus is a magisterial work, impressive alike for its tone of
authority and its richness of fact and argument. It made a great
impact ,on European and American anthropologists of India,
and gradually also on my own students and younger colleagues;
but it did not alter very substantially the perceptions of Indian
sociologists in the rest of the country. As for myself, I was more
impressed than convinced by the body of Dumont's work.
Let me say at once that I find it impossible to believe that
Indians, or any others, for the most part 'behave in a coherent
or rational manner': life would be enormously more simple - and
more dull - ifthey did so. Anyone looking at Indian society today
is likely to believe that their manner of behaviour is closer to the
opposite. It can ofcourse be argued that Dumont,s observations
apply to traditional and not contemporary India. But I would
find such an argument unconvincing, either because it cannot be
easily proved true, or because, even if true for traditional India, it
may not be relevant to contemporary India, or for both reasons.
Preoccupation with the Hindu categories of thought has
figured prominently also in the recent writings of a number
of American anthropologists of the University of Chicago,
or trained in that university. Most influential among them
is Professor Marriott whose ideas have made an impact as
much through his own writings as through the work of his
many able pupils. 17 Marriott began his work on India under
the influence ofRedfield, but subsequently changed its direction
to address himself more specifically to the distinctive features
of Hindu thought with special attention, as I understand it,·to
Hindu systems of classification. The term 'ethnosociologf has
been used to characterize the work of the Chicago school of
anthropologists; ts
IO Society and Politics in India
The preoccupation with modes of thought, systems of clas-
sification, and so on has been associated with the attempt to
define the unique, invariant and unchanging properties ofHindu
culture, with special emphasis on traditional Hindu culture.
Such an attempt has had· attractions, though not exactly of the
same kind, for both Indian and Western scholars. But it has also
had the unfortunate effect, at least from the sociological point of
view, of exaggerating the contrast between contemporary India
and other modem societies. India has emerged as the repository
of unchanging cultural values in a world in which change is
conspicuous everywhere. The method adopted is more a method
of contrasts than the comparative method as I understand it. 19
The anthropological approach has acquired a certain influ-
ence in contemporary intellectual life, but in the process it
has shifted its attention somewhat from social structure, social
institutions and social processes to modes of thought, systems
of classification, and the symbolic order in general. Through
this change, however, it has retained or even strengthened
the definition of the object of its study as the Other. 20 This
has led to a peculiar tension among anthropologists in India:
their theoretical concerns tend to be determined by the general
intellectual ambience of their discipline; at the same time, it is
difficult to consistently play the. part of 'the· astronomer of the
social sciences' while studying one's own society and culture.
Since I maintain that my position as a sociologist of India
must be in some sense different from that of 'the astronomer
of the social sciences', I must indicate, however briefly, the
place I assign to moral judgements in my kind of work. I do
not believe that sociology, as I view it, has much scope to
become a 'policy science'; my tasks are mainly confined to
description, comparison, interpretation and analysis, as against
prescription. 21 There is, however, one essay, 'Equality as a right
and as a policy', that touches upon questions of policy, but it
is mainly an account of the social and poµtical conditions out
of which certain policies arise, and of their intended as well as
actual consequences; it is an exercise in policy analysis rather
than policy prescription.
 Introduaion 11
It now remains for me to make a few brief comments on the
essays included her~ in the light of the general observations
made above. The first two essays on race and caste deal with
the same subject, but not quite in the same way. (I was surprised
to find how much they have in common, despite the interval of
nearly twenty years in their writing.) They are both comparative
in outlook, 'Race, caste and gender' being more self-consciously
so than 'Race, caste and ethnic identity'. I have argued that much
can be learnt about both caste and race by comparing the two,
and I now believe that our insight into each and both can be
deepened by a consideration of gender and its place in societies
divided by race or by caste.
Caste and race are about inequality, a subject in which I
have had a long-standing interest, but they are both also about
collective identity. Collective identities are of great importance
in India, and their place in society and the part they play in
politics are recurrent themes in this collection. In the essay,
'Race, caste and ethnic identity', I try to give a brief account of
the different forms taken by collective identities in India, using
the morphological concept of structure as my point of departure;
that approach figures as the point of departure in several other
essays as well.
While groups having a 'high degree of consistency and con-
stancy' are particularly salient features of the social morphology
of India, it would be a mistake to regard them as being unique
to it or as being absent from advanced industrial societies.
Collective identities based on race and ethnicity are important
features of many societies, and there is no doubt that some
individuals suffer disadvantages while others enjoy advantages
by virtue of their ascribed membership in particular groups in
the United States as well as in India; the 'group-disadvantaging
principle' and its basis in social morphology are very briefly
discussed in 'Equality as a right and as a policy', a lecture
delivered in the London School of Economics in the summer
of 1986.
In the essay on 'The concept of tribe', prepared for a seminar,
also in the London School of Economics in 1986, I examine
the special significance of collective identities in the social
morphology of· India, past and present. The morphological
12 Society and Politics in India
approach has been used to good advantage in a recent work
on Indian society and politics by a French social and political
theorist. 22 My own approach in this paper is of somewhat older
vintage. It draws its basic inspiration from a paper published in
1941 by N.K. Bose23 who had devoted a considerable part of
a lifetime to tribal studies; the point on which I depart from
Bose is in the use of the comparative approach, but that, in my
judgement, only confirms his basic insight.
'Caste and politics in Tamilnadu', written shortly after my
monograph on a Tanjore village, is an attempt to interpret
regional political processes in the light of my fieldwork experi-
ence. Political anthropology had dwelt largely on politics in
small-scale societies or at the local level, whereas political
science, with a different disciplinary orientation, had been
concerned more with the state and its institutions. In my
essay I sought to examine the linkages between processes at
different levels of territorial organization. Collective identities
were again important in examining problems ofboth distribution
and process, and I argued that even the identities ofcaste did not
remain the same from one political context to another, or from
one territorial level to another.
'The politics of "non-antagonistic strata"' was written for and
first presented to Professor Louis Dumont's seminar in Paris in
May 1966. It sought to challenge the view that castes ceased
to be castes when they organized themselves for competitive
politics. I found that view to be out of tune with what was
happening in India. Many Indians, of both liberal and radical
persuasions, believed in the 1950s that caste was on its way out
and that it would be swept away by economic development and
democratic politics. By the 1960s it had become clear that caste
had been given a new lease of life by electoral politics and was,
in the process, undergoing important changes. While electoral
politics had given a new tum to the relations between castes,
I found it an unnecessary limitation of our view of traditional
Indian society to exclude in principle any competition or conflict
between its basic morphological constituents. As an extreme
point of comparison, I asked how far we could reach in an
understanding of Soviet society if we stopped short at the then
official doctrine that the relations between the social strata that
 Introduction 13
were its basic constituents were-in principle 'non-antagonistic'.
The essay, 'Networks in Indian social structure', written
jointly with Srinivas and published a year before my Tanjore
monograph, sought to examine the limitations of a morpho-
logical approach in which attention was confined exclusively
to enduring groups and the enduring relations between them.
It drew attention to the limitations of such an approach in the
study of a large, complex and changing society in which the
choices available to the individual were steadily expanding,
despite the continued existence of groups with more or less
clear boundaries.
I have included in this selection two essays on the Backward
Classes to whose study I have devoted much of my attention
in recent years. There is a vast literature on the subject -
descriptive, analytical and prescriptive - to which scholars in
many disciplines· have contributed. In my larger work, I have
tried to use the resources of both legal and social theory to
understand and interpret the changing place of this large and
very important constituent of Indian society.
The essay on the future ofthe Backward Classes is one of the
earliest, being concerned with the two related issues of social
stratification and social mobility. The approach is directly socio-
logical, showing in particular the influence of Max Weber in the
basic distinction it makes between status and power, and in its
analysis of the tensions between their respective demands. The
second and much later essay, 'Equality as a right and as a policy',
shows my more recent and still somewhat uncertain interest in
legal theory. I had started with stratification or inequality, and its
different forms or dimensions; it took me close to two decades to
recognize that equality as well could be conceived in more ways
than one, and that there were fundamental tensions between the
different conceptions of equality.
The two last essays deal more with ideas, beliefs and values
than social structure in the morphological sense. Equality is
more an ideal than a fact; or, rather, it is a social fact in so far
as it is collectively acknowledged as an ideal and a value. To the
sociologist, what is interesting about equality is that it does not
have the same force as a social ideal in all places or at all times.
This has led some scholars to categorize whole societies as either
14 Sodety and Politics in India
egalitarian or hierarchical, a proceeding I have criticized in detail
elsewhere. 24
'Individualism and equality' deals with the relationship be-
tween the two values comparatively and historically. The focus
of the essay is on equality, which I insist is a complex and not
a simple idea. It is not enough to say that it goes hand in hand
with individualism, for, as I argue, following Simmel, there is
an 'individualism of inequality' as well as an 'individualism of
equality'. At the same time, and with all its inherent tensions,
a
equality is powerful ideal in the contemporary world as a
whole, and no longer only in the West. My essay evoked a sharp
response from Professor Dumont who has written authoritatively
on the subject; I have reproduced my reply to Dumont because
in it I develop a little further the arguments presented in my
original essay.
The focus in the last essay is not on equality or inequality,
as in much of my other work, but on individual and person.
These are difficult concepts whose meaning and significance
have varied enormously from one society to another; yet the
individual, at least as citizen, occupies an important place
in all modem societies. My essay has two main objectives:
(1) to point out how difficult it is to reach agreement on the
significance of these ideas, rich in ambiguity as they are, when
societies of different types are being· compared; and (2) to
question the conventional view that they have had little meaning
or significance outside Western history and culture. I develop
my argument through a critique of the celebrated essay on the
person by Marcel Mauss. Thus, I end as I began, with the plea
for a differentiated view of each and every society as a basis for
the comparisons and contrasts we make between them.
 1 Race, caste and gender
Any attempt today to bring together race and caste for compari-
son and contrast is likely to meet with a cold reception. Such an
attempt invites the opprobrium specially reserved for positivism,
empiricism and eclecticism by the theoretically well tuned. They
will readily acknowledge the similarities between caste and race
when they are pointed out; what they will deny is that these
.similarities can have much significance for the understanding
at least of caste. It may be safely said that, although the subject
of caste has been discussed threadbare by students of Indian
society and culture, the comparison with race has hardly figured,
if at all, in the last twenty to twenty-five years.
Yet the fruitfulness ofcomparing race with caste was taken for
granted by American and other sociologists studying the 'Negro
problem' in the United States in the 1930s, 1940s and 1950s.
The pioneer in this regard was Lloyd Warner (1936) who wrote
about caste and class in the United States, saying that it was
more appropriate to describe blacks and whites as castes than
as races or classes. Warner directed and inspired a number
of monographic studies of what came to be known among
sociologists as the problem of caste in the US South (Davis,
Gardner and Gardner, 1941; Cayton and Drake, 1945). The
psychologist John Dollard used Warner's conceptual scheme in
his outstanding monograph, Caste and Class in a Southern Town.
The major work of the 1940s on the blacks in the United
States was Gunnar Myrdal's An American Dilemma. Myrdal,
too, used the same conceptual scheme as Warner, and justified
the characterization of blacks and whites as castes rather than
races on the ground that they were socially, and not biologically,
defined categories. Monographic studies were accompanied by
discussions in general and comparative terms. Kingsley Davis
(1941) published a paper in which he contrasted the 'primarily
physiognomic, usually chromatic' basis of the caste system of
the United States with the 'purely socio-economic' basis of the
16 Society and Politics in India
caste system prevalent in India. None of these formulations
was wholly satisfactory, although several of them illuminated
interesting features of both _systems. The point I wish to stress
is that in all these writings 'caste' was used not merely as a
metaphor but as a concept, and attempts were made, though
never very successfully, to formulate the concept precisely.
Students of caste in India have drawn on insights from the
study of race in two quite different ways. There were the earlier
anthropologists, of whom Risley is perhaps the most notable
example, who constructed elaborate arguments to prove that
the caste system originated from the encounter of races (Risley,
1908; also Ghurye, 1969). I shall have nothing to say about the
part played by racial difference in the origin of the Indian caste
system. My concern is with the approach in which insights from
the study ofrace in the United States and from the study ofcaste
in India are used to illuminate each other. A good example of
what I have in mind is the work of G.D. Berreman (1966; 1967;
1968). The approach adopted there showed promise when it
first appeared, but it went into a decline after the 1960s1 and
has never really recovered its voice.
Berreman's essays and other studies which sought to present
caste as a form of stratification were dismissed as examples
of 'butterfly collection'2 in which superficial similarities were
allowed to conceal profound differences. It must at once be
pointed out that those who introduced the concept of caste into
the study of racial stratification in America were acutely aware
of the differences between India and the United States, which
some of them stressed to a degree that may not have been fully
justified. Warner (1936) pointed out that caste in America dif-
fered from its Indian prototype because the former, presumably
unlike the latter, existed not by itself but in conjunction with a
system of classes. Myrdal (1944), in his tum, pointed out that,
unlike in India, caste in the United States existed in a moral
environment governed by the principle of equality.
Berreman (1960) brought his experience of life in Mont-
gomery, Alabama in 1953-5 to the study of a village in Debra
Dun district in which he lived in 1957-8, and found that the
first experience greatly illuminated the second. He noted in
particular the deep resentment of the underprivileged groups
 Race, caste and gender 17
in both cases even where they appeared to acquiesce in their
social subordination. He went on to construct a formal typology
of kin groups, local groups, castes and classes, summarizing their
similarities and differences in a somewhat mechanical manner
(Berreman, 1967; 1968). To make matters worse, he appeared
to be arguing that the real objective of the comparative method
was to reveal similarities between systems.
A change of attitude towards such studies came about in the
mid- l 960s reflecting to some extent a change of outlook and
orientation among anthropologists in general. Behaviourism and
empiricism came under attack, while a case was being made at
the same time for redefining the whole field of sociology as
the sociology of ideas. In Indian studies this meant a slow,
often unperceived and generally unacknowledged shift from
the 'field-view' to the 'book-view' of society, culminating in
the assignment of a privileged position to traditional 'structure'
over contemporary 'reality'.
What is of specific interest to the present argument is the
redefinition of the aim of comparison, viewed now as being
directed .to 'typification' rather than 'classification' (Dumont,
1967). I would say that a fundamental shift of orientation came
about in anthropology with the dominance ofan intellectual style
in which 'difference' became the primary object of attention.
The major figure in this shift was, of course, Levi-Strauss,
and the same shift made its impact on studies of caste through
the writings of Louis Dumont. Levi-Strauss made his point
about 'difference' most sharply while contrasting the aims of
anthropology and history: 'It is true that a discipline whose main,
if not sole, aim is to analyse and interpret differences evades all
problems when it takes into account only similarities' (1963, p.
14). This seems a very arbitrary requirement that a discipline
should either only interpret differences or only take similarities
into account.

Like race in the United States, caste in India is perceived by

millions of people today as a particularly rigid and oppressive
form of inequality. Many practices, described in earlier text-
books as integral to the normal functioning of caste, would now
be considered invidious and discriminatory, and might invite
18 Society and Politics in India

legal and political sanctions. Fifty years ago it might have made
sense to say that discrimination based on race was pathological
while discrimination based on caste was normal. To insist on the
same contrast would be misleading today.
When we consider caste and race together, we are struck at
once by the remarkable similarity in the contrasting attitudes
towards women oflower and higher ranks characteristic of men
in privileged positions in both systems. My argument is that
inequalities of caste are illuminated in the same way as those of
race by a consideration of gender. There are two aspects of the
problem. There is, firstly, the sexual use and abuse of women,
which is an aspect of the inequality of power, seen in its most
extreme form in the treatment of women of the lowest rank by
men of the highest; this is the _aspect of the problem that has
received most attention. There is, in addition, the unremitting
concern with the purity of women at the top, associated with
ideas regarding bodily substance that have been discussed
separately in studies of American kinship (Schneider, 1968),
and of caste and kinship in India (lnden and Nicholas, 1977;
Marriott and Inden, 1980); we can deepen our understanding of
both caste and race by exploring these ideas more systematically
and in comparative terms.
If we believe that the position assigned in thought and life
to women is of crucial significance to the understanding of
both caste and race, we are much better placed today than
anthropologists were a generation ago to pursue the comparison
between the two in greater depth. The position of women
in society, particularly in modem or contemporary society,
received very little scholarly attention from sociologists and
social anthropologists in the decades when comparisons of
race and caste were most extensively made. It is true that
Dollard (1957) wrote about the 'sexual gain of caste' in the
U S South and Berreman (1960) later wrote about the sexual
exploitation of both black and untouchable women. But these
observations were either lost or ignored in the absence of an
adequate conceptual framework for the comparative study of
gender.
It may well be the case that such a framework does not
exist in a fully developed form even now. But there is no
 Race, caste and gender 19
doubt that the climate has altered vastly so that the plea
for a serious consideration of these issues can no longer be
as easily ignored as in the past. The advances achieved in
women's studies in the last two decades have implications not
only for a fuller understanding of the relations between the
sexes, but also for a deeper insight into the general problem
of inequality, of which caste and race are two particular forms.
I am talking now not only about new facts but also about
new ways of looking at facts that have long been taken for
granted.
The sexual use of women of inferior rank by men of
superior rank would not acquire its characteristic forms in
societies divided by caste or race if the ordinary relations
between men and women were not marked by asymmetry.
The asymmetry characteristic of such relations in general is
merely reinforced when the man belongs to a superior race
(or caste) and the woman to an inferior one. The normal
requirement of asymmetry would be seriously upset if the
woman belonged to a superior and the man to an inferior
rank. The stricter the demand for asymmetry in the ordinary
relations between men and women, the more severe will be
the sanctions against the reversal of roles. I would surmise
that the distances required to be maintained between castes
or between races are likely to vary directly with the dispar-
ities established between men and women in the society as
a whole.
We have to be careful, however, to distinguish between
relatively stable societies and those undergoing rapid change
as a result of changes in the legal and political systems and
in the general climate of opinion. Such changes have been
marked in the last four or five decades not only in the
United States, but also in India. In these changing conditions,
small and gradual reductions in disparities are periodically
met with sudden and violent reprisals which bring established
patterns into sharp relief. It is difficult, when this is happen-
ing, to demonstrate or even to discern any clear direction of
change.
The asymmetry inherent in the link between race and gender
is nicely brought out in Dollard's study of Southerntown.
20 Society and Politics in India

In simplest terms, we mean by a 'sexual gain' the fact that white

men, by virtue of their caste position, have access to two classes of
women, those of the white and Negro castes. The same condition
is somewhat true of the Negro women, except that they are rather
the objects of the gain than the choosers, though it is a fact that
they have some degree of access to white men as well as to men
of their own caste (Dollard, 1957 (1937), p. 135).
This asymmetry sustains and is sustained by contrasting images
of the sexuality of black and white women of which exact
parallels may be found in the contrasting images of lower and
upper caste women in India.
Leaving aside the facts of interaction for the moment, we may
turn very briefly to the logic of the asymmetry indicated above.
That logic is articulated very well in the Hindu Dharmashastras.
The traditional Hindu theory of marriage clearly reveals the
dual subordination of inferior to superior varnas and of women
to men in the distinction it maintains between anuloma and
pratiloma unions. An anuloma union is one between a man
of a superior and a woman of an inferior varna, and, subject
to certain conditions, it is accepted. The rule in its broadest
interpretation allows a Brahman man to take, in addition to
a Brahman wife, a Kshatriya, a Vaishya and a Shudra wife; a
Kshatriya man is allowed to take, over and above his Kshatriya
wife, a Vaishya and a Shudra wife; a Vaishya man may take, in
addition to a wife from his own varna, one also from the Shudra
varna; a Shudra man has to be content with only a Shudra wife
(Manu, 1964, p. 77). Pratiloma, on the other hand, is the union
of a woman of a superior varna with a man of an inferior one,
and it is condemned in the severest possible terms. The lowest
of human beings, akin to beasts, are the Chandalas who are
described as the offspring of pratiloma unions between Brahman
women and Shudra men (Manu, 1964, p. 405).
It must be pointed out that scriptural authorities are by and
large uneasy about anuloma even though they acknowledge its
consequences. We may say that there is a norm of anuloma only
in the sense that its consequences are acknowledged but not in
the sense that the act itself is commended. Or, we may say that
the act itself is viewed very differently from pratiloma which is
 Race, caste and gender 21
clearly condemned. The contrasts here are strikingly similar to
the contrasts encountered in the conventions governing unions
between whites and blacks.
By its acceptance of polygyny, Hinduism gave itself room to
construct an elaborate formal structure for defining the relations
between men and women belonging to superior and inferior
varnas. Protestantism, with its strict code of monogamy, left
itself little room for elaborating a theory of hypergamy, but it
gave a kind of piquancy to sexual relations between the races
by making them in varying degrees unsanctioned. It must be
remembered that in the US South all sexual unions between
whites and blacks were extra-legal; but the extra-legal domain
itself was not homogeneous, being differentiated according to
recognized, if not well-defined, principles.
We must not make the mistake of believing that Indian
practice adhered strictly to Hindu theory, and that all inter-
caste unions were according to the recommendations of the
Dharmashastras. We have seen that the Dharmashastras them-
selves were uneasy about unions between varnas. P.V. Kane
(1974, pp. 449-52), our leading authority on the subject,
suggests that anuloma unions came to be viewed with increasing
disfavour by authors of legal digests and commentaries from
around AD 900, although we know that such unions in various
forms were legally recognized as marriages until our own time.
What is germane to the issue is that, with or without anuloma,
a large number of extra-legal unions took place between men
and women of different castes everywhere and at all times,
and that these unions were governed by the same unwritten
rules which, according to Dollard and many others, governed
extra-legal unions between the races.
A great deal has changed, in law as well as politics, in the last
forty years, not only in the United States but also in India. The
Hindu Marriage Act of 1955 has set itself against the theory
of anuloma by allowing inter-caste marriage and disallowing
plural marriage. But the asymmetry of which anuloma was the
theoretical expression is still very much in evidence in social
practice. Inter-caste marriages are infrequent if not rare, and
it is difficult to make any categorical statement on the basis of
the limited and rather uneven information available.
22 Society and Politics in India

A great many sexual unions take place outside marriage,

including some between persons of different castes. These
range from permanent companionship at one end, through
semi-permanent and casual liaison, to seduction and rape at
the other. In the absence of detailed information collected
systematically, one can go only by general impressions. Such
impressions clearly indicate that there is a wide measure of
tolerance of extra-marital relations between men of superior and
women ofinferior rank, particularly between men oflandowning
castes and women of landless, including untouchable, castes
(Freeman, 1979), whereas the reverse relationship generally,
though not invariably, meets with reprisal.
I would like in passing to point to an important change in
the attitude of the courts in these matters as indicated in the
judgement of the Calcutta High Court in Mongal Chandra
versus Dhirendra Nath (AIR, 1976, p. 129). Mongal Chandra,
the illegitimate son of a Shudra named Bhadreshwar by his
Brahman concubine, Urmila Bala, claimed succession as a
dasiputra (son ofa female servant or slave) to a part ofhis father's
estate. The arguments against him were twofold. Firstly, it was
pointed out that the dasiputra had a recognized claim only among
Shudras, i.e. it would hold only if he had been the illegitimate
son. of a Shudra by his Shudra concubine. Secondly, since the
union of which he was the offspring was a pratiloma union, no
claim could possibly arise. The High Court rejected both the
arguments and upheld Mongal Chandra's claim to a share in
his father's estate equal to half the share due to his legitimate
half-brother.
Although it is difficult to be categorical, it would appear
that upper caste men have less easy access to untouchable
and tribal women .than they did in the past. From this I
am inclined to infer that material sanctions are more decisive
than ritual ones in restricting such access. When the balance
of political power made the risk of material sanctions relatively
small, ritual sanctions were not very effective in preventing the
sexual use of untouchable·or tribal women by upper caste men.
The balance of power has now changed, though perhaps not
very radically, and this has altered not so much the attitudes
of upper caste men as their horizon of possibilities. We have
 Race,casreandgender 23
accounts of similar changes taking place in the US South in
the 1930s.
Relations between castes are changing rapidly and these
changes are accompanied by reports of caste violence, including
atrocities against untouchables and tribals in many parts of the
country. The new legal and political systems have not eliminated
the traditional hierarchical order, but they challenge it at many
points. Disputes lead to clashes between members of different
castes. It is difficult to assess the extent ofchange, because caste
clashes are now reported much more extensively than before
although, clearly, not all such clashes are reported even now.
The disputes that lead to atrocities against untouchables and
tribals arise from many causes. Some of them clearly are
engineered by interested political parties. There are many others
that arise from the conditions of land tenure and of agricultural
work. But there can be no doubt that there has been an increase
in the clashes that arise out of attempts to control and use the
sexuality of lower caste women. It is a sign of the changing
times that annual statistics of atrocities against untouchables
and tribals, including atrocities against their women, are now
officially published in India. 3 These statistics are defective on
many points, but they are illuminated to some extent by reports
of increasing violence against women in general. 4

It is clear that there is some pattern in the use and abuse of

the sexuality of lower caste women, even though the pattern
is changing. This has to be seen in conjunction with the
jealous attitude towards and strict control over the sexual and
reproductive capacities of upper caste women. The jealous
concern of white men for the purity of their own women has
been noted by most students of race and stratification in the
United States (and also South Africa). The purity ofwomen has
of course been long recognized as the cornerstone of the Hindu
theory ofcaste and kinship. We are now in a position to compare
the two systems at a deeper level as a result of advances recently
made by cultural anthropologists, mainly American, in the study
of ideas regarding bodily substance in the United States as well
as in India.
When we compare caste and race at a deeper level, we find
24 Society and Politics in India
in both systems a prevalence of values and symbols relating to
blood and natural substance, and beliefs regarding the strong
constraints imposed by them on human character and conduct.
These beliefs, values and symbols are deeper in the sense that
they remain relatively unaltered even when the asymmetries
described earlier change due to changes in law and politics.
Hindus regard differences of caste as being in some sense
differences of substance, and believe that these latter impel
members of different castes to act differently. There are, as
I shall show, parallel beliefs that differences of race express
differences of natural substance which constrain character as
well as conduct. One might still contrast caste and race by
arguing that ideas about natural substance and the constraint
imposed by it on social conduct are central to Hindu culture
and peripheral to American culture, but I doubt that such an
argument can be easily sustained.
I would like to enter here into a brief discussion ofSchneider's
account ofAmerican kinship (Schneider, 1968). It has stimulated
a body ofwork on caste and kinship in India, and the authors ofan
important essay on caste systems have acknowledged its seminal
influence on their work (Marriott and Inden, 1980).
Schneider describes American kinship as a part of American
culture, which for him is a system of symbols and their
corresponding meanings. There are two symbolic features,
described for short as substance and code, which singly or in
combination define the domain of kinship in American culture.
Americans think of kinship in terms of shared biogenetic
substance, typically blood; they think of it also in terms of a
distinctive moral code, expressive of diffuse enduring solidarity,
or love. Father, mother, brother, sister, son, daughter, etc. are
relatives in the full sense of the term because they share the
same blood and also because they love each other or ought
to do so. In-laws, step-children and foster-parents are also
relatives but not in the fullest sense because, although there is
love between them, they do not have the same blood; husband
and wife constitute a special case because although, like in-laws,
they are brought together by marriage rather than blood, there is
nevertheless a transmission of substance between them. Natural
relatives (e.g. genitor and illegitimate offspring) are the obverse
 Race, caste and gender 25
of in-laws because between them there is no recognized code, or
so Schneider would have it, although there is shared substance.
According to Schneider, Americans believe that 'relationship
as substance' belongs to the natural order whereas 'relationship
as code' belongs to the social order. The natural order has its
own compulsions as does the social order, although the two
compulsions are not of the same kind. The interpenetration
of the 'natural' and the 'social' orders within the domain of
American kinship is a subject of crucial importance on which,
unfortunately, Schneider's account does not throw much light.
The significance of Schneider's work is that it has drawn
attention to a fundamental feature ofAmerican, indeed Western,
culture as a whole. It is true that in one sense kinship is to a large
extent segregated from other aspects of American culture, but
in another sense the dichotomy between substance and code is
of general significance. Clearly, the pivot on which the relation
between race and stratification turns is the question of rightful
kinship, i.e. with whom one may rightfully have kinship, with
whom one may not, and for what reasons. It goes without saying
that 'rightful' is not the same thing as 'legal' or 'by law' or even
'legitimate'; unfortunately, these distinctions are obscured rather
than clarified by Schneider's manner of exposition.
The intimate, though negative, relationship between race and
kinship is nicely brought out by Everett Hughes. After drawing
attention to the highly flexible nature of American marriage and
the kin ties arising from it, he goes on to say:
'But on one point of difference the grandly flexible system is
hard and unyielding. The essence of the race line in North
America is that no person identified as Negro will be admitted
as effectively social kin of any person classified as white' (Hughes,
1965, p. 1136).
I must, however, point to the asymmetry which is not brought out
with sufficient clarity by Hughes. The risk ofbeing kinless does
not weigh equally with all children of racially-mixed unions; it
is likely to weigh much more where the genitor is a black than
where he is a white.
The ethnography of the US South to which I have already
alluded provides fairly detailed information on extra-marital
26 Society and Politics in India
sex, concubinage, and illegitimate offspring which may be
used for re-examining the place of substance and code in
American culture. The one point that is stressed above all
others is the strict governance of the relations between whites
and blacks by the rule of endogamy. This is the reason given
most frequently for choosing the term 'caste' for the system.
The same literature also points out with unfailing regularity that, -
although marriage was by definition confined within the caste,
sexual unions commonly took place across it. This at once raises
the question of the social position of the concubine or mistress
and of the natural children borne by her.
There is ample evidence of the presence in many cases,
though by no means in all, of bonds of affection - perhaps
even of love - between a man and his mistress, and between
him and his natural children. Davis, Gardner and Gardner
recount the story of a white man who stood by as the house
of his black mistress was burning down. Unable to bear the
sight any longer, he rushed into the house, calling out, 'Let me
in to save my children', and earning permanent ostracism from
his own community (Davis, Gardner and Gardner, 1941, p. 31).
Summing up their observations, our authors state:
Furthermore, the white man accepts the children as part of the
relationship; he cares for them and exhibits much the same
affection as if they were legitimate. Thus there is formed a family
group which, at least within the home, ignores caste restrictions
(Davis, Gardner and Gardner, 1941, p. 38).
Here we see the great significance of the distinction between
the 'politico-jural' and the 'domestic' domains; what has to
be denied in the former may nevertheless be acknowledged
in the latter.
Settlements were made of house and other property, and
sometimes even of land, for the maintenance of the concubine
and, less frequently, for the upkeep -and future well-being of
the natural child (Davis, Gardner and Gardner, 1941; Dollard,
1957). The ownership of a black plantation could on occasion
be traced to a gift from a white landowner who had fathered
a coloured child. 5 One can see that a black mistress and her
children might claim an obligation on the part of the white
 Race, caste and gender 27

husband-father to give them protection and patronage. But why

should the white man acknowledge a claim that had no basis
whatever in the law? The answer seems to me to be obvious. The
very fact that American culture places a high value on 'bioge,netic
substance' means that there is some obligation towards one's
own substance, even when that substance is generated clearly
outside the law. One is compelled by American culture to
acknowledge a part of oneself in one's natural child.
The flaw in Schneider's argument, it seems to me, lies in his
belief that code can be completely separated from substance
within the framework of American culture. The two may
indeed be considered separately for many purposes and in
many contexts, but only up to a point and within certain limits.
A father cannot disown his son - or a son his father - however
much he may be socially embarrassed by him, precisely because
son and father are of the same substance. 'Owning' here means
owning an obligation which can only be expressed in social terms
and which does not cease to be social simply by being outside the
law. Other codes may be violated or disowned; but in American
culture, and I suspect in Indo-European culture generally, it is
impossible to disown completely the code that is inherent in an
immediate relationship by blood.
I have argued that the distinction between 'legal' and 'extra-
legal' is by no means simple, at least so far as kinship by
blood is concerned. The extra-legal not only has its own
code, but is itself internally differentiated. I may illustrate
the point by adapting the distinction, formulated by Fortes,
between 'illegitimate' and 'illicit' (Fortes, 1969, p. 252). Only
the children of legally-wedded spouses ofthe same colour-caste
are 'legitimate' in the restricted sense of having full legal title.
The child of a white man and his black commonlaw wife is
illegitimate; but neither the union nor its fruit is illicit; the
mother can transmit status to the child. A sexual union between
a black man and a white woman would be in a wholly different
category; like incest, itwould be illicit, and neither parent could
transmit status to the offspring.
Schneider's mistake has been magnified by some ofthose who
have carried his conceptual scheme into the study of Hindu caste
and kinship. I will take as an example the account of kinship
28 Society and Politics in India

in Bengali culture by Inden and Nicholas which begins with

a handsome acknowledgement of indebtedness to Schneider.
Substance and code, according to the authors, are fundamental
features of Bengali kinship, but their mutual relationship is quite
·different in Bengali culture from what it is in American culture.
This is so because the premise of Bengali culture is altogether
different from that of the American:
As a consequence of this cultural premise, no distinction is made,
as in American culture, between an order of "nature", defined by
shared biogenetic substance, and an order of "law", defined by code
for conduct (lnden and Nicholas, 1977, p. xiv).
The authors proceed to underline 'the inseparable relationship
of code and substance in Bengali culture' (lnden and Nicholas,
1977, p. xiv), suggesting clearly their separability in American
culture.
I have already indicated, and will try to show by further
illustration, that the assumption of the radical separability of
substance and code in American culture is open to question. We
have seen that in American culture some social obligations are
entailed in the natural kinship between father and illegitimate
child. We must now ask whether Bengalis - or Hindus in
general - are able to distinguish between 'artificial' and 'real'
kinship, and the answer to that question will show that there
are circumstances under which they are able and willing to treat
substance and code as separate.
I shall avoid the obvious trap of adoption, because in Hindu
law adoption was governed traditionally by strict conditions,
including the condition that adoptor and adoptee be ofthe same
caste - a point in favour ofthe argument by Inden and Nicholas. 6
The ties of kinship may, however, be extended artificially in
other ways than by adoption. There is, firstly, what is broadly
described as ritual kinship. Adrian Mayer tells us: 'There are
several ways in which people of different castes can be linked
as kin through ritual acts' (1960, p. 139); and, further, 'there is
no great feeling that ties should be made inside or outside the
caste' (1960, p. 142). Then there is 'village kinship', through
which terms and some forms of courtesy are extended to
co-villagers. Obviously, the strength of these ties varies greatly.
 Race,casreandgender 29
Mayer himself classifies kin ties· into three kinds according to
their strength. The strongest are real ties and certain kinds of
ritual ties where 'there are definite obligations with a minimal
amount to be fulfilled on pain of general public disapproval'
(Mayer, 1960, p. 146). Then there are 'minor ritual kin ties
(rakhi) and strong friendships which have become expressed in
a kinship idiom' (Mayer, 1960, p. 146); these too entail definite
obligations, though not ofthe same kind or the same strength as
in the first case. Finally, there is 'village kinship' where the idiom
of kinship is extended mainly as a form of courtesy.
I would like to stress the point, to which Mayer has also
alluded, that kin terms and corresponding modes of behaviour
are commonly extended, sometimes across caste, in a highly
differentiated manner. This means that a certain person may be
treated as mother's brother and another person as father's sister
even though they both belong to castes other than one's own. It
is difficult to see how this could happen if code and substance
stood in an 'inseparable relationship' in Hindu culture.
Marriott has used Schneider's unit ideas of substance and
code to formulate an elaborate and complex argument about the
'transactional and transformational culture of India' (Marriott,
1976, p. 111). This argument may be viewed as a first step
in an ethnosociology of Indian culture which will lead to the
construction of a more informed general sociology free from
the distortions inherent in the use of categories derived from
one civilization for the study of all civilizations. This first step,
however, entails an accentuation of the contrast between Indian
and Western thought and culture. It is with this accentuation
of the contrast, rather than with other aspects of Marriott's
important essay, that I am concerned, since it impinges directly
on the comparative study of caste and race.
Turning back for a moment to the example of 'artificial'
kinship, it can certainly be argued that a man may well treat
a person as mother's brother or father's sister, but he will
surely not eat food cooked by either if these 'artificial' kin
both belong to an inferior caste. Thus, the code of kinship
may be. extended artificially up to a point, but not beyond that
point, for there is also a code of food transactions with which
it has to be congruent. Extending the sentiment of kinship may
30 Society and Politics in India
not go very far, it may be argued, if it runs counter to the code
of food transactions.
We have to be careful in dealing with such an argument for
it does indeed point to a very important part of Hindu culture.
There is no doubt about the general importance of food trans-
actions in traditional India and about their specific importance
in the operation of kinship and caste. But a number of further
points need to be made. The code offood transactions was never
observed with the same strictness in all parts of India, and it is
now undergoing change to such an extent that ethnographic data
become rapidly out of date. Undoubtedly, the traditional code of
food transactions was unusually elaborate, but the elaborateness
of a code is not the same thing as its social significance. It is
not at all clear how far the structure of caste (or of kinship)
is dependent for its continued existence on the survival of the
traditional code of food transactions. There are now thousands,
if riot millions, of Indians who ignore or repudiate the traditional
code in both principle and practice, but that certainly does not
mean that they have given up caste.
Marriott is surely right in asserting that Hindu thinking denies
the 'easy separability' of substance and code, and of actor and
action. But I am not sure that Hindus are quite unique in that.
It is true that the separability of actor and action is much
more in tune with modem capitalist that with traditional Hindu
culture. But there is at least one significant area of American
life, concerned with race, where it is precisely this separability
that is widely denied, implicitly if not explicitly.
The doctrine of the separability of substance and code and,
more generally, of actor and action is a liberal doctrine whose
importance in modem Western culture cannot be denied. But
this does not mean that the doctrine is never disregarded in
either theory or practice. Nor is it the case that the inseparability
of substance and code, of actor and action, is affirmed only in
the context of race. The attitude towards the destitute in early
nineteenth-century England was not wholly dissimilar to the
attitude towards the blacks in early twentieth-century America.
It was a common argument, familiar to every reader of Dickens,
that the destitute were unthrifty and improvident by nature and
not due to circumstance, and that charity, whether private or
 Race, caste and gender 31
public, would only harden their nature and not alter their
conduct. Echoes of the same kind of argument are heard today
in the debate about gender; but that is too large a subject for me
to enter into here.
Although attitudes to race in the United States vary greatly
among both whites and blacks, the ethnographic literature on
the US South reveals the persistent belief that whites and blacks
are different by nature:~there are beliefs of inherent difference in
regard to every conceivable attribute, from size of genitals to
aptitude for music. Moreover, whatever white men may believe
about the separability of substance and code, they do not apply
that belief uniformly to themselves and to others.
The explanation of 'Negro conduct' in terms of an unvarying,
indeed unalterable, 'Negro nature' is commonly reported in
the ethnography of the 1930s and 1940s. It is true that racial
stereotypes are now less commonly and less crudely expressed,
at least in public, and perhaps also less widely held. To some
extent this is paralleled in India by the fact that caste stereotypes
are out, at least on the public platform, although they are
widely held and frequently expressed in private. There can be
little doubt that the upper caste Hindu typically believes that
untouchables perform poorly at school and at work - which in
fact they do - because they are made of an inferior substance.
This is surely paralleled by the American belief that the poor
scholastic achievement of the black is due to his inherently
inferior intelligence. 7
As pervasive as, and perhaps deeper than, ideas about black
intelligence are white ideas - and fears - about black sexual-
ity. The idea that black men are governed by ufitamed and
untamable natural sexual urges and that black women are sex-
ually 'hot' and white women 'cold' is a commonplace of southern
literature (Dollard, 1957). These ideas, being consistent with the
asymmetry of power between the races and between the sexes,
no doubt served to maintain that asymmetry. But it would be
an oversimplification to treat them merely as rationalizations
designed to justify and maintain an unequal structure of power.
The idea of substance manifests itself most insistently in the
context of miscegenation. It is true that miscegenation has taken
place extensively in the United States, as also in India, but in
32 Society and Politics in India

both cases it has occurred largely outside the law. In each

case the fact of miscegenation brings out deep-rooted fears
about its effect on the purity of race or caste. Myrdal has
discussed in detail the fear of miscegenation or amalgamation,
and the arguments against it. 'The basic role of the fear of
amalgamation in white attitudes to the race problem is indicated
by the popular magical concept of "blood"' (Myrdal, 1944, p.
587). He also tells us that the standard response of the man
on the street to the plea for racial equality was the presumably
unanswerable question: 'Would you like to have your daughter
marry a Negro?' Myrdal is quick to show us where his own
sympathies lie. But in the light of the discussion now available
on American kinship, we cannot as easily dismiss the popular
American concept of blood as 'magical'. That concept is of
fundamental importance for understanding not only kinship but
also race and stratification in the United States.
Myrdal's own liberal presuppositions prevent him from seeing
in full the real contradiction between the American Creed and
the American attitude to 'blood'. For him, there is a liberal view
of race and a conservative, or even a reactionary, view of it. The
liberal view, which is also his own view, is the rational one;
it has gained ground steadily and is bound to prevail in the
end. In the meantime, the conservative view, arising out of the
'popular magical concept of "blood"', is still entrenched in the
South which, in any case, is known to be backward, although it is
showing definite signs of progress. This view may not be wholly
wrong, but it is superficial and can be misleading.
My reading of Schneider, which differs somewhat from that
of Parsons (1975), tells nie that American attitudes to race are
pervasive and enduring because they are tied up with American
ideas about blood which are deep and fundamental. This does
not, of course, mean that these ideas are unalterable, but only
that their rhythms of change are not the same as the rhythms
of change in what is popularly described as political ideology.
All aspects of a society do not change at the same rate or
even in the same direction. There is abundant evidence of a
change in the relationship between race and the occupational
structure which is a central part of the American system
of stratification (Freeman, 1976; also Pinkney, 1985). But it
 Race, caste and gender 33

would be a mistake to read that evidence to mean that there

has been a corresponding change in the American attitude to
miscegenation, which belongs to a different domain of culture.
All the evidence suggests that, by and large, Americans
continue to adhere to the belief that race is a biological fact.
Why should this be so when every undergraduate student of
anthropology knows, or ought to know, that race is a cultural
and not a biological fact? (Montagu, 1974). It is impossible, in
the face of this evidence, to agree with Marriott (1976) that
'biological substantialism' is a peculiarity only of the Hindus.

I am now in position to return to the original objective of this

paper. That objective was not to reach any definite conclusion
about caste or about race, or about the similarities and differ-
ences between them. My purpose was to raise certain questions
about attitudes towards the comparative method held by influ-
ential students of Indian society and culture; and, at the same
time, to enter a claim for the validity of limited comparisons,
w-hen made systematically and with an open mind.
Comparisons between caste and race have been all but ban-
ished from the field of Indian studies for the last twenty-five
years on the ground that, since Indian and Western civilizations
are so radically different, such comparisons cannot be fruitful
and must be either superficial or misleading (Dumont, 1964;
1966). It is a part of this argument that caste is 'normal' in India
whereas race is 'pathological' in America (Dumont, 1961). Such
an argument is itself misleading because in a rapidly changing
world it is difficult, if not impossible, to determine what is
normal and what is pathological; and it introduces evaluations
that tannot be defended and are not really necessary.
The marked stress on 'difference' and 'contrast' prominent
in anthropological writings on Indian society and culture in the
last two decades is associated with a return to the Indological
approach, or, as I indicated at the beginning of this essay, to the
'book-view' as against the 'field-view' oflndian society. I do not
mean by this that the anthropologists who have contributed most
to this return - whether Dumont or Marriott - have ever denied
the importance of fieldwork; indeed the fieldwork done in the
1950s by each of these anthropologists was outstanding, if not
34 Society and Politics in India

exemplary. 8 It is none the less true that they have increasingly

taken their orientation from Hindu thought rather than Indian
life, however crude that distinction might sound. Moreover, the
growing preoccupation with thought rather than action has led
them, and their followers, to go back to the past and to locate
its basic elements in classical Hinduism and its religious and
philosophical literature.
The book-view or Indological approach assigns a privileged
position to the past as compared to the present. Obviously, a
great deal offieldwork has gone into the anthropological writings
to which I have been referring. Butwe have to consider not merely
the quantum of fieldwork but also its orientation; not just how
much fieldwork one does but also where one sets one's sights.
It hardly needs to be argued that in anthropological fieldwork
what one observes and what one sets out to observe are never
wholly unrelated. In particular, there is a marked tendency in the
anthropological writing which has emerged from the fieldwork
to which I refer to push to the margins whatever is distinctive of
modem or contemporary Indian life.
All this has meant that the predominant anthropological
representation of Indian society and culture over the last twenty
years or so has had a certain timeless character. Attention
has shifted away from technology, politics and law and has
been focused on ritual, ceremony and religious thought. It is
maintained, directly or indirectly, explicitly or implicitly, that the
changes taking place in India now are confused and confusing,
that they affect the surface of Indian life without touching its
core. If in my discussion ofrace in the United States I have used
mainly the ethnography of the 1930s and 1940s this is in part
deliberate, since we have to believe that some things in America
remain relatively fixed even though many things change, just as
some things in India change even though many things remain
unchanged. ·
What emerges from the literature that I have criticized is a
'structural' view of Hindu culture against the backdrop of a
'historical' view of Western civilization. This fits in very well
with the emphasis on 'difference' and 'contrast' to which I have
drawn attention. My view is that if we are to develop the study of
Indian society and culture within the framework of comparative
 Race, caste and gender 35
sociology, we must put back the Indological approach where it
properly belongs. I mean by this not that we should ignore the
past or treat it as unimportant, but simply that the present and
not the past should be the point of departure in the sociology
of India as it is, or ought to be, in the sociology of any society.
A sociology of India that has its orientation to the past and
disregards or devalues the present is bound to be unfruitful and
in the end self-defeating.
It is, of course, the most difficult thing, in applying the
comparative method, to maintain a proper balance between
comparison and contrast. One must try nevertheless to be
faithful to the facts and fair to scholars with a different
intellectual orientation from one's own. Perhaps there may be
genuine differences of orientation between sociologists devoted
to the study of their own society and those devoted to the study
of other cultures. I would not, however, push that point too
far, because implicit in it is the presumption that societies
or civilizations (India and the West in the present case) are
somehow like substances and, as such, mutually impenetrable;
such a presumption itself becomes an obstacle to comparison.
Those who adopt the 'typifying' or the 'distinctive features'
approach do not renounce comparison in either principle or
practice. It is true all the saine that their approach leads to a
sharpening of contrasts in the short run if only because they
hope thereby to make their comparisons more fruitful in the
long run. For what does it mean to typify if not to engage
in 'one-sided accentuation' (Shils and Finch, 1949, p. 90)
for establishing clear contrasts? We know how effectively that
technique was used by Max Weber for constructing ideal types
of great analytical value.
However, it is one thing to engage in 'one-sided accentuation'
for constructing ideal types of, say, economic action or legitimate
authority, and quite another to accentuate in a one-sided way
the peculiarities of a whole nation or a whole civilization. In
the former case it is easy to keep in mind the fact that 'market
rationality' or 'charismatic authority', as the case may be, is
a construction that we have made for a particular analytical
purpose. In the latter, it is easy to lose sight of the distinction
between the construct and the reality, for it is useless to pretend
36 Society and Politics in India

that human beings - including anthropologists and philosophers

- can be persuaded to regard India, France, Europe, America
or the West in the same way as, or with the same detachment
with which, they might regard 'market rationality' or 'charismatic
authority'. It is here that the 'typifying' or the 'distinctive fea-
tures' approach, with its inclination for 'one-sided accentuation',
may become a source not only of intellectual error but also of
political mischief.
 2 Race, caste and ethnic identiry
The attempt to view race and caste within the same framework
of understanding could take us in two different directions. In
the first place, we might consider to what extent systems of
stratification based on caste (as in India) and on colour (as
in the Southern United States) can be regarded as analogous
in structure; this is a problem in comparative sociology. In the
second place, we might ask how far in India caste distinctions
correspond to differences in physical or racial type; this problem
is of more special interest to students of Indian society and
history. 1
When American social anthropologists, mainly under the
influence of Lloyd Warner, began to study the Deep South
of the United States in the 1930s, they found it useful to speak
of a caste system in representing the cleavages between blacks
and whites in rural and urban communities there. 2 Gunnar
Myrdal employed similar terms and categories in his classic
study of the American Negro made at about the same time. 3
The metaphor of caste has since then been widely used in
describing multi-racial societies in other parts of the world,
notably South Africa.4
There are certain obvious parallels between the Indian caste
system and the system of stratification based on colour, whether
in the US South or in South Africa. In studying the US South
both Warner and Myrdal were struck by the rigid distinctions
maintained between blacks and whites which seemed to them to
be in marked contrast with the more flexible pattern of relations
in a class system. Their purpose in labelling as 'caste' the system
of stratification based on colour was not so much to explore its
similarity with the Indian system as to emphasize its difference
from the class system in America and other Western societies.
It might be useful to explore a little further the similarities
between the Indian caste system and what I shall call for short
the colour-caste system. In both systems the component units
38 Society and Politics in India

are differentiated from each other by clearly defined boundaries.

Differences between castes are reinforced by a measure of
homogeneity within the caste.
Caste systems may be described as systems of cumulative
inequality. Advantages of status tend to be combined with
advantages of wealth and power, and those who are socially
underprivileged also tend to be at the bottom of the economic
and political scales. There are many exceptions to this in the
colour-caste system where poor whites co-exist with well-to-do
blacks, 5 but exceptions of the same kind have existed in Indian
society for a long time. 6
In both systems the component units maintain their social
identity through strict rules of endogamy. In a class system
individuals tend to marry within their own class but there are no
prescribed rules which require them to do so. In the US South
marriages between blacks and whites were strictly forbidden and
this is still the case in South Africa. In India the principle of
endogamy was in certain areas mitigated by the practice of
hypergamy (anuloma) by which a man from a higher caste
could under prescribed conditions marry a girl from a lower
caste. It must be emphasized that traditionally the practice of
hypergamy was governed by strict rules which recognized the
distinctions between castes as well as their hierarchical order;
and, as Mrs Karve has pointed out, it 'is found in certain parts
of India among only certain castes and i_s not a general practice in
any region'. 7 Those who define systems of stratification in terms
of the rigidity of marriage rules are bound to be struck by the
similarity between the Indian and the colour-caste systems.
Closely associated with the rules regulating marriage are
certain attitudes towards women characteristic of both types
of society. A very high value is placed on the purity of women
belonging to the upper strata and they are protected from sexual
contamination by men of the lower strata by sanctions of the
most stringent kind. 8 On the other hand, there is a strong
element of 'sexual exploitation' in the relations between men
of the upper strata and women of the lower. Berreman notes
that the 'sexual advantage' enjoyed by high caste men in an
Indian village studied by him are similar even in their details
to those enjoyed by white men in the town studied by Dollard
 Race, caste and ethnic identity 39
in the US South. 9
We might at this stage sum up the characteristics of castes by
saying that they are hierarchically ranked groups or categories
based on hereditary membership which maintain their social
identity by strict rules of endogamy. The fact of hereditary
membership is of great importance. It fixes the social. status
of the individual at birth and prevents his movement from one
group or category to another. In spite of many exceptions, these
factors combine to fit the social divisions in a caste society into
an uncommonly rigid mould.
If I began by considering the similarities between the two
types of social stratification, this was not to imply that I
consider these to be in some sense more fundamental than their
differences. Opinion is sharply divided on the significance to be
attached to these similarities and differences, 10 and scholars like
Dumont11 and Leach 12 would consider it misleading to describe
systems of stratification based on colour as caste. For the:m, the
institution of caste in the true sense of the term is a unique
feature of the pan-Indian civilization.
The differences between the two types ofcaste system - using
the same term for convenience - are obvious enough, but it has
not proved easy to sum them up in a formula. Some would
draw the distinction by saying that one represents a 'cultural
model' and the other a 'biological model' .13 The caste system
in India is certainly a cultural phenomenon, but is it adequate
to represent the colour-caste system in the US South (or in
South Africa) simply in biological terms? Both Wamer 14 and
Myrdal15 had first considered and then rejected the view that
the groups they were studying be described as races. A quick
look at their argument will throw some light on the complex
relations between race, culture and society and help us to probe
a little deeper into the subject of our study.
Warner insists that in the stratification system of the Deep
South the categories black and white are socially and not
biologically defined. Persons who are socially defined as blacks
might be biologically classified as white and people who are
regarded as blacks in one society might in another society be
viewed as whites.1 6 Myrdal's position is similar. He points out,
first, that 'the "Negro race" is defined in America by the white
40 Society and Politics in India

people' and second, that 'this definition of the Negro race in

the United States is at variance with that held in the rest of
the American continent.' 17 What is significant is not merely the
presence of physical distinctions but also the manner in which
they are socially recognized which is essentially conventional.
Neither blacks nor whites in the US South can be regarded as
races in the strictly biological sense of the term.
Kingsley Davis sought to characterize the distinction which
we are considering as being between 'racial' and 'non-racial'
caste systems.
A non-racial caste system, such as the Hindu, is one in which
the criterion of caste status is primarily descent symbolised in
purely socio-economic terms; while a racial system is one in
which the criterion is primarily physiognomic, usually chromatic,
with socio-economic differences implied.IS
We have just seen why it is not wholly satisfactory to describe the
caste system in the United States as racial; and it is not entirely
clear that the chromatic differences there are more fundamental
than the socio-economic ones as Davis would seem to suggest.
Nor is it wholly satisfactory in this context to view 'race' and
'descent' in opposition for in both cases we are concerned with
the cultural definition of biological processes.
It is true, none the less, that visible physical differences are
much more conspicuous in the colour-caste system than in
the Indian. An outsider in the US South will not have much
difficulty in deciding in the majority of cases who belongs to
which caste merely from appearance. In India he will find
it difficult, if not impossible, to do this beyond a certain
point. But this in itself would not establish the absence of ·
more fundamental genetic differences between castes in Indian
society. Indeed, their complete absence would be surprising in
view ofthe fact that members of most castes are believed to have
practised strict endogamy for countless generations.
Those who emphasize the differences between the Indian
and the American systems would base their argument on
the uniqueness of Hindu cultural values. In fact, one might
distinguish between the 'structural' view of caste which draws
attention to broad similarities and the 'cultural' view which
 Race, caste and ethnic identity 41
regards the caste system in India as unique. 19 There is no
doubt that in India caste is embedded in a system of religious
values which has no counterpart either in the US South or in
South Africa.
Western scholars have been struck by the importance of
hierarchy in the Hindu scheme of values. zo Central to this are
the notions of dharma and karma. 21 These are both complex,
philosophical notions and it is difficult to put them in a nutshell.
Very briefly, dharma implies right conduct in accordance with
one's station in life, defined largely by one's caste; karma
explains-and justified-one's birth in a particular station in terms
of one's actions in a previous life. In other words, moral rules
and standards of worth would differ from one caste to another.
Most Western observers have been struck by the iniquity of the
system, but scholars like. Leach would point out that it ensured
a measure of material and psychological security to all sections
of society, particularly to those at the bottom of the hierarchy. 22
In contrast to the values of traditional India, the American
creed has always placed the highest social value on the equality
of men. Thus, the moral environment in which rigid social
distinctions exist in America is quite different from the moral
environment in India. One may say that the American system
is disharmonic; inequalities exist in fact although rejected by
the normative order. The traditional Indian system was, by
contrast, harmonic; rigid social distinctions not only existed
but were generally accepted as legitimate. If this argument is
correct, then the two types of system would show very different
patterns of tension and conflict.
The values of a society are not easy to describe in an objective
way. They are often ambiguous and made up of conflicting
elements. It is difficult to believe that hierarchical values were
accepted in the same way by all strata of Indian society. Most of
what we know about traditional Indian values is based on texts
written by people who belonged to the top of the hierarchical
system. Perhaps we will never know in quite the same detail how
the order of caste was perceived by people at the bottom of the
hierarchy.
·-B_crreman, who, unlike most students· of Indian society, has
studied a village community by living with the lower castes,
42 Sodety and Politics in India
would contend that there are sharp differences of perspective
between. the lower and the upper strata.23 Others also have noted
the presence of tensions and conflicts between castes which
would not be expected if everyone accepted without question
the position assigned to him within the hierarchical order. 24
However, most of these tendencies have been recorded within
the past twenty years and their emergence in contemporary India
would not contradict the assertion that traditionally the Indian
caste system approximated to the harmonic type.
Berreman also rejects the view that the American value system
can be defined unambiguously in terms of its emphasis on
equality. 25 He quotes Spiro's critique of Myrdal to support his
argument:
The assumption of egalitarian culture norms is untenable unless
one adopts an idealist conception of ideal norms which are irrel-
evant to human behaviour and aspirations. Actually discrimination
against the Negro is not in violation of southern ideal norms; it is
in conformity with them. 26
There is also the question of the colour-caste system in South
Africa. Can we say, perhaps, that here we have a normative
order which accepts the existing structure of inequality between
groups as legitimate?
Differences between the colour-caste system and the Indian
system are not confined to the realm of values. There are
important differences in the structure and composition of the
groups which constitute the two types of system. In the US
South there are only two principal castes, blacks and whites;
in South Africa there are four, Africans, whites, coloureds
and Asians. 27 In India the caste system comprises a large
number of groups whose mutual relations are of an extremely
complex nature.
In India it is not at all uncommon for a single village to
have as many as twenty or thirty castes. 28 Each linguistic
region in the country has between 200 and 300 castes. Many
of these are divided into sub-castes which might in tum be
further subdivided. 29 If we leave the village and take a larger
territorial unit, it becomes impossible even to determine the
exact number of castes in it. The distinctions between caste,
 Race, caste and ethnic identity 43
sub-caste and sub-sub-caste become blurred. The same caste
might be called by different names and different castes by the
same name.
There is no single rank order for all the castes and sub-castes
which applies in every region. Perhaps all that can be said·very
firmly for the country as a whole is that Brahmins rank at
the top and Harijans at the bottom. There is a great deal of
ambiguity in the middle region. The different cultivating castes
make competing claims to superior status. The Brahmins (like
the Harijans) are themselves divided into a number of castes
and sub-castes whose mutual ranks are by no means easy to
determine. JO All this is not to deny that a certain measure of
consensus in regard to caste ranking does exist within the local
community.3 1 This consensus was probably stronger in the past
than it is today.
It can be argued that structurally there is a basic difference
between a dichotomous system and a system of gradation in
which there are many terms. Once again, the two types of
system are likely to display very different patterns of social
conflict. Theories of social class and of conflict assign a
crucial significance to the dichotomous division of society.32
Where the contending parties are two in number, the conflict
tends to be intense; where they are many, a shifting pattern
of coalitions reduces the intensity of conflict. The same theory
can be extended to caste. Where the community is divided
into blacks and whites, the conflict is likely to be sharp;
where it is divided into twenty or thirty groups, no particular
conflict is likely to absorb the energies of the community as
a whole.

Next we shall try to see if any relationship can be established

between caste distinctions and physical differences in the Indian
population. It might be said at the outset that if such a
relationship exists it is not likely to be either simple or direct.
Physical differences are not polarized in India but are spread
over a continuum. The population cannot be readily divided
into races or even into clearly recognizable physical types.
The caste system in its tum is a system of great complexity.
It is divided and subdivided into innumerable groups and a
44 Society and Politics in India

consideration of these might provide a convenient point of

departure.
The word 'caste' is used in India to refer to groups and
categories of very different kinds. Two types of distinctions
are particularly ·important. The first is between vama and
jati and the second is between caste and sub-caste. The
difference between vama and jati can be briefly described
as the difference between a model or a conceptual scheme
on the one hand and a set of real social groups or cat-
egories on the other. There are only four vamas which are
arranged in a particular order whereas jatis are many and
their rank order is both more ambiguous and more flex-
ible.33 Jatis should not be viewed as having grown out of
divisions and subdivisions within a set of four original vamas.
Rather, as Mrs Karve has argued, vama and jati have co-
existed as two different but related systems for at least 2,000
years. 34
The distinction between caste and sub-caste is of a dif-
ferent kind. Both are real social divisions, but one is more
inclusive than the other. If we take potters or carpenters as
examples of castes, we will find that in any given region
there are two or three different kinds of potters or of car-
penters, differentiated according to technique or provenance
or sect or some other less tangible factor. These different
divisions we might refer to as sub-castes. They are similar
in structure to the more inclusive groupings and are generally
endogamous. Scholars like Ghurye would maintain that the
different types of potters are sub-castes, being products of
segmentation within the potter caste. 35 Mrs Karve, on the
other hand, has argued that the different types of potters
are often unrelated and that each should be called a caste
and the potters as a whole a 'caste cluster'. 36 Her argument
is important in this context because she has tried to support it
with anthropometric data. 37
Sometimes there are several levels of differentiation and not
just two. Thus, the Tamil Brahmins are of three main kinds:
(1) temple priests; (2) domestic priests for the non-Brahmins;
and (3) scholars and landowners. The last are divided into
Smartha and Shri Vaishnava. Smartha Brahmins, in their
 Race, caste and ethnic identity 45
tum, are further subdivided into Vadama, Brihacharanam,
Astasahashram and Vattima. The Vadama, finally, are divided
into Vadadesha and Chozhadesha Vadama. 38 This kind of
differentiation makes it useful to view caste as a segmentary
or structural system. 39 For even though each segment is endoga-
mous, the social distance between segments is variable. Thus,
the social distance between Vadama and Brihacharanam is
smaller than the distance between Vadama and a Shri Vaishnava
segment which in tum is smaller than that between any Brahmin
segment and any non-Brahmin segment. This way ofviewing the
system leads us to ask if there is any relationship between social
distance and racial distance.
Most anthropologists who have analysed caste from the
biological point of view would concede that some physical
differences do exist between castes. But they are sharply divided
on the significance they attach to these differences. On the
whole, earlier scholars emphasized the differences in physical
type they observed between castes. Contemporary scholars are
more inclined to stress the fact that most castes are more
or less heterogeneous in their physical composition and that
variations within the caste are sometimes greater than variations
between castes.
It is not enough to know that castes differ from each
other in their biological make up. We would like to know
in addition whether the extent to which they differ in this
regard is related to their social distance. Castes which are
socially adjacent might be quite different in their biological
composition while those which are at opposite ends of the
social scale might show very little difference biologically. To
answer this kind of question satisfactorily we will need a
great deal of systematic empirical material. The evidence that
we now have is scanty and does not all point in the same
direction.
The first serious effort to study physical or racial differences
between castes in a systematic way was made towards the
end of the last century by Sir Herbert Risley.40 Risley not
only believed that such differences existed but argued that
they were systematically related to differences of social rank
between castes:
46 Sodety and Politics in India

If we take a series of castes in Bengal, Bihar, the United Provinces

of Agra and Oudh, or Madras, and arrange them in the 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.41
Risley was also struck by the fact that the upper castes were in
general lighter skinned than the lower and drew attention to a
number of local proverbs in which this distinction was given
recognition.
Risley developed an elaborate theory to explain the social
ranking of castes. He argued that the caste system was the
outcome of the encounter between two distinct racial groups,
one representing a light-skinned, narrow-nosed, 'Aryan' type,
and the other, a dark-skinned, broad-nosed, 'non-Aryan' type.
The Aryans, according to the theory, were not only the dominant
group but also adopted the practice of hypergamy. This practice
led to the formation of a series of intermediate groups whose
social rank varied directly with their amount of Aryan blood.
Risley sought to support his arguments with anthropometric
data. His conclusions were challenged by later scholars who
found fault with both his data and his methods. 42
Ghurye criticized Risley's work but did not reject his argu-
ment altogether. He emphasized the importance of regional
variations and noted that a caste which ranked very high in one
area might closely resemble in its physical features a caste which
ranked very low in an adjacent area. He pointed out that in many
parts of the country there was no clear relationship of the kind
which Risley had sought to demonstrate: 'Outside Hindustan in
each of the linguistic areas we find that the physical type of the .
population is mixed, and does not conform in its gradation to the
scale of social precedence of the various castes'.43
But Ghurye agreed that in the Hindi speaking area itself there
was a close correspondence between the 'physical hierarchy' and
the 'social hierarchy'. Here the Brahmins were long-headed and
narrow-nosed, and very low castes like the Chamar and Pasi
were broad-headed and broad-nosed. On the basis of such
evidence, Ghurye was prepared to conclude that here, at
 Race, caste and ethnic identity 47
least, 'Restrictions on marriage of a fundamentally endogamous
nature were thus racial in origin'. 44
The most comprehensive single investigation so far carried
out is the anthropometric study of Bengal made jointly by an
anthropologist, D.N. Majumdar and a statistician, C.R. Rao. 45
The data were collected from a defined cultural region, Bengal,
comprising both West .Bengal and Bangladesh. Sixty-seven
groups were investigated, including Muslims, Christians, a
few tribal groups and a large number of Hindu castes. These
groups were studied with regard to sixteen basic anthropometric
characters and a number of indices derived from them. Some
serological data were collected in addition. The anthropometric
data were analysed by means of rigorous and sophisticated
statistical tests.
In spite of many qualifications, Majumdar concluded that
there was some clustering of groups according to their social
proximity. The tribal and semi-tribal groups tended to be
clustered at one end and at the other end were the higher castes
such as Brahmin, Baidya and Kayastha. 46 Majumdar pointed out
that these data confirmed the observations made by him in two
other areas in India, Gujarat and Uttar Pradesh. 'In all the three
surveys, it has been found that some correlation exists between
the order of social precedence in a state or region, and the
ethnic constellations based on anthropometric data.'47 It must
be emphasized, however, that the relationships which emerge
from the study by Majumdar and Rao are of a far more complex
nature than the one which Risley believed he had established.
Studies made more recently do not all support Majumdar's
conclusions. Karve and Malhotra have published the results of
a detailed comparison between eight Brahmin 'sub-castes' in
Maharashtra, taking anthropometric, somatoscopic and sero-
logical data into account. 48 Their data show the existence of
significant differences among some of the Brahmin 'sub-castes'.
Comparing their findings with those of other scholars, they
conclude that there is no necessary relationship between social
distance and physical distance. Thus there is no justification for
assuming that the distance between the Brahmin 'castes' under
investigation is less than the distance between a Brahmin 'caste'
and a non-Brahmin 'caste', for some Brahmins are closer to
48 Society and Politics in India
members of other 'castes' than to each other. 49 It would appear
that the more closely we look at the system the less firm we can
be about the linkage betwee~ caste and race.
The shift from morphological to genetical indicators would
seem to confirm the view that the linkage between social and
physical distance is tenuous and uncertain. As my last example
I shall take a study by Sanghvi and Khanolkar which examines
the distribution of seven genetical traits among six endogamous
groups in Bombay. 50 Of the six groups, four are Brahmins; one
is a high non-Brahmin caste, Chandraseniya Kayashth Prabhu
(CKP), ranking next only to the Brahmins; and the other is a
Cultivating caste, Maratha (MK), belonging to the middle level
of the hierarchy. As the authors point out, all these groups have
been regarded by earlier anthropologists as being of the same
physical type.
The results of the analysis show a rather complex pattern of
variations. Some of the Brahmin groups are quite close to each
other, and one ofthem is very similar in its genetical composition
to the non-Brahmin Marathas. The Koknasth Brahman (KB)
are, on the other hand, quite distinctive in their genetical com-
position as are also the Chandraseniya Kayasth Prabhu (CKP).
Moreover, these two groups are marlc.edly different from each
other. 'The magnitude of differences between the groups KB
and CKP for each one of the seven genetical characters is more
or less similar to that between American whites and American
Negroes.'5 1 Although the Chandraseniya Kayasth Prabhu are
non-Brahmins, they rank very high and might be regarded as
being socially proximate to the Koknasth Brahman.
This leads us to a consideration of the social significance
of genotypical as opposed to phenotypical differences. Earlier
anthropologists such as Risley sought to establish a relationship
between the social rank of a caste and the physical appearance
of its members. They were encouraged in their pursuit by
beliefs widely held in Indian society about the existence of
such a relationship. 52 Upper castes are universally believed
to be light-skinned and narrow-nosed and lower castes to
be dark-skinned and broad-nosed. It would now appear that
two socially adjacent castes whose members are very similar in
their physical appearance might nevertheless be quite different
 Race, caste and ethnic identity 49

in their genetical composition.

Genetical differences are likely to acquire social significance
only if their existence is widely known or if they are reflected
in clear differences in physical type. As I have indicated, certain
broad differences in appearance exist between castes at opposite
ends of the hierarchy in many parts of the country and equally
significant are the beliefs and stereotypes regarding these dif-
ferences which persist in spite of much evidence to the contrary.
Beliefs which· are technically wrong or inconsistent sometimes
assume crucial significance in social life. As Passin has argued,
The relation of caste to race is not simply a question ofwhether the
groups are in fact racially different, but rather that there seems to
be some disposition to attribute racial difference to even the most
marginal cues in caste and caste-like situations. 53
This is particularly true in the Indian context where in some lan-
guages the same word is used to denote both caste and race. 54
What is important in social life is the sense of solidarity which
people feel when they belong to the same community and the
feeling of distance which separates members who belong to
different·communities. The sense of community is often based
on the feeling that its members have a common origin. This
feeling may be vague or it may be consciously formulated in an
ideology. It may be strengthened if the community is marked out
by distinctive physical features, but this is not a necessary con-
dition for its existence. Sometimes a strong sense of community
can exist even in the absence ofvisible physical indicators. This
leads us to a consideration of ethnic groups and identities.

The systematic use of the concept of ethnicity is of relatively

recent origin in sociology and social anthropology although the
presence of ethnic groups in the United States has been widely
discussed for many years.
An ethnic group is a distinct category of the population in a
larger society whose culture is usually different from its own.
The members of such a group are, or feel themselves, or are
thought to be bound together by common ties of race or nationality
or culture. ss
As this description suggests, there is no single criterion by which
50 Sodety and Politics in India

ethnic groups can be defined.

In the United States the term 'ethnic group' came into use
to describe immigrants from the different parts of the world.
Examples of these would be the Irish, the Italians and the Poles
who settled in the country in successive waves of migration.
These groups were not all differentiated by visible physical
indicators. Initially there were major differences of language,
culture and religion among the groups. As some of these dif-
ferences began to diminish among second- and third-generation
immigrants, it was felt that a culturally homogeneous population
would emerge out of the melting pot of American society. But
in spite ofa high degree ofmobility, both horizontal and vertical,
and a certain amount of intermarriage between groups, ethnic
identities have proved to be remarkably persistent in American
society. 56
The presence of ethnic groups is of course not a unique
feature of American society. They exist in all societies where
cultural differences are given a particular meaning and are
organised in a particular way. Ethnic differentiation has been
a conspicuous feature of the so-called plural societies of South
and South-East Asia. 57 Sometimes this differentiation is asso-
ciated with the presence of large groups, such as the Chinese
and the Indians in Malaysia, which differ markedly from each
other in language, religion and provenance. The co-existence of
such disparate groups is likely to generate tensions and conflicts
which might, in the extreme case, threaten the integrity of the
political framework ·itself.
Ethnic identities might persist even when ethnic groups are
not visibly different or politically organized. In a recent collec-
tion ofpapers Barth and his colleagues have argued persuasively
that ethnic identities do not depend for their survival on any
particular aggregate of cultural traits. 'It is important to rec-
ognise that although ethnic categories take cultural differences
into account, we can assume no simple one-to-one relationship
between ethnic units and cultural similarities and differences.' 58
Eidheim gives a graphic account of the manner in which an
ethnic boundary is maintained between Lapps and Norwegians
even in the absence of any readily visible physical or cultural
differences between them. 59
 Race, caste and ethnic identity 51
Ethnic groups are generally endogamous and in that sense
they tend to be biologically self-perpetuating. 60 Even in the
complete absence of diacritical distinctions endogamy could
of course serve to keep ethnic boundaries intact. When all
marriages do not take place within the group, ethnic boundaries
might still be maintained if intermarriage is governed by the rule
of hypergamy; the practice of hypergamy acts as an·important
boundary maintaining mechanism among certain sections of the
hill Rajputs in India. 61 Far from dissolving ethnic boundaries
altogether, intermarriage might under certain conditions serve
to bring these boundaries into sharper relief.
Thus, the concept of ethnic group is somewhat broader in
its scope than that of race. Ethnic differences might be based
at least partly on race as in the case of Malays, Chinese and
Indians in Malaysia or of blacks, Indians and whites in the
Caribbean. They might also exist in a society which is racially
more or less homogeneous as in the case of the Pathans in West
Pakistan and Afghanistan or of some of the multi-tribal systems
in East Africa.
The caste system, in its tum, may be viewed as a particular
case of ethnic differentiation. Whether or not 'racial' differences
exist between castes, they are often differentiated from each
other culturally, in their dress, diet and rituals. Where even
these distinctions are feeble or absent, the boundaries between
castes are maintained by the rules of endogamy and hypergamy.
However, even ifwe regard caste as a system of ethnic groups, it
is a system in which the different groups are all integrated within
a hierarchical order.. Ethnic groups are not necessarily arranged
in a hierarchy and they are not always integrated within a unitary
system.
•We notice a close similarity between caste in India and ethnic
groups in the United States when we examine the part they play
in the political process. 62 In the United States ethnic solidarities ·~
are widely used for mobilizing political support and ethnic
rivalries have to be taken into account in formulating electoral
strategies. 63 In India caste enters into the political process in
a number of ways. 64 Caste associations have not only acted
as pressure groups but, in at least one area, have transformed
themselves into political parties. 65 Rivalries between parties are
52 Society and Politics in India
sometimes heightened when they base their support on mutually
antagonistic castes. 66 However, in both India and the United
States the relationship between caste or ethnic identity and
the political process is complex and ambiguous. The political
process brings out not only the cleavages between such groups
but also the possibilities of coalitions among them.
The Harijans provide a particular example of solidarity based
on caste or ethnic identity. In the past the barrier of pollution
kept them segregated from many areas of social life. These
barriers have now been legally abolished but the Harijans retain
much of their traditional stigma and continue to be socially
and economically underprivileged. But they are now provided
with opportunities to organize themselves politically. 67 This has
enabled them to gain some advantages but it has also brought
them into confrontation with the upper castes whose members
are not always in a mood to accept them as equals. The situation
of the Harijans in contemporary India - like that of the blacks in
the United States - reveals a paradox. The lessening of cultural
distance has in both cases been accompanied not by a decrease
but by an increase in tension and conflict.
India has not only a Harijan problem, there is also an Adivasi
or tribal problem. Harijans and Adivasis are officially grouped
together as the Backward Classes and their separate identity is
given constitutional recognition. 68 The tribal people numbered
about thirty million at the 1961 census and they constituted over
6 per cent of the Indian population. They are divided into a
large number of separate tribes, differing in race, language and
culture. They are concentrated in particular areas in the country
which tend to be geographically isolated but there is no policy of
keeping them in reservations.
The tribal population of India does not belong to any single
racial or physical type. The differences between the 'Veddoid'
type common among certain tribes in central and south India
and the 'Palaeo-Mongoloid' type found in the north-east hill
areas might be greater than the differences between the tribal
people and their non-tribal neighbours in any particular area.
But Furer-Haimendorf has rightly pointed out that differences
of the latter kind also exist69 and Majumdar's anthropometric
data seem to point in the same direction. 70
 Race, caste and ethnic identity 53
After drawing attention to differences in physical type between
the tribal and the non-tribal population, Furer-Haimendorfsays,
'It is all the more remarkable that despite racial differences no
less fundamental than those found in countries with acute race
problems, there have never been any cases of racial tension in
India'. 71 One important factor is the very great variety ofphysical
types which has prevented a polarization of the population along
racial lines. This does not mean that differences do not exist or
are not socially recognized. In fact, tribal solidarity is perhaps
being given a new lease of life by democratic politics. But the
conflict is transferred on to a different plane where the cleavage
between tribals and non-tribals becomes one among a number
of politically relevant ethnic distinctions.
We have so far considered ethnic differentiation among
groups which are hierarchically arranged, for, although the
Adivasis are in the strict sense outside the caste system,
they are almost everywhere ranked below the caste Hindus.
We may now turn to ethnic diflerentiation between groups
which are not hierarchically arranged, such as those based on
religion or on language. In some sense these provide the most
fundamental cleavages in contemporary Indian society. When
one talks about 'national integration' in India one has primarily
in mind the problems of holding together the different religious
and linguistic communities. While one can distinguish analyti-
cally between ethnic identities of different kinds - hierarchical
and non-hierarchical - in reality these often tend to become
confused.
India has been described as a multi-religious nation. The
Hindus are in an overwhelming majority, accounting for around
80 per cent of the population; the Muslims constitute a sig-
nificant minority with a little more than 10 per cent of the
population. There are other religious groups which are of
significance in particular regions, such as the Sikhs in the
Punjab and the Christians in Kerala. But for the country as
a whole the cleavage which has greatest significance is the one
between Hindus and Muslims. If there is a 'communal' problem
in the country its prototype is the one which grows out of the
relations between these two communities. 72
Hindus and Muslims in India do not belong to separate
 54 Society and Politics in India
races. In fact, they are both racially very mixed. This is only
to be expected since the majority of Indian Muslims are the
descendants of converts from Hinduism. Spear argues that there
were two main types of conversion: clan or group conversion as
. a consequence of which castes such as Rajputs,Jats and Gujjars
in north India have Hindu as well as Muslim sections; and
mass conversions through which low-caste Hindus, particularly
in Bengal, embraced Islam. 73 The last point finds confirmation
in Majumdar's anthropometric data referred to above; the
low-caste Namasudras are closer in their physical appearance
to the Muslims than they are to the upper caste Hindus. 74
Hindus and Muslims have co-existed as communities in
different parts of India for a millennium. Religious differences
have been associated with a host of other differences in ways of
life. These differences have not always been the same, but the
fact of difference has remained, heightened at times and sub-
dued at others. Hindus and Muslims might not differ in physical
type but religious ideology has provided each community with a
basis for consciously organizing its identity in opposition to the
other. Over the centuries the two communities have borrowed
much from each other and during the last few decades they
have been exposed to similar forces of change. But this has
not erased the boundaries between them. In fact, the pattern of
Hindu-Muslim relations in recent Indian history would seem to
show that groups might become more conscious of their opposed
identities precisely at a time when external differences between
them are being reduced.
The population of India is also divided on the basis of
language. The divisions of language and religion generally
cut across and do not reinforce each other as they do to a
large extent in countries like Malaysia and Sri Lanka. This,
in addition to the fact that both Unguistic and religious groups
are many and not two each,. tends to make the conflict between
communities diffused rather than polarized.
Over a dozen major languages are spoken in India but there is
none which is the mother tongue of a majority ofthe people. The
speakers of the different languages are not randomly distributed
throughout the country. Each language has its 'homeland' so that
linguistic differences largely coincide with regional differences.
 Race, caste and ethnic identi'ly 55
The different states which constitute the Union of India are
in effect linguistic units. This means that the ethnic identity
provided by language has both a cultural basis and a political
organization.
Differences between linguistic groups can give rise to two
kinds of tensions. At one level are the disputes between the
different linguistic states over particular issues, for instance the
question of boundaries or the distribution of river water. 75 At
another level one encounters the problem oflinguistic minorities
in practically every state; these problems are likely to be
particularly acute in large metropolitan cities like Bombay or
Calcutta which attract people from all over the country. Ethnic
boundaries based on language are in a way crucial; they restrict
communication between people in the literal sense of the term.
Differences of language have in reality very little to do with
differences of race although in one important case linguistic dif-
ferences have been represented in a racial idiom. The different
languages of India belong to two major families, the lndo-Aryan
languages spoken in the north by about three-quarters of the
population and the Dravidian languages spoken in the four
southern states by about a quarter of the population. People
in the southern states have, particularly since independence,
sometimes expressed a fear of domination by the north 76
and a separatist political movement developed there although
its influence has been confined almost wholly to one state,
Tamilnad.77 One of the arguments advanced by leaders of
this movement was that South Indians, being Dravidians, had
a separate identity in race, language. and culture and should free
themselves from the domination of the Aryan North Indians. 78
Tamil separatism has now become subdued and one no longer
hears the racial argument very frequently but language barriers
are in other respects no less significant than they were before.

We have moved a long distance from a consideration of racial

differences to differences of quite another kind which are at
times expressed in a racial idiom. Ethnic identity must not be
thought of as something which defines the character of one
group in opposition to another for all time. In India the same
individual has a number of different identities according to
56 Society and Politics in India

caste, religion and language and any one of these might become
more important than the others, depending upon context and
situation. It is not enough to know that boundaries exist between
groups, one must also examine the situations under which some
boundaries are ignored and others become significant. Thus, in
one context Tamil speaking Hindus and Muslims might unite
to defend themselves against 'Aryan' domination; in another
context Hindus from both north and south India might regard
Muslims as aliens among them.
Although ethnic differences have a bearing on social conflict,
a knowledge of the former is not enough to predict the pattern
of the latter. In order to understand the scale and intensity of
conflicts between ethnic groups we have to take a number of
factors into account. These are (1) the objective differences
between them; (2) the social awareness of these differences; .
and (3) the political organization of this awareness.
As we have seen, the objective differences themselves are of
many kinds. They may be roughly grouped together as physical
or cultural. Cultural differences in tum can be based on religion,
language or region. There is no direct relationship between the
degree of these differences and the extent to which people are
aware of them. Differences of colour might exist to the same
degree in two societies and yet people might be acutely aware
of them in one society and not in the other. Cultural differences
are more difficult to measure. And, in any case, there are no
satisfactory criteria by which one can compare the awareness of,
say, religious differences with that of linguistic differences.
People might be highly conscious oftheir differences, whether
physical or cultural, without their consciousness acquiring a
political form. In traditional Indian society there were not only
differences between castes, but -people were universally aware
of these differences. Yet castes were not always organized into
mutually antagonistic groups. They began to organize them-
selves into associations at a time when people were beginning
to feel that caste consciousness would fade away. The course
of political conflict remairis unpredictable. There is no general
theory which can enable us to delineate in exact terms the
relationship between cultural differences and their organization
into mutually antagonistic groups.
 3 The concept oftribe with special
reference to India
Anthropologists have been from the very beginning engaged in
the study of tribes, and it is in some sense to this study that their
discipline owes its distinctive identity. When historians, political
theorists, sociologists and others have to deal with tribes, they
tum to anthropologists for expert opinion on what tribes are and
how they are constituted. In some countries what constitutes a
tribe is of concern also to administrators and policy makers, and
they too expect advice and guidance from anthropologists. Yet it
cannot be said that anthropologists are themselves in agreement
about the concept, and their disagreement is, if anything, even
larger today than it was in the past.
A possible reason for the widening of disagreement is that
anthropologists now study not only a wider range of topics but
also a wider variety of societies. Until the 1930s or 1940s they
confined their attention mainly to simple, pre-literate, small-
scale and isolated societies in Australia, Melanesia, the Pacific
Islands, North and South America, and sub-Saharan Africa.
This was true as much of Boas, Lowie, Wissler and their pupils
in the United States as of Rivers, Radcliffe-Brown, Malinowski
and their pupils in the Commonwealth. The tribe was here the
centre and focus of attention and, except where note had to be
taken of recent changes introduced as a result of exposure to
European culture, it was considered as a self-contained unit.
A major change of orientation came about in the 1940s
when anthropologists began to claim that their discipline had
a distinctive contribution to make to the understanding of not
only tribes but also civilizations. This move was first made in the
United States where the work of Robert Redfield 1 stands as a
kind ofwatershed, although he had an illustrious predecessor in
A.L. Kroeber. 2 Redfield popularized the study of the peasantry,
especially in Latin America and Asia, and the view that peasants
58 Society and Politics in India
are 'part societies' and 'part cultures'.3 The interest in studying
large wholes gathered strength from other sources as well, and
the study of civilizations in India, in the Arab world and in
China, Japan and Indonesia has become as important as, if not
more important than, the study of tribes.
The study of civilizations has given anthropologists a new
awareness of the importance of history. The anthropologist
who studies tribes in India today does not necessarily confine
himself to the present or even the recent past, but might try
to go back to medieval if not ancient times. 4 An important
issue in the study of tribes today is how we understand the
relationship between tribe and civilization. It is here that I
would like to make a distinction between two approaches
which I will describe as the 'evolutionary' and the 'historical'.
The evolutionary approach takes a long-range view of the
passage of time and stresses the succession of social formations.
Evolutionists no doubt recognize the presence of survivals, but
these are regarded as anachronisms, which they probably are
on a sufficiently extended time scale. The historical approach
limits itself to a particular fr~ework of space and time and
stresses the co-existence of different social formations within
that framework. What is regarded an an anachronism in the
evolutionary perspective may appear as a necessary component
in the historical framework.
Morgan's conception of the tribe 5 and Durkheim's concep-
tion of the polysegmental society6 were both rooted in the
same evolutionary perspective. Their successors chose their
examples not from India, China and the Islamic world, but from
Australia, the Pacific Islands and North America where recent
historical experience brought out the disjunction rather than
the co-existence of tribe and civilization. Western civilization
penetrated these areas in modem times, and tribe and civiliza-
tion stood opposed there in every particular of race, language
and culture. In many parts of the Old World the situation
was different. There tribe and civilization had co-existed for
centuries if not millenia, and were closely implicated in each
other from ancient to modem times. In India the effort to
disentangle tribe from caste began in a systematic way during
British rule and led to unforeseen results.
 The concept oftribe with special reference to India 59
Ethnographic material from India did not figure prominently
in the general discussion regarding the definition of tribe.
The problem in India was to identify rather than define
tribes, and scientific or theoretical considerations were never
allowed to displace administrative or political ones. This is
not to say that those engaged in drawing up lists of Indian
tribes did not have their own conceptions of tribe, but those
conceptions were neither clearly formulated nor systematically
applied.
Lists of Indian tribes were, in fact, drawn up, with or without
benefit of clear and consistent definitions. These lists are not
only in current use, but provide constitutional guarantee of
tribal identity to those included in them. The present list
shows more than 400 tribes with an aggregate population
of over fifty million persons accounting for 7.76 per cent
of the total population. Even a cursory look will show that
it includes the widest variety of social formations, from small
food-gathering bands to vast populations of settled agricul-
turists comprising three million persons and more. 7 Indian
anthropologists have been conscious of a certain lack of fit
between what their discipline defines as 'tribe' and what they
are obliged to describe as 'tribes', but they have sought a way
out of the muddle by calling them all 'tribes in transition'. 8 This
does not settle the issue because in India tribes have always
been in transition, at least since the beginning of recorded
history.

Among recent attempts to provide a general definition of

tribe, those by Ellman Service9 and by Marshall SahlinsIO
have received some attention. What is interesting about their
definitions is that they are part of a general classification ofsocial
formations set in an evolutionary framework. Underlying the
classification is the dichotomy between 'State' and 'non-State'
societies, but non-State societies are not viewed as being all
of the same kind: they are divided into bands, tribes and
chiefdoms.
The principal novelty in this scheme, it seems to me, is
not the distinction between band and tribe or between tribe
and chiefdom, but the definition of the tribe as a segmentary
60 Society and Politics in India

system. As Sahlins put it, 'A tribe is a segmental organiza-

tion. It is composed of a number of equivalent, unspecialized
multifamily groups, each the structural duplicate of the other:
a tribe is a congeries of equal kin group blocs' .11 Although
the study of tribes had been the staple of Anglo-American
anthropology from Morgan to Malinowski, the concept of
segmentary system entered the scene relatively late, with the
publication in 1940 of The Nuer12 and of African Political
Systems. 13 It is true that the idea of segmentary system had
been used extensively in the 1890s by Durkheim, but it did
not catch the imagination of students of tribal societies until the
1940s.14
The idea of segmentary system has had a very wide appeal
among anthropologists working in different parts of the world.
It has given a focus to the study of tribes in the Islamic
world, particularly in the Maghreb. 15 It has enriched our
understanding of the social system of the Pathans on what
used to be the north-western frontier of India although there,
as Barth has shown, tribe is already to some extent inter-
woven with caste. 16 There are, however, large parts of the
world where the segmentary principle by itself does not greatly
illuminate our understanding of what are commonly recognized
as tribes. A tribe is best described as a segmentary system
where it has a particular type of clan and lineage structure
most fully developed among the Arabs but also found .in
parts of sub-Saharan Africa and elsewhere. There are various
difficulties in making the segmentary principle as described by
Evans-Pritchard in his studies of the Nuer and the Cyrenaicans
the defining feature of tribes in all parts of the world. On
the one hand, Sahlins himself argues for a very restricted
use of the term 'segmentary system'; 17 and, on the other,
the contributors to African Political Systems give no indication
of using 'tribe' and 'segmentary system' as interchangeable
terms.
These or similar considerations must have weighed with
Sahlins, for in 1968 he modified his original fourfold scheme
of 1961 into a threefold scheme, comprising band, tribe and
State. The category 'tribe' was expanded to include tribal
chiefdoms in addition to the segmentary tribes which had
 The concept oftribe with special re[erence to India 61
earlier made up the whole category. It is not as if Sahlins
saw no difference whatever between segmentary and chiefly
tribes but he felt on reconsideration that the difference entailed
no discontinuity, that they were 'permutations' of the 'same
general model'.18 But, as Godelier has rightly pointed out,
if Sahlins went so far in one direction to include chiefdoms
in the category of tribes, he might have gone a little in the
other direction to include bands as well within the same
category. 19
The need to define tribe in an evolutionary perspective has
been reiterated by Godelier in a recent critical essay. For
Godelier the tribe is at one and the same time a type of
society and a state of evolution. As he has put it, 'The link
between the two uses ofthe term tribe, seen as a type ofsociety and
as a stage ofevolution, is very clear since, each stage of evolution
is characterised by a specific mode of social organisation'. 20
He attributes the failure of anthropologists to arrive at a clear
conception of the tribe to their lack of a consistent theory of
evolution.
For Godelier there is a fundamental difference between tribe
and chiefdom because inequalities of class which are absent in
the former become established in the latter. Work is organized
differently in the two types of society and the primacy of
kinship is undermined by the appearance of class in the
evolution from tribe to chiefdom. 21 The assumption here is
that wherever we find a chiefdom we will also find a division
into classes. This, as the editors of African Political Systems had
warned us, is far from the case. Even in the African chiefdom,
'Distinctions ofrank, status or occupation operate independently
of differences of wealth'. 22 Moreover, as recent students of the
subject have realized, a segmentary system need not necessarily
be acephalous. 23
Godelier's essay exposes the limitations of the evolution-
ary point of view. The evidence is too thin to support his
assumption of the uniform co-variation of mode of liveli-
hood, kinship structure and political system within the range
of societies with which we are concerned. Both segmentary
tribes and tribal chiefdoms are supported by various modes of
livelihood.
62 Society and Politics in India

The Tallensi and the Bemba are both agriculturists, the Tallensi
having fixed and the Bemba shifting cultivation, but they have very
different political systems. The Nuer and Logoli of Group B and
the Zulu and Ngwato of Group A alike practise mixed agriculture
and cattle husbandry.24
There is, if anything, an even greater range of variation in the
nature and function of kinship in each of the two types of
society.
It would be a great convenience if some simple scheme could
be devised for classifying the 400 odd communities designated
as tribes in India. Various classifications have been proposed
on the basis of ecological, racial, linguistic, religious and other
criteria. Perhaps the most convenient is the one persistently
recommended by N.K. Bose on the basis of mode of livelihood:
hunters and gatherers, animal herders, shifting cultivators and
settled agriculturists. 25 But very little is to be gained by trying to
fit this kind ofclassification into any rigid evolutionary scheme. It
is now being increasingly recognized that tribes with some ofthe
simplest technologies have been more closely integrated with the
wider society than others with a more advanced technology. 26 If
isolation, self-sufficiency and autonomy are characteristics of the
tribal condition, there is no simple correlation between these and
the level of technology.

Not all anthropologists who write about tribes adopt an evol-

utionary perspective. Morton Fried has presented us with a
rather different notion of tribe.27 He has not only criticized
existing definitions but argued that the very concept of tribe,
so extensively used by anthropologists, is inherently ambiguous.
His criticism is very far reaching, and we must examine briefly
the factors behind the ambiguities it has exposed.
It was Morgan's argument, which Godelier repeats with
approval, that the tribe is a 'completely organised society'. 28
When we speak of the tribe as a 'completely organized society',
we assume that there are boundaries separating tribes from each
other and from other types of society. What· is the nature of
these boundaries? Can they be easily recognized? Do they
really exist? No student of tribal life in India can afford to
 The concept oftribe with special reference to India 63
ignore these questions. But the boundary problem has its own
peculiar features in India, for what is unclear there is not so
much the boundary between one tribe and another as that
between tribe and non-tribe. The boundaries between tribes
are perhaps a little more clear than Fried would allow for,
because in India tribes have since time immemorial lived in the
shadow of a civilization which has been strict about maintaining
the boundaries between castes.
Fried rejects the argument, apparently popular among some
physical anthropologists, that the tribe is a breeding population
whose boundaries circumscribe the range of sexual relations. 29
He marshals evidence to show that not only unsanctioned sexual
unions but also socially approved marriages frequently take place
across the boundaries of what are called tribes. But the point
about outmarriages ought not to be taken too far. There are
probably genuine differences here between, let us say, Amazonia
and South Asia. In India tribes such as Khasi, Garo, Santai,
Munda, Oraon and many others are known to have a marked
preference for endogamy which is not surprising in view of
the high value placed on endogamy everywhere in India.30 As
I shall point out later, there have been innumerable cases of
tribes becoming castes - the Indian equivalent of 'passing' -
and when this happened, the rule of endogamy was probably
reinforced.
Fried also questions the test of language, or the view that
each tribe has its own distinctive language which defines its
boundaries. 'The idea,' he says, 'that tribes, whatever else they
may be, are somehow minimal speech communities, turns out to
be no sounder than the notion that they are basic breeding popu-
lations'. 31 There are well-known examples, even in small-scale
societies, of people speaking the same language being divided
into several endogamous groups as well as examples of people
intermarrying though their native languages are different. Again,
the significance oflanguage in the definition of tribe will depend
on the extent to which the tribe is implicated in the civilization in
its vicinity.
In India the test of language has always been an important
one in the identification of tribes. There are fifteen officially
.recognized languages listed in the Eighth Schedule of the
64 Sodety and Politics in India
Constitution of which four belong to the Dravidian family and
the rest to the lndo-Aryan family. Besides these, there is an
assortment of several hundred languages, usually not counted
as literary languages, and spoken by smaller populations, though
in some cases these may comprise as many as a couple ofmillion
persons each. By and large, it is the communities associated with
these languages that are recognized as tribes in India.
Two brief comments must be made here about the relation-
ship between civilization, language and tribe. The Dravidian
languages of India include not only Tamil, which is one of the
oldest literary languages of the world, but also languages spoken
by a number of tribes such as the Baiga and the Kond who
lived by shifting cultivation until the other day. The Tamils are
proud of their ancient and medieval civilization which created
elaborate irrigation works on the one hand, magnificent temples
on the other, but their cultural affinity with some of the simplest
tribes of peninsular India would appear to be beyond dispute. In
a recent impressive work, Trautman has shown how the Tamils
share the same fundamental structure of kinship and affinity
with the Baiga and the Kond. 32 Further, Dravidian kinship, in
both its 'civilized' and 'tribal' forms, is markedly different from
North Indian kinship.33 Nothing could demonstrate more effec-
tively the hazards of using kinship as a basis for discriminating
between 'tribe' and 'civilization' or 'tribe' and 'State'.
The second point to note is that some of the tribes, including
a few very large ones, have no separate language of their own
but use the language prevalent in the region they inhabit. This
is particularly common in the western part of India, in the states
of Rajasthan and Gujarat. There obviously has been a loss of
language in some cases but it is impossible to date this loss in
most cases. When the loss of language is accompanied by a loss
of other cultural traits, a sort of invisible threshold is crossed,
and the tribe ceases to function as a tribe although it does not
thereby lose its identity as a community.
Economic relations can of course be easily shown to overflow
the boundaries of the tribe, no matter how we define 'tribe'.
Fried would argue that the political boundaries of what are
called tribes are usually far less clear than is assumed in the
model of either the tribal chiefdom or the segmentary tribe.
 The concept oftribe with special reference to India 65
It may be recalled that Evans-Pritchard's original model of
the segmentary political system was presented as a kind of
solution to the boundary problem among the Nuer who lack
all constituted political authority. 34 If a conflict took the form
of feud, it was within the tribe; if of war, it was between one
tribe and another. This method ofsolving the boundary problem
cannot be easily applied everywhere, for even if we admit that
the segmentary principle is a kind of universal principle, its
operation does not have everywhere the same consistency and
constancy that Evans-Pritchard reported it to have among the
Nuer. The model of the tribe as a segmentary system is a
tempting one, but one faces many pitfalls in yielding to the
temptation.
Fried's argument would therefore be that what are generally
designated by anthropologists as tribes represent neither a
definite type of society nor a definite stage of evolution. They
are too amorphous and too assorted to qualify for either role.
The tribe, according to Fried, is much better regarded as a kind
of secondary phenomenon which in the typical case acquires its
form and identity from some external source.
While being bold, I shall go on to say that most tribes seem
to be secondary phenomena in a very specific sense: they may
well be the product of processes stimulated by the appearance of
relatively highly organized societies amidst other societies which are
organized much more simply.JS
The formation of a tribe as a secondary phenomenon is
familiar to anthropologists. Perhaps the best known example
in the literature is that of the Makah Indians who, according
to Elisabeth Colson, were given their present identity by the
Indian Service of the American Government: 'The people
regarded as Makah, by themselves and by those who are not
of their group, are such by a political definition framed to
organize a group of people with political rights as members
of the Makah Tribe'. 36 Extending her observations to another
continent, Colson argues that 'contemporary African tribes are
either new forms of political organization created for adminis-
trative purposes by the modem states within which they exist
or they represent the emergence of self-conscious nationalistic
66 Society and Politics in India
movements comparable to those of Europe and Asia' .3 7
It is true that in some parts of the world colonial rule has had
cataclysmic effects, dissolving old identities and precipitating
new ones. But to say that tribes as we now know them are all
of recent origin or are all secondary phenomena would be to
take a holiday from the lessons of history. Tribe and civilization
have encountered each other for centuries in many parts of the
world, and it is only in some areas and at certain periods that the
encounter has been sudden and cataclysmic.
In the historical relations between State and tribe, the State
has not by any means always had the upper hand. Not only has
it often lacked the strength to resist encroachment by tribes, but
the tribal way of life has appeared to many as superior to that
represented by the State. Writing about the Pakhtuns in what
used to be the north-western frontier oflndia, a recent observer
notes, 'Indeed as far as they are concerned, it is "state" and
not "tribe" which occupies the periphery of things, and it is
to the state all the characteristics of the peripheral attach, most
especially dissipation'. 38
We should neither ignore the civilization in the background
when we write about tribes, nor argue as if the relationship
between tribe and civilization were the same everywhere. It is
here that we see the flaws in both of the two schemes proposed
by Sahlins. It is one thing to make a direct contrast between
tribe and State. Sahlins does not stop there but proceeds
to divide tribes into tribal chiefdoms and segmentary tribes,
adding bands at the other end. Surely, this should oblige
him to consider corresponding distinctions among what he
calls states, for these differ among themselves as much as do
stateless societies. To take two neighbouring countries, the
Soviet State is very different from the State in Afghanistan
in many respects, and certainly in relation to tribes. It may be
preferable in a historical context· to speak of civilization rather
than State as the complement of tribe, for certainly in the Indian
case, civilization has shown far greater unity and continuity than
any state.

The co-existence of tribe and State has been discussed by

students of the Islamic world from lbn Khaldun down to the
 The concept oftribe with special reference to India 67
present.3 9 Indeed, what has been stressed in their writings
is not merely the co-existence of tribe and State but their
complementarity. As a recent student of the Middle East has
put it, 'The tribes have always been a part of, as well as being
in varying degrees apart from, the Iranian state'. 40 Given the
acknowledged significance of the tribal component in the origin
and growth of the Islamic State, it would be misleading to refer
to these tribes, whether in lbn Khaldun's time or in our own, as
secondary phenomena. Nor can we regard them and the states,
which for centuries they regularly overran, as two separate stages
of social evolution.
The Islamic tribes were external to the Islamic State in a
way in which they were not external to Islamic civilization.
Conversely, subscribing to the values of Islam did not require
submission to the discipline of the State. One might even say
that some of the most fundamental values of Islam - equality,
community, brotherhood - found a more authentic expression
in the 'tribal' than in the 'civilized' sector oflslam. The Islamic
case teaches us to use the concept of civilization in two senses:
in the first or unmarked sense civilization includes tribe and in
the second or marked sense it is contrasted with tribe.
India has unfortunately had no Ibn Khaldun to record and
reflect on the relationship between tribe and civilization in
pre-modem times. The Indian intellectual tradition, which is
both ancient and rich, is remarkable for its lack of historical
sense, and hardly any history was written before the advent of
Muslim rule. 41 We are therefore forced to rely on conjecture
and reconstruction to a far greater extent than in the case of
Europe, China or the Islamic world.
Both historians and anthropologists have noted that in tra-
ditional India tribes were not only recognized to exist but
were given a definite designation: jana as againstjati.42 How-
ever, it is not easy to determine the exact connotation of the
term jana, and the distinction between jana and jati must
have been even less clear in ancient times than the cor-
responding distinction today between tribe and caste. Each
category was heterogeneous and there was always some overlap
between the two. The historian Niharranjan Ray has noted
that:
68 Society and Politics in India

in Indian historical tradition there were two sets of janas, one

who are still recognised by anthropologists and sociologists as
tribes... and another set who were at a relatively higher level of
socio-economic and political organization and of aesthetic and
religious culture.43
It would be rash to seek to identify these two sets of jana with
segmentary tribes on the one hand and tribal chiefdoms on the
other, but it is clear that some of them founded states and joined
the mainstream while others either remained isolated or were
pushed into marginal areas.
Historians of both ancient and medieval India have spoken
repeatedly of the rise to power of tribal dynasties in various
places and at various times. Some of these, like the Ahom in
the thirteenth century, came as intruders from outside. Others,
like the Chandela, allegedly of Gond origin, rose to power from
within. It is obvious that the term 'tribe' has been used in a
loose sense, meaning different things to different historians,
but it is possible to reconstruct with a degree of accuracy the
tribal origin of some at least of the ruling dynasties of the
pre-British period.
The rise to eminence of a tribal dynasty did not lead
necessarily or even generally to a radical change in the mode
of life of the tribe as a whole. Only some sections of it
would become Hinduizr.d while others might survive more or
less in their previous condition. What is characteristic of the
relationship between tribe and civilization in India is that there
was virtually no way in which a tribal dynasty could legitimize
its rule without becoming Hinduized. This meant, among other
things, bringing in Brahmin priests, Barbers, Washermen and
the rest, and replicating in due course of time the hierarchical
structure of caste. Although only the ruling families or lineages
became fully Hinduized, some ofthis rubbed off on their poorer
cousins who might continue in their previous tribal condition,
sometimes in the remote hills and forests. This kind of survival
cannot in any meaningful sense be described as a secondary
phenomenon even where it was affected by influences from a
newly established kingdom.
As the State became more powerful and society better
 The concept oftribe with special reference to India 69
organized, the scope available to tribal lineages to establish
new dynasties became more restricted. But the State was not
the most durable product of Hindu civilization. The weakness
and decay of states always left room for the emergence of tribal
chiefs whose aim was to create not tribal chiefdoms so much as
kingdoms after the Hindu model.
When a tribal lineage established a Hindu kingdom there
occurred what may be described as integration at the top.
Although the historical record is likely to furnish evidence of
integration mainly at that level, what must have been far more
common is integration at the bottom.44 This kind of integration
is by its nature difficult to document for the ancient and medieval
periods. It took place whenever a tribe or a section of it, usually
through force of economic circumstances, became involved in
the larger division of labour by providing specialized products
associated with their habit, such as lac, honey, ropes, baskets,
mats, etc., or a regular supply of manual labour.
Despite changes in the fortunes of individual tribes and
despite incursions into tribal territories by Hindu kings and
Hindu ascetics, the tribal identity never became fully effaced
in any of the major regions of the country. It is remarkable
how close to such renowned ancient and medieval centres of
civilization as Gaya, Ujjain and Madurai tribes could still be
found living in their natural setting, so to say, well into the
present century. The Hindu kingdom - and to a large extent its
Muslim successor - did not seek to eliminate tribes but allowed
or even encouraged them to live on its margins. This is not a
setting in which one can proceed very far by viewing tribe and
state as two distinct and successive stages of evolution.
Students of Indian society and history have been struck
repeatedly by the presence of survivals at every level. As D.D.
Kosambi, the historian of ancient India has put it, 'India is a
country of long survivals'. 45 Kosambi tried to use these survivals
for developing a method for reconstructing, India's past. Our
concern is not so much with that method, whose limitations
are well known, as with the co-existence down to our times of
diverse social formations, tribal as well as non-tribal.
Kosambi provides a fascinating thumbnail sketch of this
co-existence in his own time - the 1950s - and around his
70 Society and Politics in India
own home in the city of Poona.46 Nearest in location he found
a group of tent-dwelling nomadic families belonging to the Ras
Phase Pardhi tribe, divided into six exogamous clans bearing
Brahmin and Rajput names; they led a precarious existence
by snaring small animals, working as casual labourers and
begging. Then there were Ramosis who were rapidly discarding
their 'tribal' customs and becoming like the general Maratha
peasantry in appearance, language and religious observance. In
addition to these there were two Telugu speaking tribal groups,
the Vaidu who were snake-charmers and medicine men and the
Vaddar who worked as stone cutters.
It is now widely acknowledged that what were until recently
regarded as hunting and gathering tribes were in many cases
reduced within the last hundred years or so to an economic
condition in which they are forced to survive by foraging,
begging, thieving and other such activities. 47 These would
appear to correspond to the kind of secondary phenomena to
which Fried has so convincingly drawn attention. It is almost
certain that such secondary phenomena have increased in scale
and intensity from the·middle of the nineteenth century to the
present, and the idea of 'ex-criminal tribes', if not of 'criminal
tribes', sounds definitely modem. But it would in my view be
a mistake to believe that such secondary phenomena have no
precedents in the past. Indeed, there is reason to believe that
both detribalization and retribalization occurred in the distant
as well as the recent past. But they did not alter everything,
for collective identities often outlived changes in economic and
political fortunes.

I have so far been comparing tribe with State - or tribe with

civilization - in a very general way. I must now make a little
more specific my comparison of the morphological features
of tribal society with those of Hindu society. But what is the
morphological analogue oftribe in Hindu society? Is it the whole
of Hindu society, or is it the individual caste which is, as it were,
the building block of that society?
As soon as we enter into a close comparison, we are struck
by a paradox. When we place tribal society beside Hindu
society as a whole with its elaborate arrangement of castes,
 The concept oftribe with spedal reference to India 71
we observe the sharpest possible contrast. Tribal society is
homogeneous, undifferentiated and unstratified; Hindu society
is heterogeneous, differentiated and stratified. The polarity of
equality and hierarchy is much more clearly represented here
than in any other comparable case. Muslim society would appear
to stand at the other end of the scale, at least on the plane
of values.
When, on the other hand, we compare individual castes,
which are the constituent units of Hindu society, with indi-
vidual tribes, we observe a certain homology. Caste and tribe
emphasize and perpetuate collective identities in strikingly
similar ways; a caste or tribe may change its name and,
within limits, also its mode of livelihood and yet retain its
collective identity. Other societies, whether Islamic or Chris-
tian, are not made up of segments that at least outwardly
resemble tribes to such a large extent. Traditional Hindu
society was at one and the same time both hierarchical and
segmental. It is no accident that observers down the ages
have so persistently mistaken castes for tribes, and tribes
for castes.
From the end of the nineteenth century onwards, ethnog-
raphers, who were mainly British civil servants working in
India, began to publish descriptive catalogues of the tribes
and castes in the different parts of the country.48 A perusal
of these will show how unclear the line of division some-
times is between tribe and caste. There are, firstly, castes
of tribal · origin in areas in which the caste-based division
of labour is well established, and these include both agricul-
tural and artisan groups. But one encounters also the con-
verse phenomenon, namely, the growth of occupational spe-
cialization and the emergence of caste-like groups in the
interior of a tribal area. I am not saying that it is gener-
ally difficult to distinguish a tribe from a caste, only that
the difficulty often arises at the margins of Hindu civiliza-
tion.
A number of historians have argued that the Hindu social
structure owes its uniqueness to the manner in which it was
built up, block by block so to say, by the accretion of tribes.
D.D. Kosambi has put it in the following words:
72 Society and Politics in India

The entire course of Indian history shows tribal elements being

fused into a general society. This phenomenon, which lies at the
foundation of the most striking Indian feature, namely caste, is also
the great basic fact of ancient Indian history.49
Kosambi was a pioneer among Indian historians in carefully
observing the present as a way of understanding the past.
Professor lrfan Habib, our leading authority on Moghul India,
has in a recent paper tried to follow Kosambi's lead in tracing the
origin ofcaste. 50 Habib's argument is that the structure oflndian
society was not fundamentally different from the structure of
other societies, such as Safavid Iran, at comparable levels of
material advancement. How then did India come to have a
caste system of such great rigidity? Habib makes the interesting
point, contra Dumont, that virtually all outside observers, from
Megasthenes to Bernier were struck not by the "hierarchy"
of the system, but its hereditary occupation'. 51 In addition to
this, he stresses rigid endogamy which he regards as a 'tribal'
characteristic. The caste system was in this view the outcome of
the fusion into the general society of tribal communities which
were from the start rigidly endogamous.
The argument about the caste-based social order in India
being built up by the cumulative accretion of tribal commu-
nities answers some questions but raises many others. Tribal
components have contributed to the origin and development
of civilizations everywhere: why did the accretion of tribes lead
to the formation of a caste system only in India? Why did the
fusion of tribal elements into the general society not lead to the
formation of a caste system in medieval Morocco or medieval
Poland?
The argument that a strict rule of endogamy was carried over
from tribe to caste is not wholely convincing. As Fried has ably
demonstrated, endogamy is not a universal characteristic oftribes.
If there is evidence of strict endogamy among Indian tribes, that
may be an Indian rather than a tribal characteristic. Hindu society
no doubt carries the marks of tribal culture, but tribal society also
carries the marks of Hindu culture in India as it does of Muslim
culture in Morocco. The Indian case reveals not only the co-
existence of tribe and civilization but also their interpenetration.
 The concept oftribe with special reference to India 73
What is important therefore is not that tribal elements fuse
into the general ·society but that collective identities survive
the conversion of tribe into caste. It is this process that N.K.
Bose set out to analyse in his brilliant paper published forty
years ago on 'The Hindu method of tribal absorption'. 52 The
argument of that paper was elaborated in a book first published
in Bengali and later translated into English as The Struaure of
Hindu Society. 53
I must repeat before presenting Bose's argument that there
are over 400 named tribes in India whose conditions vary so
much that it would be naive to expect the argument to apply
equally well in all cases. At one extreme are the indigenous
tribes of the Andaman Islands - the Onge, Jarawa and others
- who, until the nineteenth century, remained almost completely
isolated from the mainland and therefore unaffected by the
Hindu method of tribal absorption. Then there are the tribes
in the north eastern hill areas - Konyak, Abor, Dafla and many
others - who, because of their location on the frontier of more
than one civilization, were better able to withstand the pressure .
to become castes, although the Ahom, now regarded as a caste,
were once clearly a tribe, and the Khasi, still regarded as a tribe,
were developing a state with unmistakably Hindu features. 54
Even in the north west, in what is now Pakistan, the Swat Pathan,
though primarily a group of Muslim tribes, have divisions that
are not wholely unlike Hindu castes. 55
The tribes that have been affected the most by the Hindu
method of tribal absorption are the ones in the interior hill and
forest areas where influences from other civilizations, whether
Islamic or Chinese, have been feeble or absent. These tribes
comprise a large array - Bhil, Munda, Santai, Oraon, Saora,
Juang and numerous others - and account for the bulk of the
tribal population of the country. It was Bose's argument that
there was a symbiotic, though unequal, relationship between
these tribes and the larger society of castes which became
apparent as soon as one viewed them on a sufficiently broad
geographical canvas and a sufficiently long historical scale.
Bose took pains to describe the economic context of the
symbiosis between tribes and the wider society, arguing that
the absorption of the former by the latter was generally of some
74 Society and Politics in India
material advantage to both, though not in the same way or to the
same extent. He believed that the caste-based economy and its
division of labour enabled it to support populations of greater
size and density than in a tribal economy whose material base
was at a lower level. When that material base became precarious
due to expansion of population or for some other reason, a
tribe or a section of it sought economic security through closer
attachment to the wider society. This attachment was generally
granted by the wider society on the condition that the newly
attached group took the lowest position in it. Bose combined
fieldwork in the tribal areas with a study of the classical texts
to establish his argument.
Bose believed that the whole order of Hindu society, since
at least the time of the Mahabharata, had been structured in
such a way that a tribe or a section of it was not only allowed
but encoura,ged to maintain a hereditary monopoly over its
occupation. Whole groups would thus function as basket makers,
or rope makers, or collectors of lac, or dealers of some other
product, usually of the forest. But, while specialized occupations
were extremely important in maintaining boundaries between
groups, such boundaries could be maintained even in their
absence through the enforcement of strict rules of endogamy.
Bose laid great emphasis on the economic ethic of the wider
society which put a high value on hereditary occupations, pro-
tected the occupational monopolies of groups and discouraged
competition between individuals in the occupational sphere.
The distinctiveness ofwhat has been called the Hindu method
of tribal absorption is seen when we examine the American
approach to the assimilation of tribes till the time of the Indian
Reorganization Act of 1934. Here I turn again to Colson's study
of the Makah Indians. The goal of the Indian Service, she
tells us, 'was the complete assimilation of the Indians, and
therefore the Makah, into American society in as short a period
as possible', and this goal would be frustrated 'in so far they
emerged from the moulding process one jot different from the
ideal average American'. 56 In the event, the goal was frustrated
as it was bound to be: 'Today American Indian tribes continue to
exist. They are deviant groups within American society'. 57 The
Hindu method of tribal absorption did not seek to efface the
 The concept oftribe with special reference to India 75

tribal identity fully or in the shortest time, and the end product
was a caste, usually of the lowest rank but not a deviant group.
Bose has been criticized for dwelling too much on the symbi-
otic nature of the relationship between the tribe and the wider
society and not enough on its asymmetrical and exploitative
character.58 The precariousness of the tribal economy was not
always its natural condition. It resulted sometimes from a tribe
being pushed back by its more prosperous or better organized
neighbours from a better to a worse location where it could
survive only on the lowest economic plane. As a recent writer
has tersely observed about a major area of tribal concentration
in Bihar, 'it can be surmised that the people of Chota Nagpur
remained primitive so that their neighbours could grow'. 59 But,
whether these tribes were pushed out or pulled in, their lives
were never wholely unaffected by the larger currents of Hindu
civilization.
Our understanding of the transformation oftribe into caste or
the fusion of tribal elements into the general society will remain
incomplete without an appreciation of the role of the economic
ethic of Hinduism. Evidence of oppression and exploitation
cannot be used to discount its hold over the minds of people
in the entire subcontinent down to our own times. The economic
ethic was part of a wider system of beliefs and values which
we describe broadly as Hinduism and which overflowed the
boundaries of Hindu society in the narrow sense.
When a tribe stood on the threshold of closer economic
interaction with Hindu society, it would normally not be a
complete stranger to the beliefs and values of the society on
whose threshold it stood. Hindu saints and ascetics have from
time immemorial gone into the remotest hill and forest areas in
the pursuit of their religious vocation. It would be a truism to
describe the forest as a category of Hindu civilization. The forest
and its people had as important a place in it as the desert and its
inhabitants had in Islamic civilization. Just as the desert was the
great source of political renewal in Islam, so was the forest the
perennial source of religious renewal in Hinduism.
Sometimes the forest retreat of a saint or an ascetic became
a famous centre of pilgrimage, ensuring a regular flow of traffic
through the tribal area surrounding it. This traffic might
76 Soci.ety and Politics in India

influence the religious life of the tribal people even when

they remained largely outside the organization of economic
activities based on caste. There is evidence of deep-rooted
and widespread Hindu beliefs and practices throughout the
tribal areas in the interior of India. 60 Within the traditional order
strict sanctions were developed for maintaining the bounda-
ries between one community and another, but there were no
comparable sanctions for maintaining the boundaries between
Animism and Hinduism, or between tribe and Hindu society.

In the light of what I have said above, it will be easy to

see why I prefer the historical to the evolutionary approach
in the definition and identification of tribes. Where tribe
and civilization co-exist, as in India and the Islamic world,
being a tribe has been more a matter of remaining outside
of State and civilization, whether by choice or necessity, than
of attaining a definite stage in the evolutionary advance from
the simple to the complex. We cannot therefore dismiss as
anomalous the Indian practice of regarding as tribes a large
assortment of communities, differing widely in size, mode of
livelihood and social organization. They are all tribes because
they all stood more or less outside of Hindu civilization and
not because they were all at exactly the same stage of evolution.
Similarly, a recent student of tribes in Iran and Afghanistan
has noted that they range 'from fragmentary and independent
communities somewhat resembling the bands of hunting and
gathering peoples, to centralised chiefdoms involving hundreds
of thousands of people, considerable differentiation of wealth
and status, and many of the trappingS of states'. 61
I have sought to stress the permeability of the boundary
between tribe and non-tribe, not to deny the presence of
tribes in either the past or the present. The permeability
of the boundary in India, in the Islamic world and perhaps
also in China obliges us to adopt a flexible rather than a rigid
attitude towards the definition of tribe. It makes the presence of
borderline cases an inescapable feature of the system, but does
not permit us to argue as ifall cases were borderline cases. It may
be difficult to decide whether the Bhumij in eastern India or the
Dubla in western India are a tribe or a caste, but there should be
 The concept oftribe with special reference to India 77
no difficulty in deciding that the Vadama are a caste in Tanjore
or the Juang a tribe in Mayurbhanj.
The traditional social order tolerated or even encouraged
the proliferation of borderline cases, but the modern State
cannot afford to do so. It demands clear categories in place of
ambiguous ones. The Indian Constitution now recognizes the
Scheduled Tribes as a separate category with specific claims and
entitlements. The State has thus an obligation to list and label
the tribes for whose benefit it has adopted special measures, and
these tribes have in their turn acquired a new interest in being
listed and labelled.
The process of designating or 'scheduling' tribes in India
began during British rule and acquired a systematic character
from the time of the 1931 census. It became involved in political
controversy from almost the very beginning. On the one side
were the official anthropologists, mostly British members of the
Indian Civil Service, who argued that the aboriginal tribes had
a distinct identity that marked them out from the rest of Indian
society. 62 On the other were the nationalist anthropologists who
argued that they were part and parcel of Hindu society. 63 These
points of view, though apparently contradictory, have both been
accommodated in the present Constitution which recognizes
that tribes are different from castes, but treats tribals, with
individual exceptions, as Hindus all the same.
The Government of India Act of 1935 had introduced special
provisions for the tribal people and a list of Backward Tribes
was promulgated in that connection in 1936.64 After the new
Constitution was adopted in 1950 the President promulgated
in the same year a list of Scheduled Tribes which was based
very substantially on the list of Backward Tribes promulgated in
1936 by the colonial government. The list was revised in 1976
and is due for another revision soon. A list of Scheduled Tribes
is required in connection with a set of special provisions in the
Constitution some ofwhich, though not all, are for only a limited
duration of time.
The Constitutional provisions have in certain respects sealed
the boundaries between tribe and non-tribe, and given to the
tribal identity a kind of definiteness it lacked in the past. Until
recently a tribe was part of a regional system, and tribes from
78 Society and Politics in India
different regions had little to do with each other. Ao Naga,
Munda, Dubla, Baiga and Toda lived their separate lives without
a sense of their common identity. The new legal and political
order has changed this to some extent. There is now not only
a definite tribal identity enjoying a legal sanction but a political
interest in maintaining and strengthening that identity.
Political forces released in the last few decades have not only
arrested the absorption of tribes into the wider society but have
to some extent reversed the process. The case of the Mahato of
Chotanagpur illustrates the point. They had been counted as a
tribe until 1921, but had themselves declassified in the census
of 1931 when they made a bid to identify themselves with the
large Kurmi-Mahato caste of Bihar. They are now trying once
again to have themselves reclassified as a tribe. 65 Paradoxically,
the number of communities deemed to be tribes has increased
with the modernization oflndia between 1950 and 1976, and the
tribal population as a proportion of the total population has risen
steadily from 5.30 per cent in 1951 to 7.76 per cent in 1981.
It is doubtful that this assertion of tribal identity in the
political domain can be described as retribalization in any
meaningful sense. The categories involved are very different
in scale and orientation from those of the past. It is true that
collective identities have proved far more durable in the face of
economic and political change than was earlier envisaged.66 But
the collectivities to which the label 'tribe' will remain attached
will depart in both form and function further and further from
what can reasonably be described as tribes.
 4 Caste and politics in Tamilnadu
Before discussing the role ofcaste in Tamilnadu politics, I would
like to examine in somewhat general terms the relationship
between caste and the political system. Most recent studies by
social anthropologists have emphasized the role of caste in local
politics and political scientists, too, have become increasingly
sensitive to the part it plays in wider political systems. Some
kind of interrelationship between these two sets of factors is
also often assumed by journalists both in this country and
abroad.
It is a truism that the nature and content of politics undergo
transformation from one territorial level to another. What is
rather less obvious to the non-specialist is that the caste
system itself evinces several levels of differentiation. These
levels require to be specified before a proper understanding can
be achieved of the transformations in the relationship between
caste and politics from one level of organization to another.
Enough material now exists for the formulation of a common
strategy for the analytical study ofthe relationship between caste
and politics in the different parts of the country.
While public discussion on the 'communal' pattern ofpolitics
does draw attention to a significant social reality, it also tends to
obscure the true nature of the relations between caste and the
political system. Obviously these relations are more immediate
at certain levels than at others. A failure to recognize such
differences is likely to lead to hasty and unsound generalizations.
It is also likely to divert attention from the fact that the relations
between caste and politics are not static but change continuously
over time. Finally, there are many alignments other than those
based on caste which play an important part in the political
process. Their role has often been undervalued because many
ofthem are more fluid and amorphous in nature and hence more
difficult to identify than is caste. The preoccupation with caste
in the present paper is not intended to convey the impression
80 Society and Politics in India

that other factors are unimportant or play a less important part

in political life, whether in Tamilnadu or elsewhere.
Studies on factions, partic_ularly by Nicholas 1 and Brass2, have
drawn attention to the need for a more differentiated and flexible
approach to the study of the relations between caste and politics.
These studies show how alliances are often made across caste
for the pursuit of political objectives. Some of the implications
of alliances of this kind will be examined at a later stage.

I shall begin with a consideration of the political system and the

way in which it can be best viewed for analysing the part played
in it by caste and similar modes of grouping. As I see it, there
are two major problems in the study of political systems. These
are, (1) problems of distribution, and (2) problems of process.
I shall briefly consider each in tum.
Problems of distribution Every society or social segment is
characterized by a certain distribution of power. Among these
a central position is occupied, by groups which have a territorial
basis, such as the village, the district or the State. Each such
unit is characterized by the presence of individuals or groups
of individuals who can take decisions which are binding on the
unit as a whole. The first task of political analysis is to identify
these positions of power and to examine their hierarchy and
interrelations.
Structures of power are of diverse kinds. They may be
regional or local; functionally diffuse or specific; formal or
informal. One important feature of contemporary India lies
in qualitative and quantitative changes in structures of power.
New types of structures of power such as parties, panchayats
and machines have proliferated since independence and pen-
etrated into the rural areas. The two main factors behind
this are the adoption of adult franchise and the institution of
Panchayati Raj.
Not only have structures of power multiplied but they have
also become more differentiated. In the past at the local level
the dominant caste was often the principal locus ofpower. Today
there are differentiated political structures of various kinds
such as parties, panchayats and machines. Such differentiated
structures ofpower are generally more easy to identify since they
 Caste and politics in Tamilnadu 81
have often a formal organization. But it has to be remembered
that particularly at the local level real power may be vested in an
informal body such as a group of lineage elders rather than in a
formal structure such as the statutory panchayat.
Each structure of power can be examined for its caste
composition and comparisons can be made between structures
at different levels of organization. This will tell us about the
distributive aspect ofthe relationship between caste and politics.
General statements are often made about the control by the.
dominant caste or a group or coalition of dominant castes of
the political system of an area. Such statements can be tested
empirically although this is rarely done. It should be possible
to study the frequency distribution of castes in given structures
of power and to compare these with their distribution in the
population as a whole. Such comparisons can be made fairly
accurately at the village level and in somewhat more broad and
general terms at other levels.
It is often found that a particular caste is highly represented in
the village panchayat or the panchayat samiti or the zilla parishad.
This high representation may at times be due to the fact that
the caste in question is numerically preponderant and highly
represented in the population as a whole. Also it seems likely
that over-representation of castes in structures of power is more
characteristic of certain territorial levels than of others.
A second kind of distributive problem relates to changes in
the caste composition of particular structures of power over a
period of time. A study of this kind will not be handicapped
by the absence of reliable material on the distribution of castes
in the total population, because here one has only to compare
the caste composition of relatively small units at different points
of time.
Two kinds of change seem to be taking place in the relation
between caste and politics in Tamilnadu as well as in other parts
of the country. In the first kind, power shifts from one dominant
caste to another. This happened when Kallas and a few other
castes wrested control over village politics from Brahmins in
Sripuram.3 It happened on a wider scale in Tamilnadu as a
whole when Brahmins were displaced by non-Brahmins in
important political bodies.
82 Society and Politics in India

The second kind of change is perhaps more radical than

the first. Here the locus of power shifts from the caste system
itself to differentiated structures of power. As I have indicated
earlier, a vast body of new structures of power has emerged
in India since independence. Today traditional bodies such
as groups of caste elders (which are functionally diffuse) have
to compete increasingly with functionally specific structures of
power such as parties and statutory panchayats. Often there are
mechanisms which bring about the interpenetration of the two
sets of bodies.
Problems ofprocess Structures of power exist within a con-
text of events and activities. This flow of events and ·activities
creates continuous changes in personnel and, over longer
periods of time, changes in the structures themselves. The
electoral process provides a good example of the way in which
changes are brought about in the personnel of political bodies.
Changes in personnel may be of minor or major significance; it
is difficult to say at which point a change in personnel becomes
so significant as to constitute a change in structure.
It is necessary for incumbents of political office to maintain
support in order both to acquire such office and to act effectively
within it. In no society is support given automatically; it has to be
sought and cultivated more or less actively. In societies having
representative government, such as contemporary India, there
are specific institutional arrangements through which support
is given and withdrawn. But the manipulation of support is
an important feature of all societies including those areas of
Indian society which are still largely governed by 'traditional'
principles.
Support may be given in return for material benefits. But
material benefits cannot be granted directly or immediately
in exchange for, every kind of support. For this reason the
mobilization of support requires appeals to loyalties of various
kinds which do not always have a tangible material basis.
In Indian society, as in other 'traditional' societies, an impor-
tant place is held by groups based on what have been called
'primordial loyalties'.4 Castes constitute typical instances of
groups of this kind. Loyalties to caste and kin, whose psy-
chological roots we need not examine here, provide powerful
 Caste and politics in Tamilnadu 83
bases for political support. Other things being equal, people are
expected to support members of their own caste or kin group.
To the extent that traditional values persist, loyalty and obliga-
tion to caste and community are considered 'good'. Such loyalty
and obligation are expected as a matter of course in a variety
of social fields; it is natural that they should be carried over to
the field of institutional politics. A slogan used by candidates of
the Meena caste during the elections in Rajasthan illustrates the
manner in which loyalties from one field are carried over into
another: 'Do not give your daughter or your vote to anyone but a
Meena'. Similar slogans have been widely used during elections
in Tamilnadu: vanniyar Vottu anniyirkku illai (the Vanniya vote
is not for anyone else).
Even in traditional Indian society there were other loyalties
besides those based on caste. Such for instance are the loyalties
ofclass and of patron-client relationships. The latter in particu-
lar often cut across caste and are in conflict with it. Thus, the
Palla client of a Vellala patron has one set ofloyalties to his fellow
Pallas and another to his patron who is of a different caste. Such
divided loyalties, which are an inherent feature of the system,
give to it a certain indeterminacy and always leave room for some
manipulation of support. This is one reason why political events
are difficult to predict though it is possible to interpret them.
Caste may enter into the political process in a number of
ways. Firstly, appeals may be made to caste loyalties in a
general way as when Vanniyas are exhorted to vote for Vanniya
candidates. The force of this kind of appeal is made evident in '
Tamilnadu where rival parties often match caste with caste in the
selection of candidates for electoral office. Secondly, networks
of interpersonal relations are activated both during elections and
at other times for mobilizing support along caste lines. Since
kinship, marriage and commensality often stop short at the
boundaries of caste, intra-caste relations are very important.
Finally caste associations such as the Vanniyakkula Kshatriya
Sangam may seek to articulate caste interests in an organized
manner.

I now tum to a brief consideration of the nature of caste.

The caste system in Tamilnadu is both elaborate and deeply
84 Society and Politics in India
segmented. Segments of different orders assume importance at
different levels of the political system. It seems that the political
process itself plays some part in bringing about changes in the
nature of segmentation in the caste system. Organized politics
often necessitates the fusion of adjacent segments and this
fusion in the political context is likely in, the long run to affect
other aspects of inter-caste relations, such as commensality or
intermarriage. s
The caste system evinces some of the structural properties of
segmentary systems. 6 There are several levels of differentiation;
large units are divided into smaller ones and these further
subdivided on the basis ofcleavages which are fairly enduring in
character. The divisions and subdivisions either merge with one
another or are placed in opposition, depending upon context and
situation. Although in a given context a unit of a lower order may
lose its identity through merger with an adjacent unit, it tends to
reappear as an independent entity in a different context. Thus
the system as a whole retains a degree of continuity over time.
The segments themselves are differentiated according to
styles of life. Each segment - whether sub-caste, caste or
caste-group - is characterized on the one hand by certain
diacritical distinctions and on the other by a set of syncretic
values. The diacritical distinctions 'define the unity of the
segment in terms of differentiation from other segments',
whereas syncretic values 'define the unity of the segment in
terms of internal solidarity'. 7 One caste differs from another in
matters of dress, diet and other habits, while within the caste
there is a consciousness ofcommunity. It is this which facilitates
the mobilization of support on the basis of caste as contrasted
with other social categories such as class.
Diacritical differences between segments have been greatly
elaborated in the caste system in Tamilnadu. In the past they
were reinforced by a variety of sanctions, some of which are
still in existence. Food habits, types of habitation, styles of dress
and many other customs varied from one caste to another. The
higher castes jealously preserved their traditional styles of life,
even to the extent of cooking or serving food in a particular
manner. The lower castes could imitate these ways to some
extent, but ritual and other sanctions prevented such imitation
 Caste and politics in Tamilnadu 85
proceeding beyond a certain point. In Tamilnadu such sanctions
were employed with considerable force even thirty years ago. 8
The unity which a caste derives from its diacritical distinctions
is more in evidence at higher than at lower levels of segmenta-
tion. Members of a broad division such as the Brahmins share
only a few diacritical elements in common, whereas those of a
subdivision of the Smartha Brahmins, for instance, share many.
Similarly, internal solidarity is likely to be more intense within
a sub-caste than within a group of related castes. The highest
order of segmentation in the caste system is represented by a
small endogamous unit whose members are the bearers of a
homogeneous cultural tradition and are, in fact, related to each
other by ties of kinship and affinity. At the other extreme are the
primary segments (viz. Brahmins, non-Brahmins and Harijans)
whose members share a few common customs and are bound
together by a broad feeling of community.
We can present here only a very briefand summary account of
the caste structure ofTamilnadu. The population ofTamilnadu
can be broadly divided into three groups, the Brahmins, the
non-Brahmins and the Harijans (or Adi-Dravidas). In the
villages the three groups are generally segregated in different
residential areas. The Brahmins live in brick-and-tile houses in
a separate part of the village known as the agraharam and are
marked off from the others by distinctive patterns of speech,
dress and diet; within the agraharam there is a fairly intense
community life from many areas of which non-Brahmins and
Harijans are excluded. The Harijans in their turn live in
their own streets known as cheris which have a unity no
less distinctive than that of the agraharam. The non-Brahmins
represent a broader spectrum of cultural variations and appear
to be on the whole less cohesive than the two other primary
segments.
Each primary segment, which appears as a unit in relation to
the others, is internally subdivided. The Brahmins in Tamilnadu
are subdivided into Smartha, Shri Vaishnava, etc. Each of these
evinces a greater measure ofunity (both diacritical and syncretic)
than the Brahmins taken as a whole. The Smartha Brahmins in
their turn are similarly subdivided into Vadama, Brihacharanam,
Astasahashram and Vattiman. The Vadama are subdivided into
86 Sodety and Politics in India

Vadadesha Vadama and Chozhadesha Vadama. Segmentation

among the non-Brahmins follows a more complex and less
clear-cut pattern and the Harijans as a unit appear to be
less deeply segmented than the Brahmins. But everywhere the
pattern is broadly similar. 9
At which level of segmentation does caste enter into the
political process? When we seek to analyse the role of caste
in politics, which should be our unit of investigation, a broad
grouping such as the Brahmins or a small subdivision such as
the Vadadesha Vadama or the Pramalai Kalla?
Some have been inclined to argue that castes can successfully
enter politics only when they combine into fairly large aggre-
gates; too much segmentation, in their view, tends to reduce
the viability of castes in the competition for power. There
can be little doubt that organized politics at the State level
has tended to bring about a certain aggregation of adjacent
segments within the caste system. This phenomenon appears
to have an all-India character and has been commented on by
many. Srinivas for instance writes, 'In order to be able to take
advantage of these opportunities, caste groups, as traditionally
understood, entered into alliances with each other to form bigger
entities'. 10 He gives the example ofthe Okkaligas in Mysore who
now include, in the context of State politics, several related but
distinct entities. The same can be said of the Vellalas or the
Kallas in Tamilnadu.
The Non-Brahmin Movement provided a common platform
not only for a wide variety of Hindu castes but also for Christians
and Muslims. It also added a new dimension to the unity of
the Brahmins which grew in response to the non-Brahmin
challenge. However, it would be a mistake to view this unity
as something absolute or as having significance in every context.
It is well known how the unity of the non-Brahmins was
loosened after they gained ascendancy in Tamilnadu because
of competition between rival dominant castes among them for
power and office. It is hardly surprising that cleavages which are
ignored in the interest of a particular objective should reassert
themselves once that objective has been achieved.
The unity one encounters in the caste system is in a very real
sense relative. Although a minor segment of a caste may be too
 Caste and politics in Tamilnadu 87
small to act as an independent unit in state politics, it may be
a viable unit in the village. Unity in one political context does
not preclude opposition·in another. The fact that a group of
sub-castes unites against a like group over a certain issue does
not mean that they cannot be divided over a different issue.
Conversely, the fact that two sub-castes contend for power in
a particular arena does not mean that they cannot unite against
a different caste in a wider arena. In fact, the complementary
processes of fission and fusion are an important feature of caste
politics in Tamilnadu. Further, since caste is a highly structured
system, fission and fusion are inclined to follow clearly defined
patterns.
Many have observed that the unity provided by caste has
different degrees of inclusiveness. 11 Srinivas tries to account
for this in a way which appears to be characteristic.
The point which needs to be emphasized here is that for purposes
of sociological analysis a distinction has to be made between caste at
the political level and caste at the social and ritual level. The latter
is a much smaller unit than the former. 12
To my mind, Srinivas draws the distinction at the wrong place.
What is important here is to distinguish not between political
and social levels, but between different levels oforganization in a
'merging series', 13 viz. State, district and village, or caste-group,
caste and sub-caste.
Srinivas' conclusion that 'caste at the political level' is gener-
ally a large aggregate derives from his preoccupation with the
role of caste in state politics. But caste may also play a part
in village politics and there the effective unit of organization
may be fairly small. Nor is it correct to maintain that larger
aggregates have no social or ritual unity. Such broad groupings
as the Brahmins and the Harijans (and to a much lesser
extent the non-Brahmins) do have a measure of diacritical and
syncretic unity and it is this fact which largely accounts for their
persistence at every level of political organization.
Srinivas also appears to suggest that the larger aggregates not
only are specifically political in function but are somehow new
to the Indian scene. Even this position cannot be well sustained.
Broader groupings such as Brahmins, Shudras and Panchamas
88 Society and Politics in India

were relevant to a wide variety of contexts even in traditional

society. 14
We conclude that caste may be significant to the political pro-
cess at every level of segmentation, although organized politics
at State and district levels has often led to a quasi-permanent
aggregation of segments. Further, there seems to exist some
broad relationship between the arena of politics and the level
of segmentation at which caste enters into it. At the State (or
even district) level, minor segments merge with one another so
as to be able to operate as viable units. But this kind of merger
easily comes about precisely because a basis for it already existed
in the traditional structure. At the village level a major segment
may subdivide and its component units be opposed to one
another, again because the cleavage was present even in the
past. Thus, while fusion may occasionally be 'structural', very
often it is merely 'situational'.
The principle of segmentation operates even within the sub-
caste, viewed as the smallest unit of endogamy. In such cases
the units which stand in opposition to each other are generally
lineages. (A sub-caste which is an endogamous unit is often
subdivided into lineages which are exogamous. In certain parts of
Tanjore district where the Kallas are decisively dominant, rival
candidates for political office often belong to different lineages
ofthe same sub-caste. (In other contexts the three related castes,
Kalla, Marava and Ahamudiya, may together operate as a single
unit). When a sub-caste is large, culturally homogeneous and
decisively dominant, the cleavages within the lineage system
often assume great importance. (A very good example of this
is provided by the Jats of Northern India.) Thus for certain
purposes lineage, 15 sub-caste, caste and caste-group may be
viewed as constituting a single series.1 6
Although both the territorial system and the caste system show
a similar pattern of division and subdivision, it would be wrong
to assume a high degree of correspondence between levels of
segmentation in the two. Even a broad grouping within the caste
system such as the Brahmins may be relevant to village as well as
state politics. Much depends upon the caste composition of the
village or other territorial unit in question. A village which has
only a few castes of which one is decisively dominant will show
 Caste and politics in Tamilnadu 89
a different kind of alignment from one where there are many
castes among which none enjoys decisive dominance. Every-
where, however, the caste system provides one set of cleavages
along which units tend to merge or subdivide. Whether they do
merge or subdivide will depend upon a variety of other factors,
some ofwhich are extraneous to the structure of caste. It cannot
be too strongly emphasized that political alliances, at every level,
often cut across caste and are frequently based upon affiliations
which have little direct connection with caste.

The threefold division of society into Brahmin, non-Brahmin

and Harijan provides perhaps the broadest basis for 'commu-
nal' politics in Tamilnadu. (In the present analysis I ignore
altogether the Hindu-Muslim cleavage; on the whole it has
been of less importance than the cleavage between Brahmins
and non-Brahmins or between non-Brahmins and Harijans.)
This division provides the basic framework for the analysis of
problems of both distribution and process. It casts its shadow at
every level of political organization from the village to the State.
Its importance is given tacit recognition in the composition of a
variety of political bodies. The Madras Municipal Corporation,
for instance, has long followed the convention of rotating
the Mayoralty between one Brahmin, two non-Brahmins, one
Harijan and one Muslim.
To what extent are we justified in treating such broad divi-
sions as castes? I have shown that a certain measure of unity is
associated with each category, and Brahmins and Harijans at
least are certainly viewed as castes in a variety of contexts by
the Tamil-speaking people. The non-Brahmins admittedly are
a very broad and heterogeneous division and sometimes (though
not generally) they are so broadly defined as to include even the
Harijans. Even so, the Non-Brahmin Movement gave them a
certain coherence and unity, which seem to have outlived the
Movement itself. For that reason it becomes necessary to analyse
their role in Tamilnadu politics even though they constitute a
kind of residual category. (The term 'non-Brahmin' has been
criticized by many precisely on this account; however, it has
become a part of the political history of T amilnadu and it would
be unreal to dispense with it, at least in this context.)
90 Society and Politics in India

Perhaps the most important consequence ofthe Non-Brahmin

Movement (of which more later) was the introduction of a
'communal' or caste idiom into South Indian politics. The
composition of political bodies was changed by it, sometimes
artificially, through reserved seats, and everywhere communal
loyalties became important in giving or withdrawing support. It
is doubtful whether the Movement even attempted to organize
politically the entire body of non-Brahmins. But it certainly did
succeed in creating a lasting impression that in virtually every
political context it was important whether a person was a Brahmin
or a Non-Brahmin. It is this which gives the politics of South
India - including Maharashtra but probably excluding Kerala -
its distinctive character when compared with the politics of the
North. The special position ofthe Harijans in the political system
is acknowledged everywhere in India and is, in fact, sanctioned
by the Constitution.
The Brahmins constitute a convenient starting point for a
discussion of the role ofcaste in Tamil politics for their position
is in many ways unique. As a social stratum they were the first
to be politicized and up to the l 920s they enjoyed a dominant
position in the former Madras Presidency. Their representation
in most of the political bodies which then existed was far in
excess of their proportion in the population as a whole. The
changes in their political fortunes over the last fifty years bring
into focus not only the role of caste in politics but certain major
shifts in the bases of power in Tamil society. Certainly, no
section of Tamil society of comparable size 17 has for so long
occupied the storm centre of political debate and controversy
and it is doubtful whether any other section has undergone a
more radical change in its relation to the distribution of power.
To what do the Brahmins owe their exceptional position in
Tamil society? I have already commented on the diacritical
differences between the Brahmins and the others. These are
certainly sharper in Tamilnadu, and in South India as a whole,
than in North India. Two of them may be considered to begin
with: the real difference in linguistic usage and the imputed
difference in racial origins. These distinctions, in part real and
in part imaginary, have combined to create a popular and wide-
spread belief that the Brahmins represent an 'Aryan' element
 Caste and politics in Tamilnadu 91
superimposed on an indigenous 'Dravidian' sub-stratum. This
belief has far-reaching consequences for the development of
political attitudes in Tamilnadu.
The opposition to Brahm.ins has been posed in economic as
well as ethnic terms. On the whole, Tamil Brahm.ins enjoyed a
favourable position in the traditional economic system. A con-
siderable section of them owned land, though they were not all
landowners and there was a large class oflandowners - both big
and small- among the non-Brahmins. Even in Tanjore district,
which has the highest concentration of Brahmin landowners,
the three biggest landowners prior to the fixation of ceilings
were non-Brahmins 18 but the proportion oflandowners among
the Brahmins was much higher in Tanjore district than among
any non-Brahmins caste or caste-group of comparable size (with
the possible exception of the Mudaliyars).
It seems probable that large or moderately large Brahmin
landowners were concentrated primarily in the Tanjore-Trichhy
area and that few were to be found elsewhere. But Brahmin
landowners, whether large or small and whether in Tanjore
district or outside, have been related to the productive organi-
zation in a significantly different way from the non-Brahm.ins
since they are debarred by scriptural injunction from the actual
work of tillage. Hence a Brahmin who owns even a small parcel
ofland has to depend for labour on non-Brahmins and Harijans,
whereas a non-Brahmin landowner in a comparable situation
may not only till his own land but will probably take some
additional land on lease and perhaps also work as a part-time
agricultural labourer. The majority of non-Brahmin landowners
are peasant farmers, tenants or agricultural labourers or some
combination of the three. Among Brahmin landowners, on the
other hand, a large proportion consists of rentiers and many
are absentee landlords. This contrast between Brahmin and
non-Brahmin landowners, arising from differences not so much
in size of holding as in styles oflife, has played a most important
part in the development of the Non-Brahmin Movement
The initial consequence of British rule was probably to
increase the structural distance between Brahm.ins and the
rest of Tamil society. Brahm.ins were the first to take to
Western education and Western educated Brahm.ins entered
92 Society and Politics in India

the professions and services in large numbers. Those who

entered Government and other services used the ties of kinship
and affinity to recruit more Brahmins. It is difficult to form an
accurate estimate oftheir representation in professional, admin-
istrative and managerial positions, but there is little doubt that
during the first quarter of the present century it was extremely
high. This was projected as a major issue by the Justice Party,
which emerged in 1917 as a champion ofnon-Brahmin interests
and demanded more equitable representation for them in the
educational system, in local bodies and in the services.
In this initial phase (which may arbitrarily be considered as
ending with the formation of the Justice Party in 1917) the
cleavage between Brahmins and non-Brahmins was widened
in two important ways. As Brahmins entered the institutions
of higher learning, the professions and the services, everywhere
they formed cliques from which non-Brahmins were excluded.
Between 1892 and 1904, out of sixteen successful candidates
for the I.C.S., fifteen were Brahmins; in 1913, ninety-three
out of 128 permanent district munsifi were Brahmins; in 1914,
452 out of the 650 registered graduates of the University were
Brahmins. 19 In a system which was ostensibly competitive but
in which the scales must have seemed heavily weighted against
non-Brahmins, the latter inevitably developed deep feelings of
resentment.
There was another important consequence ofWesternization.
As its pace mounted, Brahmins began increasingly to look
outwards to the towns and cities. They left the agraharams
in large numbers - at first temporarily but with an increasing
measure of permanence - and joined schools, colleges and
offices in the urban centres. They had at no time had the
same intimate relations with the land as the non-Brahmins and
Harijans, and Westernization loosened considerably such bonds
as they did have with tenants and labourers in their ancestral
villages. But although they began to tum outwards, they did
not dispose of their land to any great extent, at least not in
the initial period. Rather, they became rentiers and absentee
landowners, returning to the village from time to time and
even while there, keeping one eye on a job as a clerk or
school teacher in a neighbouring town. It is evident that even
 Caste and politics in Tamilnadu 93
within the village relations between Brahmin landowners and
non-Brahmin tenants were weakened as a general consequence
of the Brahmins' Westernization. This is certainly true of the
Tanjore-Trichy area and appears to be broadly true of other
districts as well, although the changes in some districts seem to
have been far less marked.
This then is the social background out of which the Non-
Brahmin Movement emerged. The Brahmins were politically
isolated first because they constituted a separate ethnic entity
and then because they occupied privileged positions in the
economy, both as landowners and as professionals and admin-
istrators. In addition, they formed a very small minority, only
about 3% of the total population of the old Madras Presidency. _
Their proportion in contemporary Tamilnadu cannot be very
much higher.
Once non-Brahmin opposition was organized, it did not take
long to dislodge the Brahmins from their privileged positions. In
1921 Madras Presidency came under the control of the Justice
Party. Non-Brahmin representation in political bodies increased
and they were favoured by a series of legislative and executive
actions. The Brahmins suffered on two accounts: they were in a
minority and they were largely behind the Congress Party which
either boycotted the elections or refused to form ministries.
However, their decline during this period was only relative, in
terms of their dominance prior to the formation of the Justice
Party in 1917. In 1937 when the Congress staged a come-back,
C. Rajagopalachari, the veteran Brahmin leader, became Chief
Minister of Madras and other Brahmins were appointed to the
Ministry.
For all this, the Brahmins were never quite able to regain the
dominant position which they had once occupied in the Presi-
dency. After a brief revival between 1937 and 1939, their posi-
tion again declined, this time perhaps irrevocably. The Congress
itself came gradually to be dominated by non-Brahmins. The
1940s saw the emergence of the militant Dravida Kazhagam
which preached and at times practised violence against the
Brahmins. With the extension of the franchise after independ-
ence and the introduction of Panchayati Raj a keen awareness
developed among the Brahmins of their weakness and political
94 Society and Politics in India
isolation. In 1916 the Non-Brahmin Manifesto had made angry
protests against Brahmin domination. In 1962·Brahmin voters
in Trichy had to be escorted to the polling booths for fear of
violence from the Dravida Kazhagam. Brahmin representation
in the Ministry, the Legislatures and the Congress Party had
dwindled into insignificance.
There is little precise information regarding the representa-
tion of Brahmins in political bodies at different levels. That
there has been a general decline is beyond question. But it
is possible that even today Brahmins are over- rather than
under-represented in at least certain types of political organs.
Political developments over the last fifty years have created
among Tamil Brahmins a strong sense of identity as a minority.
I have heard Brahmins with a flair for metaphor describe
themselves as the Jews of South India. A strong feeling has
taken root among them that they were made victims of every
kind of discrimination. But they have not sought escape from
organized politics. On the contrary, because of their feeling of
political isolation and also because of the high rates of literacy
and education among them, they are perhaps the most highly
politicized section of Tamil society.
There is ample indication that the Brahmins are rapidly
growing alive to the fact that if they are to survive politically
they must come to terms with the non-Brahmins. In both
1962 and 1967 they supported the DMK in spite of its
non-Brahmin background. Many among them see the need
to forge new alliances, transcending caste identities. In fact,
the most bitter critics of 'communal' or caste politics today are
the Brahmins. The Non-Brahmin Movement in its tum seems
to have spent itself, having achieved its principal objectives. The
DMK renounced its anti-Brahmin bias several years ago and
canvassed actively for Brahmin support in the last two General
Elections. It seems very likely that in the near future conflicts
between Brahmins and non-Brahmins will play a less important
part than they have done in the past.
Before turning to an examination of the role of the non-
Brahmins in state politics, it may be useful to take a close-up
view of the Brahmins, first at the district level and then at the
village level. I shall consider here only one district, Tanjore, and
 Caste and politics in Tamilnadu 95
this for two reasons. Firstly, Tanjore is the cultural homeland
of Tamil Brahmins and contains the highest proportion of
Brahmins in Tamilnadu. Secondly, it has been studied more
intensively by social anthropologists than any other district in
the State. In what follows I shall draw on the field materials
of Gough,20 Sivertsen21 and my own research in and around
village Sripuram. 22 Since these studies were all made in the
same culture area but at different points oftime (Gough's during
the first and mine during the third General Elections), they offer
useful material for an assessment of change.
Tanjore district has been the classic stronghold of Brahmin
mirasdars. Gough estimates that the Brahmins 'number about
200,000 in this district' and 'own land and have administrative
rights in about 900 out of a total of 2,611 villages'.23 The
Tanjore Brahmins are also highly educated and show a high
degree of political consciousness. For all this, they are not very
highly represented either in the organs of local government or ·
in the local organization of the ruling party.24
When Gough made her study in the early fifties the Tanjore
Brahmins were under attack from two quarters, the Communist
Party and the Dravida Kazhagam, both of which were then
quite strong. Further, the two parties were united in their
opposition against them, the first on the ground that they
were mirasdars and the second because they were Brahmins.
At about this time peasant riots were organized in Mannargudi
talulea, a stronghold ofBrahmin mirasdars. The alliance with the
staunchly anti-Brahmin Dravida Kazhagam seems to have made
the Communists a little wary of antagonizing.the non-Brahmin
landowners. And there can be little doubt that in Tanjore district
the Communists drew much of their strength from the Dravida
Kazhagam. In 1952, when the DK supported the Communists,
the latter held six Assembly seats out of a total of nineteen from
the district. 25 When in 1962 the DK supported the Congress,
the Communists were not able to win a single seat.
Until the mid-l95Os the Tanjore Brahmins appear to have
been solidly behind the Congress. This support was based as
much on their traditional association with the Congress as on
their opposition to the Communists and the DK, then the two
principal antagonists of the Congress. Things began to change
96 Society and Politics in India

rapidly after the mid-1950s. The veteran Brahmin leader, C.

Rajagopalachari, was replaced by the non-Brahmin Kamaraj as
Chief Minister and the Dravida Kazhagam, known and feared
for its militant anti-Brahminism, switched its support to the
Congress. The Congress enacted a series of laws curtailing
the rights of landowners and many of the Brahmin mirasdars
in Tanjore viewed these as being specifically directed against
themselves. Finally in 1959 the Swatantra Party was formed
under the leadership of Rajagopalachari and many of the
Brahmins of Tanjore turned avidly towards the new party. In
each of the half-a-dozen Brahmin villages I visited in 1961-2,
the Swatantra Party had a solid core of supporters in the
agraharam. In Tanjore district the Swatantra Party soon came
to be known as the Brahmin party, although many ofits members
were, in fact, non-Brahmins.
The 1962 elections found the Brahmins largely opposed to
the Congress and, if anything, the mood was intensified in
1967. Since the late 1950s their political attitudes appear
to have been defined primarily in terms of opposition to the
party then in power. In 1962 they supported DMK candidates
where the Swatantra Party did not put up any and the electoral
alliance between the two parties in 1967 appears to have made
their choice even simpler. In 1962 the Brahmins generally
supported the Congress in only those rare constituencies where
the Congress itself put up Brahmin candidates.
As indicated earlier, the Brahmins in Tanjore have developed
a strong sense of unity in response to their political decline.
Formerly there was active rivalry between the Smartha Brahmins
and the Shri Vaishnavas and between the two sections ofthe Shri
Vaishnava Brahmins. These are now largely forgotten. There is
a conscious effort on their part today to foster a sense ofoneness.
The Brahmins see themselves as belonging to a minority and
define their identity in opposition to the non-Brahmins and
not to a different section of the Brahmins. In this regard
their position seems to be somewhat different from that of the
non-Brahmins.
Although the Brahmins constitute a very small minority in the
district as a whole; their position is somewhat different in each
of the three villages studied in detail. This is because these are
 Caste and politics in Tamilnadu 97
all agraharam villages, i.e. villages with large concentrations of
Brahmins, unlike the majority of Tanjore villages where there
are no agraharams and at best only a few families of priestly
Brahmins. Both Gough and Sivertsen report a decline in the
traditional authority of the Brahmins in the villages studied
by them. In both cases political parties and associations have
played a leading part in organizing support against the Brahmin
mirasdars. In Thyagasamudram the Brahmins organized them-
selves into a Landlords' Association while the non-Brahmins
and the Harijans rallied round the Cultivators' Union. Some
of the bigger non-Brahmin landlords at first supported the
Brahmins, but they seem to have been pressured at a later stage
into joining the Cultivators' Union. 26
The case of Sripuram merits discussion at greater length
because changes there seem to have been more decisive in
character. At the beginning ofthe present century Sripuram was
a flourishing agraharam village, well known throughout Tanjore
district for its large and prosperous community of Brahmins.
The agraharam at Sripuram is rather unusual in the sense that it
contains Brahmins belonging to a number of different castes and
sub-castes. Fifty years ago the Brahmins of Sripuram enjoyed
decisive dominance. However, the internal cleavages between
the Smarthas and the Shri Vaishnavas, and among the latter
between the Thengalai and Vadagalai sections, were reflected
in the competition for power relating to the control of the
village temple and other local institutions. Today the power
of the Brahmins has declined considerably, the old disputes
between the Smarthas and the Shri Vaishnavas have been
largely (though not entirely) forgotten and the Brahmins try
to face the challenge of the emerging non-Brahmin leadership
with a measure of unity.
Sripuram is, in Dahl's terminology, being transformed from
a 'system of cumulative inequalities' to one of 'dispersed
inequalities'. In the past the Brahmins enjoyed the highest
positions in the hierarchies of status, class and power. Today
they continue to enjoy ritual and economic dominance but
political power has shifted to the non-Brahmins. The shift
in political power has been hastened by the introduction of
Panchayati Raj.
 98 Society and Politics in India
Until the mid-1940s the Brahmins dominated the village
panchayat. The panchayat head was always a Brahmin and the
panchayat room was situated in the agraharam. Non-Brahmin
members ofthe panchayat had more or less the status ofsecond-
class citizens. Everything changed after independence. Now the
panchayat is completely dominated by the non-Brahmins: there
are six non-Brahmin members as against three Brahmins. Both
the President and the Vice-President are non-Brahmins and
have been non-Brahmins ever since independence. Symbolic of
the transfer of power from the Brahmins to the non-Brahmins
has been the shift in the location of the panchayat hall from
the agraharam to the non-Brahmin streets. In fact, this shift
is of more than symbolic significance. In an agraharam village
Brahmins and non-Brahmins live more or less segregated in
their different residential areas and Brahmins do not normally go
to the non-Brahmin streets unless specifically invited. Now that
the panchayat hall is the venue of important political gatherings
in the village, many of the Brahmin residents find themselves
• automatically excluded from such gatherings. For instance,
when the Minister for Co-operation came to the village in
1962, he was entertained in the panchayat hall; few of the
Brahmins attended the gathering and many of them came to
know of it only after the event.
Changes in the relative positions of Brahmins and non-
Brahmins in the political system of Sripuram are of course
reflections of changes in the bases of power in the wider
system. In the traditional system power was derived largely
from landownership and high ritual status. The institution of
adult franchise, the introduction of Panchayati Raj and the
development of specialized political organs have created new
bases ofpower. In Sripuram the Brahmin landowners have been
edged out of the panchayat by popular non-Brahmin leaders who
command the support of numerically preponderant groups and
have access to leaders and party bosses outside the village.
In the 1957-62 period the Brahmins' isolation from structures
of power in Sripuram (and to some extent in Tanjore district
as a whole) was partly a consequence of their alienation from
the ruling Congress Party. In 1962 the Brahmins in Sripuram
supported the DMK candidate for the Assembly seat. The
 Caste and politics in Tamilnadu 99
influential non-Brahmins, including the panchayat President,
had then supported the Congress. This time the DMK has been
returned to power, partly on Brahmin support. This may have
some effect on the political situation of the Brahmins although
it is difficult to say how permanent or far reaching this effect will
be. In any event it seems more than likely that the levers ofpower
will for some time to come be wielded largely by non-Brahmins
whether at village, district or State level.

The non-Brahmin ascendancy in Sripuram can be made fully

intelligible only in relation to wider structures of power and
changes in the distribution of power over the last half-century.
The Non-Brahmin Movement was formally launched with the
issue of the Non-Brahmin Manifesto in December 1916. The
Manifesto presented detailed figures showing an extremely
high concentration of Brahmins in the public services, in
public bodies and in the educational system and contrasting
this with the. 3% they formed of the total population of the
Presidency. 28
The Movement gathered strength within a very short time.
The Manifesto 'was followed by the starting of three daily
papers, in English, Telugu and Tamil, the English paper
being called the Justice which became the mouthpiece of the
movement, and supplied the name also to the party itself'.29
The Party held its first Conference at Coimbatore in August
1917. Thereafter a series of Conferences was held in the Tamil
and Telugu districts of the Presidency. The Conferences were
evidently organized with expense and care and attended by a
large number of notabilities. 30
The Justice Party sent a powerful delegation to England in
1919 to present the non-Brahmin case before the Joint Parlia-
mentary Committee set up in connection with the Government
of India Bill. The Memorandum presented by K.V. Reddi
Naidu on behalf of the non-Brahmins was one of the longest
and was once again armed with facts and figures showing the
domination of the Brahmins in every field. It argued that the
interests of Brahmins and non-Brahmins were incompatible,
that they claimed different racial origins and that if power were
transferred without communal representation:
100 Society and Politics in India

that power will be utilised for the aggrandisement of the Brahmins

and to the detriment of the Non-Brahmins; that a Brahmin
oligarchy will be substituted for a British bureaucracy; . . . . and
that the Brahmin oligarchy will never be responsible to the masses
and the middle classes, the poor and the proletariat. 31
In retrospect, the Non-Brahmin Movement appears to have
achieved singular success within a remarkably short period of
time. Although the full demands for communal representation
were not met, twenty-eight out of the ninety-eight elected seats
in the newly constituted Madras Council were reserved for
non-Brahmins. 32 The Justice Party captured the polls in the
elections of 1920, no doubt partly because of the withdrawal
of the Congress. The three Indian Ministers placed in charge
of 'transferred subjects' were all non-Brahmins and Justicites.
The same pattern was repeated in the succeeding Legislature
constituted in 1923. In 1926 the Justice Party was defeated by
the Swarajists, but the latter refused to form a Ministry and
an Independent Ministry was formed under the non-Brahmin
leader, P. Subbaroyan. The Justice Party rode into power again
in 1930 and, though defeated in the elections of 1934, was
not finally dislodged until 1937 when the Congress formed a
Ministry under the leadership of C. Rajagopalachari.
After their success in the 1920 elections the leaders of the
Justice Party settled down to the task of improving the position
of Non-Brahmins through legislative and executive action. The
·debates in the newly constituted Council were replete with
questions and counter-questions regarding the representation of
Brahmins and non-Brahmins in the services and public bodies.
Forming a vast majority, the 0ustice) party concerned itself with
little more than communal questions, including the problem of
communal representation in various areas of government service,
and within one year the Council had become a forum of anti-
Brahmin propaganda.33
One of the first moves of the Council was to recommend the
appointment in every district of a Protector of Non-Brahmin
Subordinates in Public Services (GO No. 114, dated 3.3.1921).
Reservations for non-Brahmins were introduced in increasing
proportion in the services, in local bodies and in the institutions
 Caste and politics in Tamilnadu 101
of higher learning. With the Justice Party acting as the watchdog
of non-Brahmin interests, changes began to come about in the
distribution of power between castes.
Why did the non-Brahmins need a separate party and how far
did the Justice Party represent their interests? The need for a
separate party was made clear in the Non-Brahmin Manifesto.
The Brahmins had virtually monopolized some of the crucial
advantages offered by British rule and the threat of a 'Brahmin
oligarchy' must have appeared very real to many non-Brahmins.
The latter, in spite of their enormous strength of numbers and
their wealth (for most big zamindars were non-Brahmins), lacked
organization. The only existing party, the Congress, was virtually
controlled by Brahmins. As the Manifesto pointed out, 'of the
fifteen gentlemen elected from this Presidency to represent it on
the All-India Congress Committee, with the exception of one
solitary Non-Brahmin Indian, all are practically Brahmins'.34
It is small wonder then that the Justice Party met with such
immediate response when it was formed.
However, although the Justice Party did define its position·
in opposition to the Brahmins and although it did claim to
speak on behalf of the forty million non-Brahmins of Madras
Presidency, it would be a mistake to identify it with the interests
of the non-Brahmins as a whole. First of all, the forty million
non-Brahmins on whose behalf the Party claimed to speak
included Muslims as well·as Harijans, and people belonging to
three language groups, Tamil, Telugu and Malayalam. While it
is true that in its broadest definition the non-Brahmins included
all who were not Brahmins, in practice the Depressed Classes
and the Muslims were generally considered separately. Even
after the exclusion of Muslims and Harijans the non-Brahmins
would constitute a far more heterogeneous category than the
Brahmins. And those who participated in the Non-Brahmin
Movement in its first phase, supplying its leadership and
controlling its organization, were, in fact, drawn from a very
narrow social base. For the Justice Party was in a very real
sense an elite party dominated by urban, Western educated,
landowning and professional people. It contained a formidable
array of Rajas, zamindars, industrialists, lawyers and doctors. It
was by no means a mass party and it is doubtful whether any
I02 Society and Politics in India
serious effort·was made to draw peasants and workers into its
organization.
I cannot do better here than to quote K.B. Krishna, himself a
non-Brahmin:
This movement represents the emergence of the educated middle
classes who are not Brahmins.... The Non-Brahmin professional
classes are no more champions of social justice than the Brahmin
professional classes.... The Non-Brahmin movement of Madras
Presidency is no other than the movement of the later educated
middle classes who happen to be Non-Brahmins against the earlier
educated middle classes who happened to be Brahmins.35
A more recent student has made a similar point:
The leadership, financially well endowed, was drawn almost exclu-
sively from a socially stable element of the urban population. While
Chetty, Nair, Mudaliar and the early leaders of the movement
spoke of the illiterate Non-Brahmin masses of Madras, they in
no way represented them.... With the franchise limited to but a
few hundred thousand, the party made little attempt to aggregate
support at any wider level. Its demands were formulated, not so
much to attract a following, as to influence the official policy of the
British in Madras Presidency.36
Though markedly elitist in character, the leadership of the
Justice Party was heterogeneous in terms of both class and
caste. It would be a mistake to characterize it as a 'middle-class'
party for besides professional people there were in it landed and
capitalist elements. There was also a fairly wide range of castes
although most of the prominent people belonged to the upper
crust of non-Brahmin castes such as Mudaliyars, Chettiyars and
Vellalas among the Tamils, Rajus, Reddis and Naidus among
the Telugus and Nairs among the Malayalis. In the Tamil
districts some large non-Brahmin castes such as the Padayachis
and the Kallas and particularly the artisan and servicing castes
do not seem to have been very strongly represented.
A word requires to be said about the territorial framework
of the Justice Party and the Non-Brahmin Movement in South
India. Boundaries between states have changed considerably
over the last half-century and the Tamilnadu of today is very
different from the Madras Presidency of the 1920s and 1930s.
 Caste and politics in Tamilnadu 103
The older unit was not only larger but culturally much more
heterogeneous. It is difficult now to isolate the exact contribution
of the Tamil speaking people to the Non-Brahmin Movement.
But one must remember that a large number of the leaders of
this Movement were Telugu-speaking people who combined
with their Tamil- and Malayalam-speaking counterparts against
the Brahmin elite which appears to have been predominantly,
though by no means wholly, Tamilian.
Two broad conclusions emerge from a consideration of the
Non-Brahmin Movement. Firstly, it created alliances which cut
right across linguistic and cultural divisions: the significance of
this in a society in which 'linguism' and 'regionalism' played
such an important part only a short while later can hardly be
over-emphasized. The second point which emerges is that the
entire political arena in which Brahmins and non-Brahmins
stood poised against each other was a very restricted one: the
participants were drawn almost wholly from the urban, Western
educated, landowning, business and professional classes.
It seems that little change took place in the distribution
of power in the districts except in the towns. Non-Brahmin
dominance in the organs of state and municipal Government
began with the success of the Justice Party in the elections
of 1920. Yet in the villages things appear to have remained
very much as they had been in the past for many more years.
Villages in which Brahmins had been the dominant caste did
not witness a transformation comparable to that taking place in
the Provincial Legislature or the Municipal bodies. In Sripuram,
and presumably also in Kumbapettai and Thyagasamudram,
the Brahmins continued to enjoy decisive dominance and to
control the panchayat well into the 1940s. It was only after the
introduction of adult franchise and particularly ofPanchayati Raj
that the tables were turned on the Brahmins. By this time the
Justice Party had been almost forgotten and the Non-Brahmin
Movement had acquired an entirely different character.
Yet it would be a mistake to minimize the role of the Justice
Party and the Non-Brahmin Movement of the 1920s. The
Justice Party not only prepared the ground for the induction
of new social strata into the political system but also created a
distinctive idiom for South Indian politics. This idiom remained
104 Society and Politics in India
as a crucial element in the political process long after the Justice
Party itself disintegrated. It permeated every kind of political
organization including the Congress Party.
The Justice Party was routed in the elections of 1936 and
thereafter it was almost completely eclipsed. But its defeat
did not lead to a complete reversal of the Non-Brahmin
ascendancy. It is true that the helm of Madras politics was
taken over for some time by the two Brahmin Congressmen,
C. Rajagopalachari and S. Satyamurti. But the non-Brahmins
began increasingly to infiltrate the Congress and to acquire
key positions within it. Several prominent Justicites joined the
Congress after the defeat of their party. But the non-Brahmins
had to wait for independence and the first General Elections in
independent India for their control over the Congress to become
complete and decisive.
Efforts to rouse the non-Brahmin conscience were not con-
fined to the sphere of politics in the narrow sense of the term.
From the late 1920s onwards the Self-respect Movement began
to make concerted efforts to undermine the supremacy of the
Brahmins in the ritual and social spheres. The Movement was
welcomed by the Justice Party although its immediate objectives
were not political. It sought rather to create for Non-Brahmins
a climate of self-confidence and to liberate them from the ritual
tyranny of the Brahmins. In the late 1920s and early 1930s it
gained a measure of popular appeal under the leadership of
the non-Brahmin ex-Congressman, E.V. Ramaswami Naicker.
In 1944 Naicker created the Dravida Kazhagam and this asso-
ciation became the spearhead of militant anti-Brahminism in
the South. Idols were desecrated, sacred books were burnt and
some violence was practised against Brahmins. We have seen
how during the mid-fifties the DK switched its support to the
Congress which had by then come largely under non-Brahmin
control.
The non-Brahmin ascendancy reached its peak some time
in the mid-l 950s. In recent years, with the consolidation of
non-Brahmin power, internal cleavages have increasingly tended
to manifest themselves. These cleavages, which had existed as
an inherent feature of the caste structure, were to some extent
subsumed under a wider unity during the initial phase of the
 Caste and politics in Tamilnadu 105
Non-Brahmin Movement. With the introduction of Panchayati
Raj, rifts between subdivisions of the same caste are likely to
widen, particularly at the village and Block levels.
Today the category non-Brahmin has become too broad to
provide an adequate understanding of the role of caste in
politics. In order to achieve such an understanding it is necessary
to introduce the concept of dominant caste. This concept was
first used systematically by Srinivas. 37 He has defined it in the
following way:
A caste may be said to be 'dominant' when it preponderates
numerically over other castes, and when it also wields preponderant
economic and political power. Alarge and powerful caste group can
be more easily dominant if its position in the local caste hierarchy
is not too low.JS
Dominant castes have come to play an important part in every
sphere of politics and in Tamilnadu today every dominant caste
of any significance is a non-Brahmin caste.
It has been said earlier that the non-Brahmins in Tamilnadu
constitute a congeries of castes. They include landowning and
cultivating castes such as Vellala, Gaunda and Padayachi; trad-
ing castes such as Chetti; artisan castes such as Tachchan (Car-
penter), Kollan (Blacksmith) and Tattan (Goldsmith); servicing
castes such as Ambattan (Barber) and Vannan (Washenµan);
and a whole host of other specialist castes. Each of these castes
enjoys a greater measure of unity than the non-Brahmins as
a whole, although most of the major ones are themselves
subdivided.
The major Peasant castes are not evenly distributed through-
out the State but have areas ofconcentration within it. Although
there is no exact correspondence between these areas of concen-
tration and the division of the State into districts, certain castes
can be said to be dominant in certain districts. A caste which is
dominant in one district may not be dominant in another. There
are certain districts in which more than one caste is dominant
and certain castes which are dominant in more than one district.
Thus the Mudaliyars are dominant in Chingleput district, the
Padayachis in North and South Arcot districts, the Thevars
in Ramnad district, the Gaundas in Coimbatore district, the
I06 Sodety and Politics in India

Vellalas in Tinnevelli district and the Nadars in parts of Madura

district. .In Tanjore district dominance is shared between the
Kallas and the Vellalas.
Srinivas has enumerated a number of criteria on which
dominance is said to be based.39 In addition to the ones
enumerated, geographical concentration may itself be seen
as a criterion of dominance. Artisan castes are almost never
dominant because they are territorially dispersed. A peasant
caste may be dominant, although small in size, provided it is
concentrated within a limited area. A caste tends to enjoy a
higher position in a village within the area of its dominance than
outside. Thus the Kallas of Sripuram, although out-numbered
by the Vellalas, enjoy great power in the village which is in a
Kalla-dominated area.
It is probable that every State has its pattern of dominant
castes. Tamilnadu has a number of dominant castes, each
concentrated in a particular area. In this it appears to be different
from Maharashtra where a single caste, the Maratha, enjoys
dominance, or Mysore where dominance is shared between two
castes, the Lingayat and the Okkaliga.
The relationship between non-Brahmin dominant castes is
important at every level of contemporary Tamil politics but in
a special sense at the local level, particularly the village. Where
the non-Brahmins operate as a single unit, they do so generally
in opposition to the Brahmins (and today, increasingly, to the
Harijans). In the vast majority of Tamil villages there are either
no Brahmins or only a few families of priestly Brahmins who
are politically insignificant. 40 In non-agraharam villages, which
constitute the overwhelming majority of villages in T amilnadu,
the primary cleavages are often between two non-Brahmin
castes or between two sub-castes of a single non-Brahmin
caste. (The relationship between non-Brahmins and Harijans
in such villages will be considered later.)
Fairly powerful associations began to emerge among certain
non-Brahmin castes from the end of the last century. These
associations addressed themselves to social reform within the
caste and sought to secure a better position for the caste in
the wider society. A good example is the Nadar Sangam in the
southern districts which agitated successfully for the rights of
 Caste and politics in Tamilnadu 107
temple entry for the Nadars. Such associations have occasionally
provided useful bases for the mobilization of political support.
How this could be done was demonstrated effectively by the
Vanniyakkula Kshatriya Sangam (a caste association), whose two
principal branches transformed themselves into political parties
(Tamilnadu Toilers' Party and Commonweal Party), fought the
1952 elections and bargained with the Congress for positions in
the State Cabinet.41
The Vanniyakkula Kshatriya Sangam appears to have re-
mained politically inactive for a long time but was again revived
on the eve of the 1967 elections. The southern district of
Ramnad has been a stronghold of the Forward Block which
is dominated by the Thevar caste and is generally referred to as
the Thevar Party. This kind of association between a particular
caste and a political party appears, however, to be exceptional
rather than general. Caste interests - whatever their nature -
are articulated more commonly through informal networks of
interpersonal relations than through formally organized caste
associations. A caste which is dominant in a particular district
or taluka is likely to find strong representation in local bodies
as well as in local units of the major political parties. This is
so for two reasons. The dominant caste in a district or taluka
is almost always a caste which enjoys numerical preponderance.
Hence even if recruitment were made on a purely random basis,
its·representation would be high. But political recruitment is not
made on a random basis. Ties of kinship and affinity and other
personal ties play a very important part in this process at the
local level where the syncretic unity of a caste or sub-caste is
fairly strong. This unity tends to give additional weight to the
representation of the dominant caste in local politics.
Thus panchayats, Panchayat Union Councils and local units
of the Congress and DMK parties are likely to be dominated
in South Arcot district by Padayachis, in Coimbatore district by
Gaundas and in large areas ofTanjore district by Kallas.
The dominant caste operates in the political process not only
through networks of interpersonal relations but also by virtue
of an idiom which has come to be accepted by almost every
section of Tamil society. A feeling has grown among people that
members of non-dominant castes cannot compete successfully
I08 Society and Politics in India
with those of the dominant caste. Political parties act on the
basis of.this feeling and are often unprepared to take the risk
of setting up candidates from the non-dominant castes. Out of
this has emerged the familiar electoral pattern of matching caste
with caste. I am not able to provide figures to show exactly how
far this is true but there is little doubt that the feature is a very
general one. Later I shall discuss in some detail the manner in
which it has worked in a particular Assembly constituency.
In what way is caste utilized for the mobilzation of political
support? At the village level leaders of the dominant caste have
direct ties of kinship and affinity with their caste-fellows. Such
ties may also play an important part at the level of the Assembly
constituency. But electioneering at that level also involves a
more general appeal to caste sentiments. A typical example
is the election slogan quoted earlier: the Vanniya vote is for
Vanniyas alone.
The Vanniyas provide only an extreme example of what was
a very general pattern in the 1962 (and other) elections. Every-
where people discuss the chances ofcandidates in terms of their
caste affiliation. To take the example of the key constituency of
Sattur where two dominant castes, the Nadars and the Thevars,
rallied round the two principal candidates:
The intensity of communal feelings is evident from the fact that
people irrespective of political affiliation are openly discussing the
polling of votes on a communal basis. In Sattur, an enthusiastic
Nadar worker confessed that he was a D.M.K. follower, but
because he belonged to the Nadar community he would vote for
Mr. Kamraj (a Nadar). Not a single Nadar vote, he added, would
go to a non-Nadar.42
I now come to the Thiruvaiyar Assembly constituency in
Tanjore district where I was able to watch developments during
the 1962 General Elections. This is a Kalla area, the Kallas
accounting for about 30% of the total population. They have
a strong feeling of identity in relation to outsiders and Kalla
leaders in the area have close contacts with influential Kallas
in every village. In the 1962 elections three parties put up
candidates for the Assembly seat, the Congress, the DMK and
the PSP. All three candidates were Kallas. Since 1952, when the
 Caste and politics in Tamilnadu 109
First General Elections were held, only Kalla candidates have
been successful at the polls. In 1952 the Congress put up an
influential Muslim candidate but he was defeated by a Kalla. In
1957 the Congress changed its tactics, put up a Kalla candidate
and won the seat. In 1962 the Congress again won the seat but
with a different Kalla candidate. In 1967 the sitting member was
defeated by a Kalla candidate put up by the DMK.
My first insight into Kalla politics was gained from a Kalla
lawyer who had earlier contested the Thiruvaiyar seat. He
began by saying that the much publicized unity of the Kallas
was largely an illusion. This unity manifested itself only on
certain occasions: in fact, in relation to other non-Brahmins,
the three closely related castes, Kalla, Marava and Ahamudiya,
often acted as if they were one. But in the heart of the Kalla
area there were deep rivalries between villages in which different
groups of Kallas were dominant. Even a single village, he said,
may be sharply divided between different lineages of the same
Kalla sub-caste. Thus here again we see the operation of the
segmentary principle. In certain contexts all non-Brahmins
might act as a unit; in other contexts the three castes, Kalla,
Marava and Ahamudiya, act as a unit in opposition to other
non-Brahmin castes; in still other contexts Kallas may be ranged
against Ahamudiyas, or segments of the Kalla caste against one
another.

So far I have not considered the Harijans as a distinct entity

in relation to the non-Brahmins. As indicated earlier, a good
deal of ambiguity attaches to the term 'Non-Brahmin' which
by definition constitutes a residual category. A more precise
term would no doubt be preferable to it but for its historical
association with political movements and parties which have
played a most important part in South India.
The Manifesto of 1916 used the term non-Brahmin in its
most inclusive sense, to cover not only Muslims and Christians
but also the Depressed Classes. In his address to the non-
Brahmin Conference in July 1921, K.V. Reddi Naidu said, 'The
great Non-Brahmin movement transcends over caste, over reli-
gion and over the language .... It really transcended caste and
religion, and there were assembled Mohammedans, Christians,
110 Society and Politics in India

Hindus and Panchamas'. 43 Other non-Brahmin leaders also

made occasional references to the Depressed Classes and the
need for ameliorating their conditions of life.
For all this, the separateness of the Depressed Classes
remained a persistent feature of social and political life in
Madras. They had hardly any representation in the leadership
of the Justice Party; this is not surprising in view of the limited
social base of the Party to which attention was earlier drawn. In
the Memorandum presented to the Joint Parliamentary Com-
mittee in 1919 by the non-Brahmin delegation, the Depressed
Classes were listed separately from the non-Brahmins in the
Table showing the population analysis of Madras Presidency.
Further, the separateness of the Depressed Classes was given
implicit recognition in the Constitution of 1919 in which five
seats were reserved for them, to be filled through nomination
by the Governor.
It is difficult to say how far the policy of separate representa-
tion for the Depressed Classes adopted by the British fostered a
spirit of alienation among them as Gandhi had feared it would.
But it seems clear that, given the inherent cleavages of Tamil
social structure and the general atmosphere of'communal' poli-
tics, the Harijans could hardly fail to emphasize the special
character of their social and political needs. But their political
demands were to remain unorganized for a long time. During
the first three decades of the present century organized politics
was largely the prerogative ofthe Western educated urban middle
class and the representation of the Harijans in this class was
negligible.
It is to some extent an accident of history that in Maharashtra
the Harijans found a leader of the stature of Dr Ambedkar who
succeeded within a fairly short time in investing them with a
degree ofpolitical consciousness on the whole absent elsewhere.
In the meantime the British themselves gave some measure
of protection to Harijan interests for a variety of reasons. In
Madras.there were also a few Western educated Harijan leaders,
notably M.C. Raja and Dewan Bahadur Srinivasan; the latter
was chosen to represent his community at the Round Table
Conference in 1930-2 along with Dr Ambedkar. For all this,
the Harijans had to await the extension of the franchise after
 Caste and politics in Tamilnadu 111
independence before making their impact felt as a significant
force in the politics ofTamilnadu.
The issue of civic rights has played a major part in the
politicization of the Harijans in Tamilnadu. Among the non-
Brahmins politicization was spearheaded by a Western educated
urban middle class. Such a class did not exist among the
Harijans in Tamilnadu. In their case it was more the ferment
caused by the introduction (largely from outside) of liberal
social values and the confrontation of these values with the
traditional interests of the dominant castes which created the
basis for organized political action. In the traditional system the
Harijans had accepted their disabilities as a matter of course.
As these disabilities came to be removed by law and as the
Harijans sought to translate the new laws into practice, they
came increasingly into conflict with the organized opposition of
the dominant castes.
The disabilities from which the Depressed Classes suffered
with regard to the use of amenities such as wells, roads and
temples, or status symbols such as dress and ornaments, were
generally more severe in Madras Presidency than elsewhere.
Under the liberating influences of British rule and Gandhism
the Harijans made attempts to do away with some of these
disabilities. These attempts often met with reprisal from the
dominant castes. In T amilnadu, Ramnad district has been a
major arena of conflict, although such conflict has been a
pervasive feature of the relations between non-Brahmins and
Harijans throughout the State during the last three decades.
For all the stubborn opposition of some dominant non-
Brahmin castes, the Harijans have forged rapidly ahead in
their bid for civic equality. But their journey has been by no
means smooth or easy; the price of every significant advance has
been a measure ofviolence. Today one can perceive everywhere
a change of mood among them, particularly among those of the
younger generation. In Tanjore district and elsewhere young
Harijans acquired a taste of organized politics in the 1950s
when their support was mobilized by the Communists and
the DK against the Brahmin landowners. Independence, and
subsequently four General Elections, have made the Harijans
sensitive to their political rights and today they are no longer in
112 Sodety and Politics in India

a mood to have their houses burnt or their property destroyed

without retaliation.
The introduction of adu~t franchise changed their political
situation more radically than was the case with any comparable
section of Tamil society. Although fewer in number than the
non-Brahmins, they account for no less than 18% of the
population of Tamilnadu. Their situation in relation to the
non-Brahmins is, therefore, rather different from that of the
Brahmins who constitute only about 3 % of the total population.
There are talukas in Tamilnadu where Harijans are matched
fairly evenly with non-Brahmins and now a few villages in which
they out-number them.
Along with their strength of numbers, the Harijans also
evince a high degree of unity. There are still many diacritical
differences between them and the non-Brahmins and their
internal solidarity in relation to the latter is often very strong.
This internal solidarity derives from many factors of which I
shall here consider only one. In the Tanjore-Trichy area (and
elsewhere too, though perhaps less noticeably) their unique
position in rural society is made clearly visible in the settlement
pattern of the village. They live in separate streets, known by
a separate term, cheri, which are generally at some distance
from the main village; frequently they are situated in the midst
of paddy fields at a distance of three to four furlongs from
the village centre. Further, the layout of these cheris differs
significantly from that of the main village, so that two cheris
which are physically contiguous and form a single social unit
may be attached to two revenue villages which are quite distinct
from each other. These cheris are linked by social and political
ties which often cut across the boundaries of the 'village' as this
is perceived by non-Brahmins or Brahmins. 44
Perhaps because of the physical isolation of the Harijans,
traditional caste organizations seem to have survived to a
greater extent among them than among other castes. At least
in those areas of T anjore district which I came to know
directly, the traditional kuttam of the Pallas and their leaders,
the nattanmaikkarans, still exercise a measure of authority,
whereas similar institutions which once existed among some of
the non-Brahmins are now no longer to be found. The existence
 Caste and politics in Tamilnadu 113
of these traditional institutions often facilitates the mobilization
of support by Harijan leaders on a caste basis.
Along with numerical strength and organization, the Harijans
are also able to carry a certain measure of violence into political
life. The role of organized violence in politics, particularly local
politics, has not been sufficiently stressed in studies made in
India so far. Yet the support of people with a reputation for
violence is an important factor in village politics in contemporary
India. In Tamilnadu the Brahmins find the odds heavily against
them in this regard. Nothing is more repugnant to the Western
educated Brahmin than to be engaged in a village brawl with
members of the lower castes. Such consideration of self-esteem
do not deter the Harijan from confronting the non-Brahmins.
However, al~ough the Harijans constitute important reser-
voirs of political power, there are many factors which stand in
the way of this power being actualized. Their economic position
is in general very weak and this weakness is frequently used
against them by the non-Brahmins. A long tradition of servility
often prevents them from asserting their rights, although young
Harijans-are rapidly developing a spirit of challenge. Finally, lack
of education and contact with the outside world stands in the
way of their developing some of the skills which are essential for
organized politics.
In Tanjore district the Communist Party and the Dravida
Kazhagam have both played a part in the politicization of
the Harijans. An active kisan Movement was launched in the
early 1950s and Mannargudi taluka became for a brief period
the venue of bitter political strife. Young Harijan agricultural
labourers were widely mobilized by the Communists who at that
time collaborated with the DK. The kisan Movement subsided
after a short while and the Harijans soon realized that they were
up against something far more powerful than a few Brahmin
mirasdars: when they tried to challenge the landed interests of
the dominant non-Brahmin castes, they were put down with a
heavy hand. None the less, the experience ofthe early 1950s has
had some effect in kindling a spirit of challenge among them.
When this spirit of challenge confronts the entrenched inter-
ests of the dominant castes, the result is often a measure of
violence. In the area around Sripuram where I did fieldwork,
114 Society and Politics in India

the Kallas constitute the dominant caste and they are feared
throughout Tamilnadu for their violent ways. Relations between
the Kallas and the Harijans have been strained for some time.
In the village panchayat of Sripuram, the Harijans have five
members against six non-Brahmins and they generally have a
submissive attitude towards the Kalla President of the panchayat.
In fact, meetings of the panchayat are often held without the
Harijan members being informed, and the latter are too weak to
protest against this kind of irregularity. In the adjacent village of
Ponavasal, however, things are very different: the Harijans are in
a majority in the panchayat and the panchayat President is himself
a Harijan.
In another neighbouring village also the panchayat President
is a Harijan. In this village, known as Maharajapuram, hostilities
between Harijans and Kallas led to the murder of a Kalla
landowner, allegedly by or through the collusion of the Harijan
President of the panchayat. Some of the Harijan residents of the
village were arrested and the Kallas tried to storm the court
room at Tanjore in order to do violence to them. 45
Certain sections of the non-Brahmins are becoming deeply
resentful of the militant attitudes of the Harijans which, in their
view, are fostered by the ruling party and the Government. For
the poor and the landless non-Brahmins, the concessions to
which the Harijans are entitled by law are a thorn in the flesh.
For the landowners of the dominant caste, the rising demands
of Harijan tenants and labourers appear as threats to their social
and economic position. It is clear that the dominant castes in the
villages are by no means reconciled to the high ideals ofequality
and social justice which seem to motivate legislatures at the State
and the Centre. These conflicts are likely to persist irrespective
of changes in the party in power. And the party leaders in their
tum cannot afford to ignore the demands of the Harijans who
constitute such an important reservoir ofvotes.
Old conflicts between the Harijans and the non-Brahmins
are sometimes expressed in the new idiom of party politics.
Ramnad district became the centre of turmoil in 1957 when
riots broke out between the Harijans and the Thevars (who
are closely related to the Kallas discussed above). This time
the issue was a by-election in which the Harijans supported the
 Caste and politics in Tamilnadu 115
Congress candidate against a Forward Block candidate set up by
the Thevars. Ayoung Harijan leader called Immanuel was mur-
dered because he is said to have insulted the much-respected
Thevar Forward Block leader, U. Muthuramalinga Thevar. The
Thevars were put down with a heavy hand by the Government,
allegedly because they had consistently opposed the Congress
party then in power. 46
The politicization of the Harijans has, in a sense, helped
to sharpen their identity in relation to the upper castes. But
it has also drawn them into new networks of relationships
which repeatedly cut across the barriers of caste. Harijan and
non-Brahmin leaders have learnt to depend upon each other for
support and patronage. New forms of association such as parties
and panchayats are developing which are based on other loyalties
than those of caste. It is true that such associations often mirror
the cleavages of the wider society but there are clear indications
that this is by no means always the case.

In the foregoing I have given an account of the part played by

social entities of a certain kind in the political life of T amilnadu.
These entities are variously referred to as sub-castes, castes or
caste-groups. Although there are enormous differences between
a sub-caste such as the Vadama Smarthas and a broad aggregate
such as the non-Brahmins, they are similar in one important
respect: they are both based on particularistic criteria and as
such are to be distinguished from universalistic groupings
of the kind which democratic parties and governments are
in principle supposed to be. In India, where particularistic
groupings occupied such an important position in the traditional
system, it would be unreal to expect the democratic process
to operate without taking any account of them. But are such
particularistic identities the only ones which are relevant to
politics in India today? And does not the political process itself
create new identities which cut across those of sub-caste, caste
or caste-group?
Although most scholars would agree that caste and politics
are closely related in certain parts of contemporary India, their
assessment of the significance of this is likely to vary. There are
some, like Srinivas, who would be inclined to argue that the
116 Society and Politics in India
political process tends to strengthen the loyalties of caste at least
in the short run: 'One ofthe short-term effects ofuniversal adult
franchise is to strengthen caste'.47 There are others like Gough
who would argue that politics in the modern sense tends to be
disruptive of caste.48
Before turning to these questions, it is well to remember
that there are everywhere in India today forces external to the
political system which tend to erode the loyalties of caste. I shall
consider briefly some of the factors which, on the one hand,
weaken the diacritical and syncretic unity of caste and, on the
other, create interests based on income, occupation, education,
etc., which tend increasingly to become dissociated from the
structure of caste.
Castes can best be viewed as status groups and as such they
should be distinguished from classes. As status groups, castes
are differentiated from one another through the pursuit by
tradition of distinctive styles of life. Over the last hundred
years new criteria of social differentiation have been introduced
- Western education, occupation in non-traditional sectors and
so on. To the extent that the new forms of differentiation run
along traditional grooves, caste loyalties tend to be reinforced.
We saw how the introduction of Western education at first
served to increase the social differentiation between Brahmins
and non-Brahmins, leading to political conflict between them.
However, when the new forms of differentiation cut across
traditional ones, the loyalties of caste may be weakened. And
there is little doubt that castes are becoming more and more
heterogeneous in terms of income, occupation and education
and that new status groups based on these criteria are likely to
compete with caste for people's loyalties.
It appears that caste is less crucial to one's status identification
in urban as compared with rural areas. In the cities, Brahmins
frequently work with non-Brahmins in the same professions,
live in the same neighbourhoods and send their children to
the same schools. Such people are likely in their styles of life
to have more in common with one another than with their
caste fellows in the rural areas, although it must at once
be pointed out that traditional patterns of behaviour tend to
show remarkable persistence in Tamilnadu. Even though urban,
 Caste and politics in Tamilnadu 117
Western educated professional Brahmins and non-Brahmins do
share many common patterns of behaviour, they are mutually
differentiated with regard to many others.
It is probable that caste plays a less important part in urban
than in rural politics. A recent study of trade union politics
in Coimbatore tends to confirm the view that caste enters
into the political process there in only a marginal way. 50
Among textile workers in Coimbatore, income, occupation and
personal loyalties tend to play a far more important part in the
determination ofpolitical attitude than caste. The factory system
tends to break down the homogeneity of caste and to replace its
unity by unities of a different kind.
The political process seems to have a dual effect on the caste
system. To the extent that the loyalties of caste or sub-caste are
consistently exploited, the traditional structure tends to become
frozen. Thus there can be little doubt that the Non-Brahmin
Movement arrested to some extent the attenuation ofcaste iden-
tities by driving a wedge between Brahmins and non-Brahmins.
But the political process - whether in Tamilnadu or elsewhere -
does not operate by mobilizing only the loyalties ofcaste. To the
extent that it leads to new forms ofassociation and new alliances
cutting across caste, it tends to loosen the traditional structure.
There are some who have gone even further and argued that
the political process destroys irrevocably the very nature ofcaste.
Leach raises the question, 'If a caste group turns itself into a
political faction does it then cease to be a caste?'51 He answers
the question with a clear affirmative, but his answer is based on
a peculiarly personal view of caste: 'People of different castes
are, as it were, of different species - as cat and dog ... : But
with members of different grades of the same caste, the exact
opposite is the case'. 52 Enough has been said earlier about the
manner in which castes_are divided and subdivided to expose the
futility oftrying to impose a radical opposition between 'different
castes' and 'different grades of the same caste'.
In Leach's view, competition for power is antithetical to the
very nature of caste and consequently wherever castes act 'in
competition against like groups of different castes ... they are
acting in defiance of caste principles'.53 In a democratic system
virtually any kind of social identity may be used as a basis for
118 Society and Politics in India

mobilizing political support and it is difficult to deny that caste

continues to play a major part in this regard in Tamilnadu
today. It is equally difficult to see how a particular mode of
social grouping becomes its very antithesis by the sole fact of
engaging in competition with groupings of a like order. To quote
an earlier statement:
The continuity of the politically organized castes oftoday with their
forebears is not simply one of habits and tradition, but also one of
personnel. Shall we say that as soon as people start talking about the
"Padayachi vote", the Padayachis cease to be a caste, and become
its antithesis? What shall we call them then?S4
While some have emphasized the role of caste in Indian
politics there are others who have drawn attention to the part
played in it by factions. Brass has characterized Indian politics as
a politics of factional bargains. 55 In many ways factional politics
may be contrasted with caste politics. A faction is generally
mixed in its caste composition and factional loyalties cut across
caste. It happens very rarely that groups which contend for
power are homogeneous in their caste composition. A political
unit, if it is to be viable, has generally to draw its support from
a number of castes and not just one. Conversely, a caste whose
members enjoy social prominence is likely to be divided by
rival contenders for power. But even when a caste is divided
by factions, support within the faction may still be partially drawn
on the basis of caste. Faction leaders often choose their inner
circle from among persons who enjoy some support in their
respective castes. When two rival groups are similar in their
caste composition it does not follow that their leaders cease
to appeal to caste in their efforts to undercut each other's
support.
There are various ways in which participation in organized
politics tends to alter the structure of caste, although the
processes by which this comes about are only now beginning
to be investigated. Rudolph and Rudolph have drawn attention
to an important change which accompanies the emergence of
caste associations. A caste association is no longer a birth status
group in which membership is automatically ascribed at birth;
membership in a caste association has to be acquired, although
 Caste and politics in Tamilnadu 119
the base of recruitment may be restricted to a single caste or a
group of castes. 56
Party programmes can have significant consequences for the
unity of caste. They may, and increasingly do, lead to splits
within a caste and to alliances across castes. It is thus important
to investigate not only how caste acts upon the political system
but also how the latter acts upon caste.
Political alliances between castes and between castes and
political parties tend to be rather unstable. Sub-castes, castes
or caste-groups which are in the same camp today may find
themselves in opposite camps tomorrow. It is perhaps becoming
less and less common for the same caste or sub-caste to identify
itself persistently with a particular political party or movement
over any significant length of time. And to the extent that a
caste does not identify itself persistently with any particular
party but tends to divide and sub-divide and to enter into
multifarious alliances across its boundaries, its very contours
ultimately become blurred.
Politicians of every kind in Tamilnadu have learnt to
manipulate caste in the furtherance of their interests. But
politics is a dynamic phenomenon and the politician whose
only skill is caste politics is likely to become obsolete. In this
context what Dahl says ofethnic politics in the USA is particularly
relevant: 'In order to retain their positions, politicians are forced
to search for new issues, new strategies, new coalitions'. 57 This is
in many ways as true of caste politics in India as ofethnic politics
in the USA.
The disruptive effects on caste of flexible and changing
political arrangements must not be exaggerated. It is true that
political parties tend to cut across caste but so do factions and,
as Brass has rightly pointed out, factions are a feature of the
traditional order. 58 Caste loyalties have persisted in spite of
decades of factional politics and it is unlikely that party politics
by itself will lead to their immediate dissolution. This is not
to deny the significance of parties and other differentiated
structures which now operate in addition to factions, but
only to draw attention to certain persistent elements in the
cultural idiom of Indian society in general and Tamil society
in particular.
120 Society and Politics in India

Parties, to the extent that they are responsible for the aggre-
gation of interests, tend increasingly to cut through the organi-
zation of caste, uniting people belonging to different castes and
dividing those belonging to the same caste. Everywhere leaders
of the dominant caste try to capture the major political parties
and this is rarely if ever done on a basis of planned, mutual
understanding. Parties in their tum try to create an appeal for
every major group and not merely a single group. As Lipset has
argued, 'A stable democracy requires a situation in which all the
major political parties include supporters from many segments
of the population'. 59
The relevance of Lipset's argument to Tamilnadu politics can
be illustrated with a brief consideration of the changing relations
between the Brahmins and the DMK. When the DMK started
its career as a separate political party in 1949, its leaders were
closely associated in the popular mind with anti-Brahminism.
The party decided to start with a clean slate, declared itself
against discrimination and even offered to accept Brahmins as
members. The Brahmins remained for a long time suspicious
of the new party and were not in any case much in sympathy
with either its style or its policies. However, in 1962 the DMK
emerged as one of the strongest Opposition parties in any state,
and this it did only after it had come to terms with the Brahmins
and shown itself prepared to treat them with consideration. In
fact, in more than one constituency the DMK owed its success
to Brahmin support. In 1967 Brahmin support for the DMK
was, if anything, even more enthusiastic. Today the Brahmins'
attitude towards the DMK is very different from what it was in
1957 although even now not all ofthem are its ardent supporters.
Even the Congress, which has in recent years alienated itself so
much from the Brahmins, set up Brahmin candidates in certain
constituencies and gathered Brahmin votes.
Today the political system is becoming increasingly differen-
tiated. This doe~ not, of course, mean that it is unrelated to
caste and class or that it will be so in the near future. But as
the political system becomes more and more differentiated, new
loci of power develop and these tend to acquire a weight of their
own. In the past - at least at the local level - dominant caste
and faction were probably the only significant loci of power,
 Caste and politics in Tamilnadu 121
and the faction itself was largely structured by caste. With
the development of Panchayati Raj and political parties and
machines ,of various kinds, this is no longer the case. Today
it is possible for a man to acquire a certain measure of power
by virtue of his position in the party hierarchy, irrespective of
his caste or class. No doubt membership of the dominant caste
helps a great deal, but other factors are also becoming important.
A fuller understanding ofpolitics in Tamilnadu can be achieved
only by considering the changing relations between caste, class
and party in addition to other major sources of power.
 5 Thepolitics of'non-antagonistic'strata

In recent years social anthropologists have developed a certain

conception or 'model' of traditional Indian society based primar-
ily on a set of ideas elaborated in the classical Hindu texts. This
conception has come to play an important part in the interpre•
tation of social life in India in both historical and contemporary
times. I propose to examine the basic ideas underlying it and its
usefulness as a scheme of analysis, particularly in regard to what
can broadly be described as political activity.
I shall begin by describing some of the main featl,lres of the
model. (1) It is based primarily on the ideas held or expressed by
certain sections of society and not on the observed or recorded
behaviour of people. (2) It attaches a kind of primary and
universal significance to caste as this has been conceived
in the classical texts. (3) The entire system is viewed as
being governed by certain more or less explicitly formulated
principles or 'rules of the game'. (4) The different castes which
are the basic units in the system are conceived as fulfilling
complementary functions and their mutual relations as being
'non-antagonistic'. .
There is no question that this model brings into focus some of
the most important features oftraditional India society. Its utility .
has been amply demonstrated in the works of Louis Dumont,
particularly in his recent impressive analysis of the values of
traditional Indian society (see Dumont, 1966). However, there
are two dangers in the use of a model of this kind. When it
is made too general it can be applied to almost every society
and therefore does not tell us very much about the specific
properties of any society: I shall illustrate this point later with a
briefcomparison of traditional Indian society with contemporary
socialist societies. When, on the other hand, the model is made
too specific it fails to take into account certain crucial features
of economic and political life.
 The politics of'non-antagonistic' strata 123
The strength and the weakness of this model, which has been
most systematically elaborated by Dumont, is that it is concerned
essentially with ideas and values. As such it has been most
successful in the interpretation of systems of religious belief. It
has failed, however, to give its proper place to material interests
in social life. For this reason its application to the analysis of
political and economic problems leads to major distortions of
social reality.
The purpose of this paper is not to deny the validity of a
sociology of values and ideas. It is rather to draw attention
to another realm of social reality, that of material interests,
which has been somewhat neglected, particularly in studies of
traditional society and culture in India. Interests not only exist
.as much as values and ideas but can also be represented as
structures. A comprehensive sociology of India must give their
proper place to both values and interests for it is the dialectical
relation between these two which gives to human societies their
distinctive qualities.

As I have mentioned, two major sources of distortion in our

understanding of Indian society are the overriding importance
given in it to caste, and the conception ofcaste itselfas a system
of non-antagonistic groups having complementary functions and
each enjoying its own set of privileges. When the basic groups in
the social system are defined as being non-;llltagonistic very little
room is left for the analysis of either conflict or change. In fact,
this conception of Indian society is only one step short of the
popular nineteenth century view of it as integrated, harmonious
and unchanging.
Social anthropologists have• done much to popularize the view
of Indian society as a 'caste society'. This emphasis is partly
dictated by the preoccupation with native categories of thought
rather than the dynamics of real life. The classical Hindu texts
give a much fuller elaboration of fundamental religious values
than of the interplay ofmaterial interests in the relations between
individuals and groups. Field studies by social anthropologists
have also attached overwhelming importance to caste. Caste
groups being both visible and mutually exclusive are relatively
easy to describe and it is natural that they have attracted the
124 Sodety and Politics in India

attention of European and American scholars seeking new

insights from an alien society.
It is in some contexts unavoidable to characterize total
societies in terms of a unique structural principle. Thus, one
speaks of medieval Europe as an 'estate society', of the modern
West as a 'class society' and correspondingly of traditional India
as a 'caste society'. For certain purposes such characterizations
are useful but they can also be misleading. They tend to focus
attention on one particular aspect of a society and to divert
attention from the others. The view that traditional Indian
society was a caste society together with the definition of caste
in terms of a set of ideal principles has led to the construction
of models from which economic and political conflict have been
virtually excluded. One has either to adopt a broader conc.eption
of caste or to take into account other principles of organization
in addition to those ofcaste. In doing this one will have to devise
conceptual categories which will apply to both traditional and
modem India.
It is hardly necessary to repeat that even in traditional India
control over property was not always dependent on caste (see,
for instance, Metcalfe, 1969, pp. 133-41). As such, the relations
of production had a certain autonomy although they were often
subsumed by caste. The distribution of power was likewise
partially independent of caste particularly at higher levels of
territorial organization than the village. It would thus be a
mistake to try to understand traditional Indian society solely
in terms of the rules of intercaste relations. Conversely, in
'class societies' such as modem America, religious and ethnic
divisions are not only partly independent of class but also govern
social and political life in important ways. A scheme which seeks
to reduce every aspect of social reality to a single principle of
organization cannot account either for the diversity offacts or for
changes in patterns of organization and can present only a flat,
monotonous and one-dimensional picture of social life. In short,
it is misleading to represent India as a 'caste society' in the same
way and for the same reason as it is misleading to characterize
the United States as a 'class society'.

I now tum to the view that caste is a system in which the relations
 The politics of'non-antagonistic' strata 125
between groups are non-competitive and from which antago-
nism and conflict are in principle excluded. This view has been
tersely put forward by Leach who maintains that 'wherever caste
groups are seen to be acting as corporations in competition
against like groups of different castes, then they are acting in
defiance of caste principles'(l960, p. 7). There is further
the suggestion that in such cases these groups are acting
in accordance with class principles. This way of contrasting
caste with class seems to me to be based on a rather uncritical
acceptance of the classical Hindu view of Indian society (in
which all is harmony) and the Western radical view of capitalist
society (in which all is contradiction).
Much the same view has been put forward ·by Bailey who
speaks of castes as groups 'which co-operate and do not
compete'(1963, p. 121). 1 It is, of course, possible in theory
to define the systematic properties of caste in such a way as
to exclude both competition and conflict and indeed this would
correspond very well with the representation of it in classical
Hindu texts. But the question is, how far one can proceed in
the understanding of economic and political life in India on
the basis of such a definition. Very few anthropologists would
seriously contend that competition, conflict or even politics was
absent from real life in the traditional set-up. How, then, was
competition organized in the past? What were the social bases
of conflict? Did caste play no part whatever in the organization
of mutually opposed factions or political groups? Such questions
can be fully answered only by empirical research but it should be
possible to clarify certain issues deductively.
Both Leach and Bailey have sought to accommodate compe-
tition and conflict in their models of the caste system. Leach has
done this by means of a radical opposition between castes and
what he calls 'grades within a single caste'. Between different
castes there can be no competition, whereas the exact opposite
is the case between different grades of the same caste. 'In this
respect grades within a single caste have the nature of social
classes rather than of castes' (Leach, 1960, p. 7).
Leach's attempt to invest the caste system with a semblance
of vitality by contrasting caste grade with caste is bold but
not convincing. Having found the ideal conception of caste
126 Society and Politics in India

insufficient for explaining politics, he invents the notion of

caste grade but the classical texts themselves do not make
any distinction between the two, and both Dumont (1966,
pp. 85-90) and I (Beteille, 1964, pp. 130-4) have suggested
why this distinction is inappropriate. Leach rejects Yalman's
treatment of 'distinctions in grade within a single named caste
as different only in degree from distinctions between separate
named castes' (1960, p. 7) but gives no convincing reason for
doing so. Those who have done field research in India would
surely support Ya1man and not Leach.
Bailey's attempt to accommodate politics in his model ofcaste
resembles that ofLeach up to a point but it has wider theoretical
implications. In his scheme the basic distinction is between the
dominant caste and subordinate castes. 'Only the dominant caste
has an autonomous political existence, not as a corporate politi-
cal group, but as [a] field for political competition. Certainly no
subordinate caste is a corporate political group' (1963, p. 118).
We might add that a social entity need not be a corporate group
in order to be politically significant. Lineages in tribal societies,
classes in modem society or racial groups where they co-exist
often play a significant part in ordering political relations without
always being corporate groups.
Bailey argues that 'political cleavages (i.e. competitive rela-
tionships) are between vertical groups, not social strata: coopera-
tive relationships run up and down between families in the
dominant caste and families in the service castes' (1963, p.
118). This seems to suggest a classification of societies into two
broad kinds, those in which conflict takes place between strata
and those in which it is contained within the dominant stratum.
Marxian sociologists in the East European countries sometimes
use a similar scheme for differentiating 'class societies' from
societies ofother kinds. I do not believe that such a distinction is
always useful and I shall try to justify my misgivings by examining
certain assumptions implicit in Bailey's formulation.
When Bailey argues that 'only the dominant caste has an
autonomous existence' as a 'field for political competition', he
assumes that every area has always a single dominant caste. How
are we to define the terms of competition when a village has
more than one dominant caste? 'Chis is only partly a question
 The politics of'non-antagonistic' strata 127
of terminology. Suppose in a single village Smartha and Shri
Vaishnava Brahmins are evenly matched in terms of numbers
and control over land: are we to regard them as two separate
castes or as 'two grades of the same caste'? Suppose, instead,
they are Jats and Ahirs: can we reasonably extend the meaning
of 'caste' far enough to regard them as two sections of the
same unit?
However broadly we may define the term 'caste', there is
no reason to assume that every village has a single dominant
caste or that the presence of more than one dominant caste
in a village is of recent origin. It is well known, for instance,
that in Tanjore district agraharams were set up in villages which
had pre-established communities oflandowning non-Brahmins;
later, Maratha Kshatriyas were granted land and occasionally
settled in villages where agraharams were already in existence.
Thus the balance of power between castes in a village might be
altered from above. It might also change due to internal causes
such as differential birth and death rates or differential rates of
migration. Adrian Mayer (1955, p. 1147) gives the example of a
village where Rajputs are now dominant but where formerly they
shared dominance with Dhakars. I am not sure whether Leach
would consider Dhakars and Rajputs as grades within a single
caste or whether Bailey would regard them as belonging to the
same stratum.
Once we leave the village and enter into wider territorial fields
it becomes increasingly clear that we have to deal with not one
but a number of dominant castes. Even in a small chiefdom
the dominant castes might range from Brahmins and Rajputs to
fairly low peasant castes. Srinivas has recently brought together a
body of evidence to show how the fluidity ofpolitical boundaries
in the past enabled different groups to compete for power (see
Srinivas, 1966). It is unlikely that rival groups when of different
castes did not draw upon the loyalties of caste in competitions
of this kind.
A recent case study by Frykenberg shows effectively that the
allocation of power between castes was much more differenti-
ated and fluid than is allowed for in the schemes discussed above
(see Frykenberg, 1965). This analysis of Guntur district between
1788 and 1848 reveals the existence of several loci of power:
128 Society and Politics in India

landownership, the official bureaucracy, and so on. It also shows

that these were controlled and used by members of a succession
of different castes such as Nyogi Brahmins, Desastha Brahmins,
Velamas and Kammas. Frykenberg centres his study around
the administrative bureaucracy and shows how the Desastha
Brahmins used the ties of caste to keep it within their control.
But he also tells us that in the earlier phase the Desasthas had
to contend for power with the zamindars many ofwhom were not
Brahmins at all but Kammas, Velamas and Rajus. This structure
ofcompetition was evidently not created ex nihilo by British rule,
since an official bureaucracy (controlled generally by Brahmins)
had been in existence during the Maratha period and earlier.
One may, of course, argue that in Guntur district Desastha
officials and Kamma or Velama zamindars contended for power
not as castes but in accordance with the rules of a different
system. That, as we have seen, is one way of defining caste and
it is impossible to argue against a definition except by showing
where it limits or obstructs analysis. What is important is that
officials and landowners both used the ties of caste (as well
as other ties) in gathering support to further their respective
material interests.
To make a radical distinction between the dominant caste (as
constituting the field for political competition) and subordinate
castes is to prejudge the basic issue, i.e. that of the distribution ,
of power between different castes. It is to assume that a caste,
once dominant, remains in that position for all time. What the
sociologist and more particularly the social historian has to study
is how one caste is displaced by another from its position of
dominance. My argument against this approach to the problem
is that it is static and assumes a fixed and immutable distribution
of power between castes. It leaves no room for changes in the
power positions ofcastes which are an important feature of both
traditional and contemporary Indian society.
The analysis ofpolitical life is concerned to a large extent with
material interests whose demands are often at variance with the
ideal values of a society. The initial conception of caste as a
system in which groups 'cooperate and do not compete' can
hardly do justice to this type of analysis. The attempt to save the
conception by introducing artificial distinctions between castes
 The politics of 'non-antagonistic' strata 129
and caste grades (or between dominant and subordinate castes)
leads to ambiguities and inconsistencies of the kind I have
indicated above.

I have said already that some scholars make a distinction

between two kinds of society. In the first, the basic political
cleavage is between vertical groups each of which is composed
of a similar set of strata; in the second, the principal cleavage
is between the strata themselves. Caste and estate soci~ties are
viewed as examples ofthe former and class societies ofthe latter.
I now propose to examine briefly the validity of this dichotomy.
The dichotomy rests on the assumption that in estate and
caste societies the arena of politics is narrow whereas in class
societies it is greatly extended. This is a valid distinction and
it is particularly relevant in the comparison of formal political
institutions. It is true that in traditional societies support was to a
great extent determined by pre-existing relations ofpatronage so
that the principal contenders for power were either rival barons
or rival leaders of the dominant caste (or castes) each with his
respective set of followers. Traditional societies of either kind
provided limited scope for people to switch their support from
one leader (or faction) to another.
It has been argued that the distinctive feature of politics
in modem Western societies is that there the government
is "put in a kind of market situation" (Macpherson, 1966,
p. 8). Individuals and groups are able to exercise choice in
giving power to some and withholding it from others. The
exercise of this choice is formalized through the institutions
of representative government, plurality of parties and freedom
of association. While undoubtedly these institutions give a
distinctive character to modem democratic systems they do
not by themselves define the nature of politics. In any event
one must not exaggerate either the freedom of political choice
·in the modem West or its absence in traditional societies. Social
anthropologists working in Africa have been led to give a more
comprehensive meaning to the term 'politics', one which is
better suited to an understanding of the working of traditional
societies including pre-British India.
Each society is characterized at a given moment by a certain
130 Society and Politics in India
distribution of power. In one sense politics is the expression of
tensions between those who seek to maintain this distribution (in
order to use it for various ends) and those who seek to change
it. Admittedly, the limits within which changes are sought to be
made are highly variable. At one end are attempts to replace
one set of incumbents by another; at the other, to transform
the very nature of the roles themselves. I am not sure that
it is always possible to draw a clear line between the two
sets of phenomena, between what Gluckman calls 'rebellion'
and 'revolution' or what Duverger calls 'lutte dans le regime'
and 'lutte sur le regime'. Conflict over a regime is perhaps
exceptional and occurs only at certain moments in history;
conflict within a regime is an inherent feature of all political
systems, traditional as well as modern.
In order to maintain themselves in power individuals and
groups require support. This support they obtain by drawing
upon the ties of kinship, caste, religion, race and class, or by
appealing to some more universal loyalty. Comparative studies
of political systems have clearly established that support does
not flow automatically in any society. Even where people appear
to be rigidly bound by custom, it has to be actively sought
and cultivated. This is as true of the Zulu under Shaka as
of present regimes in the West. There is little reason to
believe that in traditional India leaders of the dominant caste
could always count on the automatic support of members of
subordinate castes; or that the latter never transferred support
from leaders of one dominant caste to another, altering thereby
the distribution of power between castes.
Any conception of politics presupposes a certain indetermi-
nacy both in the composition of groups wielding power and in
the support they are able to command. Politics in this sense is
a universal feature of all human societies; one cannot picture
a society in which choice and uncertainty in the pursuit of
material interest are totally absent. Such a conception ofpolitics
leads to questions regarding shifts in patterns of support and
consequent changes in the distribution of power. Given the
nature of traditional Indian society and the importance of caste
in it, it is difficult to visualize rival groups competing for support
without making use ofthe loyalties of caste. Nor can one assume
 The politics of'non-antagonistic' strata 131
that caste was the only basis on which power was allocated and
support gathered. In fact, the task of the political sociologist is
to investigate the specific ways in which caste combined with
other factors to create and maintain particular constellations
ofpower.
It would clearly be an over-statement to contend that in
traditional India ruling groups found themselves 'in a sort of
market situation'. For the limits within which choices were
made were fairly narrow, subordinate groups· were less free
to exercise choice overtly and radical changes in the character
of the ruling group were perhaps few and far between. Such
politics as did exist was clearly not party politics in the modem
sense of the term but a politics offactional bargains. Some social
anthropologists believe that factions in India are expressions of
cleavages within the dominant caste rather than between rival
dominant castes. Actually, our knowledge of the anatomy of
factions is too limited to enable us to say anything categorical
in this regard. But given a certain instability of economic and
political power, there is no reason to assign greater importance to
intracaste as opposed to intercaste factions in either the present
or the past.2
To carry the analysis a little further, one may argue that the
market in which groups contend for power in modem societies
is fully free in only the formal sense of the term. The arena of
effective political participation, though greatly extended, never
becomes coterminous with the whole society. Even in modem
democratic systems the distribution of power is never entirely
dissociated from race, religion or region. To say this is not
to deny that considerable differences exist between politics
in traditional and modem societies. It is merely to insist that
both can and should be studied within a common theoretical
framework.
The need to accommodate both material interests and ideal
values in this framework becomes evident when we examine
available explanations of the divergence between principle and
practice. Most authors seem prepared to concede that castes
may, in fact, engage in politics but that one has to make a
distinction between principle and practice or between what
Bailey calls 'the rules of the game' and 'the way it is played'.
132 Sodety and Politics in India
One may ask whether this divergence is a recent deviation in
Indian society (as Leach would imply) or inherent in the tension
between values and interests which is characteristic of all human
societies.
The confusion seems to lie in this, that in talking about
the properties of the system Dumont, Leach and Bailey are
concerned essentially with the structure of values and not with
the structure of interests. The analysis of political processes
requires that we identify structures of interests and not explain
them away as deviations from the structure of values. To extend
the metaphor a little, the rules of the game are complex because
it is not always the same game that people are playing.
The pitfalls of ignoring certain practices because they do not
conform to particular 'rules of the game' can be illustrated from
Bailey's own analysis. It is perhaps true that the rules ofthe game
do not - or did not - permit conflicts between castes. But did
they, for that matter, permit conflicts .within the dominant caste?
Political cleavages between vertical groups divide the dominant
caste into opposed lineages, opposed branches of the same lin-
eage and opposed families within the same branch. Is it correct
to assume that such conflicts - which no doubt existed - were
more in conformity with 'the rules of the game' than conflicts
between castes? Those who invoke the rules of the game are
obliged to be consistent: conflict between agnates is as much
contrary to the rules of their game as conflict between castes.

I have argued so far that a conception of traditional Indian

society as a system in which groups 'cooperate and do not
compete', based as it is on the people's own representations,
tends to obscure some of the most important features ofpolitical
life. I shall now try to show that representations of a very similar
kind have existed in all or most societies and by themselves tell
us very little about certain aspects of social reality. Societies
throughout the ages have created images of their being in terms
of harmony, order and unity. The details ofthese have varied but
the broad outlines show almost everywhere a remarkable quality
of sameness. ·
It would be all too easy io explain away these images or
conceptions as 'myths', 'ideologies' or 'false consciousness' and
 The politics of 'non-antagonistic' strata 133
it is far from my intention to do this. Their influence on social
action is important enough to assure them a place in any scheme
of analysis irrespective of one's views about their ultimate reality.
But this is not to say that they tell us the whole truth about
human societies or even provide the best point of departure for
every type of sociological enquiry.
It is unfortunate that those who have started with ideal rep-
resentations of Indian society in their analysis of it have ignored
similar conceptual schemes which exist in other societies. Such
representations are not a unique feature oflndian society or even
of traditional societies but are present everywhere in the socialist
countries today.
A very fruitful comparison can be made in terms of the stated
principles of organization between the regime of caste in tradi-
tional India and the socialist regimes of today. I am aware that
there are enormous cultural differences between the two kinds
of societies and between their actual patterns of stratification.
But I am not now concerned with these; my objective is to show
certain correspondences between representations which people
create of the societies in which they live.
I do not wish to give the impression that there is complete
unanimity in the views expressed by East European scholars
about the nature of their own society. But such differences
as do exist are slight or insignificant when compared with
the diversity of views about capitalist society expounded by
Western scholars. At any rate, if it is possible to speak about
the stated principles oforganization oftraditional Hindu society,
one should be able to do the same for contemporary socialist
society.
Soviet scholars recognize the existence of divisions in their
own society: what they deny is the presence of antagonism
between these divisions. According to them also the divisions
(which they usually call strata) are in relations of co-operation
and not conflict with each other. Their views of their own society
are thus essentially similar to what Dumont, Leach and Bailey
believe to have been the views of the Hindus of the past about
traditional Indian society.
Contemporary Marxists in East European countries make
a radical distinction between stratification in capitalist and
134 Society and Politics in India

socialist societies. This distinction centres around the presence

or absence of what is variously described as 'antagonism' or
'antagonistic contradiction'. The distinction between 'antago-
nistic' and 'non-antagonistic' contradiction was elaborated by
Stalin who argued that the relations between strata in socialist
society were non-antagonistic whereas under capitalism the
corresponding relations were antagonistic. 3
Soviet scholars generally represent their society as being
divided into two major classes, namely, workers and peasants. In
addition to these, there is a third category, the 'intelligentsia', to
which the term stratum and not class is applied (see Nemchinov,
1957, pp. 179-84). The recognition of the separate identity of
the intelligentsia and the reluctance to characterize them as a
class reflects a basic ambiguity in Soviet Marxism to which both
Marcuse (1961, pp.91-3) and Aron (1964, p. 127) have drawn
attention.
The persistence of classes in socialist society (over and above
the division into workers and peasants) is an anachronism from
the viewpoint of Marxist theory. At the same time, the existing
divisions between workers and the 'intelligentsia' or between
various groups within the bureaucracy are too obvious to be
denied recognition. Their presence is taken into account in a
model elaborated by the Polish economist, Oskar Lange. Lange
(1962, pp. 1-15) makes a clear distinction between classes and
strata, denying the existence of the former in socialist society
while recognizing that of the latter. Relations between classes
are, in his view, antagonistic, being based on property; relations
between strata, which arise from the division of labour, are, on
the other hand, complementary and non-antagonistic.
The same contrast between 'classes' and 'strata' has been
made on the same basis in a recent paper by the Polish
sociologist Wlodzimierz Wesolowski. Wesolowski (1967, pp.
145-64) maintains that there are 'workers' in both capitalist and
socialist society and in both cases they are involved in a certain
type of production (industrial production) and they perform a
certain type of work (manual work). But their relations with
other groups are radically different in the two kinds of society.
Therefore, he argues that workers in socialist society should be
described as a 'stratum' and not as a 'class'.
 The politics of'non-antagonistic' strata 135

My purpose in drawing attention to Lange's model ofsocialist

society is twofold. It is to argue, first of all, that formulations
of ideal patterns of relationship have certain basic things in
common irrespective of the societies in which they emerge. It
. is to argue further that an uncritical acceptance of statements
regarding the rules of the game may hinder rather than help the
analysis of effective social relations. Even Marxist sociologists
view with scepticism the distinction between 'antagonistic' and
'non-antagonistic' contradiction and few sociologists outside
Eastern Europe would deny the existence there of conflicts
within and between strata. Not only do such conflicts exist but
they constitute the very core of what the political sociologist has
to study, rules of the game notwithstanding.
East European ideologues are not alone among publicists
of the modem world in representing their own society as
basically harmonious, as one in which the different strata
fulfil complementary functions. The view that Western society
is tom by class conflict is as often as not an outsider's view of
it. It is not true that those who have written about capitalist
society from within have all taken the Marxist view that relations
between classes are in their essence relations of conflict. On
the contrary, the nineteenth century produced a considerable
number of writers whose mission was to show that the different
classes were complementary to each other and that basically the
social order was in a state of equilibrium.
The idea ofthe complementarity ofclasses can be amply illus-
trated from the writings of classical economists, liberal political
philosophers, and sociologists of the 'organismic' school. The
persistence of this conception ofsociety has been nicely summed
up by Ossowski and I cannot do better than to quote him:
From legendary Agrippa, from Aristotle to Herbert Spencer and
Durkheim, from St. Clement ofAlexandria to modem papal encyc-
licals, to Oxford Movement declarations and fascist manifestos, the
defenders ofthe existing order endeavoured to represent it as based
on mutual dependence resulting from the division of tasks equally
useful for the whole society (1956, p. 21).
It is true, as Ossowski points out, that there were other represen-
tations ofsociety in Western culture but are we certain that these
136 Society and Politics in India
did not also exist in India? At any rate, the classical religious texts
are not the best place to look for them.
Some Western sociologists have constructed conceptual
schemes of their own society which are not in their essentials
different from those of classical Hinduism or of contemporary
socialist doctrine. The argument of Durkheim is characteristic:
organized society is based on complementary differences and class
conflict is an abnormal form of the division of labour.
It was a common argument among nineteenth-century Con-
servatives and Liberals in the West that not only were social
classes complementary but that each had certain rights and
society had an obligation to protect the weak from being
exploited by the strong. This was the theory of it, not very
different from the one that every caste, not merely the upper
elite has its 'privileges'. It is perhaps true that legally every caste
had its privileges just as capitalists and workers have been since
the nineteenth century 'equal before the law'. Equality before
the law does not exclude the political domination of one set of
people by another; privileges conferred by traditional usage can
be set aside by the will of the strong and the powerful.
One may say, following Mannheim (1960, pp. 97-104), that
the province of politics begins where that of law ends. It is
difficult to study the realities of politics in capitalist society
if one remains confined to the maxim that all men are equal
before the law. It should be equally difficult to study politics
in traditional India unless one is prepared to see beyond the
principle that each caste is entitled to its special privileges. The
proper subject matter of politics is the ambiguity in practice
of privileges which are clearly defined in principle. In some
societies the law defines a certain fundamental equality of every
citizen; in others, it grants special privileges to various categories
of people. But in all societies there is an area of indeterminacy
beyond what is formally defined by law; the sociologist has to
analyse how choices are made within this area.
To justify or even to explain the continued existence of a
society, it is perhaps simplest to present it as a unity in which the
different components fulfil necessary and complementary func-
tions. Thus, castes are represented as complementary to each
other in traditional India; likewise classes in nineteenth-century
 The politics of 'non-antagonistic' strata 137
Europe and strata in the socialist countries today. To represent
such divisions as complementary is perhaps the easiest way of
explaining why people accept them and why they persist over
time. The Marxist theory of capitalist society was exceptional
in that it sought to explain and justify not the existence of a
particular order but its dissolution.

The kind of questions that I have raised would perhaps gain

in clarity by being related to a debate of a more general
scope in contemporary sociology. This debate centres around
the priorities to be assigned to values and to interests in the
understanding of social life. Dumont has clearly ·chosen to
assign primary significance to values and in this he seems to
be followed by Leach and apparently in some measure also
by Bailey. Anthropological studies of Indian society are clearly
dominated by a concern for values and show a corresponding
neglect of the systematic analysis of interests.
On this point Dumont makes his position quite explicit: 'A ce
niveau, le systeme des castes, c'est avant tout un systeme d'idees
et de valeurs, un systeme formel, comprehensible, rationnel, un
systeme au sens intellectuel du terme' (1966, p. 53). 4 And he
soon goes on to say 'Notre premiere tache consiste a saisir
ce systeme intellectuel, cette ideologie' (1966, p. 53).5 Leach
appears to follow Dumont quite closely but Bailey's general
position is not as simple. He was not only one ofthe first to make
detailed monographic studies of economic and political life but
also challenged Dumont's and Pocock's excessive preoccupation
with ideas and values (see Bailey, 1959, pp. 88-101).
It is, therefore, by no means true that students of Indian
society have ignored altogether either economic and political
life or material interests in general. What is lacking, however, is
a comprehensive framework for the study of interests of the kind
which Dumont has developed in relation to his study of values.
The construction of such a framework~ be a major step in the
direction of a more balanced appraisal of social reality in India.
The main points of difference in the broader field of socio-
logical theory between approaches based on values and on
interests have been summarized by Dahrendorf (1959). He
describes the adherents of the two principal approaches as
138 Sodety and Politics in India
Utopians and Rationalists, a choice of terminology which no
doubt reflects his own preference. Alternatively, Dahrendorf
speaks of two theories or metatheories which he describes as
the 'integration theory of society' and the 'coercion theory of
society'.
It is not necessary to follow Dahrendorf in every detail of
his analysis but one of the points which he makes is of
particular interest here. This is the general preoccupation of
those concerned with the study of values with order, harmony,
integration and a static view of society and the corresponding
neglect by them of differences of interest, coercion, conflict
and change. The argument that the emphasis on values at the
expense of interests leads to certain basic distortions of social
reality had been earlier put forward by Lockwood (1956, pp.
134-46).
It might appear a little harsh to label as Utopians those who
have contributed so richly to the understanding of Indian society.
But it would be hard to deny that there is a certain Utopian
element in Dumont's insistence on the universal significance
of dhanna, in Leach's contention that 'every caste not merely
the upper elite has its "privileges"' and in Bailey's definition of
castes as groups 'which co-operate and do not compete'.
The preoccupations with values on the one hand and with
interests on the other derive respectively from the sociologies
of Durkheim and Marx. The influence of Durkheim has greatly
outweighed that of Marx among social anthropologists engaged
in the study of primitive or pre-industrial societies. It is inter-
esting that Leach was one of the first British anthropologists to
lament the preoccupation among his colleagues with '"functional
integration," "social solidarity," "cultural uniformity," "struc-
tural equilibrium"' (1954, p. 7). In his own studies of caste,
however, he does not seem to have gone far enough to redress
the balance towards a more dynamic conception of society.
Recently, both Lockwood and Dahrendorf have emphasized
the importance of interests in the study of human society.
Dahrendorfs argument is that while it is useful to study both
values and interests, which particular framework one chooses
ought to depend on the problem one wishes to study. He has
in his own work shown that systematic studies of interests are
 The politics of'non-antagonistic' strata 139
as much within the proper domain of sociology as systematic
studies of values. It is important now to tum our attention to
the former kind of study in India because the role of values
themselves can never be fully assessed unless we study them
in their continuing interaction with interests.
 6 Networks in Indian social struaure
With M.N. Srinivas

Much has been written on the concept of social structure since

Radcliffe-Brown first started making systematic use of it more
than thirty years ago. Apart from writing about it in abstract
and general terms, most British anthropologists have used it as
a central concept in presenting their field material.
Evans-Pritchard's use of the concept in The Nuer1 remains
as a model of its kind. There the concept was used to mean
'relations between groups ofpersons within a system of groups',
such groups, further, having 'a high degree of consistency and
constancy'. 2 This way ofviewing social structure has enabled the
fieldworker to present his data in an economical manner. The
concept is also relatively easy to handle. One begins by locating
the enduring groups in a society, then proceeds to define their
boundaries, and finally one specifies their mutual positions, or
their inter-relations in terms of a series of rights, duties and
obligations.3
The approach outlined above has been widely used by social
anthropologists (particularly British anthropologists or those
trained in Britain) in the study of Indian village communities.
The papers by Srinivas and Kathleen Gough in Marriott's col-
lection4 may be taken as illustrations. The village is viewed there
principally in terms of a set of enduring groups and categories
such as castes, sub-castes and economic classes.
A model of society which is conceived in terms of enduring
groups and categories has also to deal with the problem of
interpersonal relations. It has, in addition, to devise ways of
depicting the relations between groups and categories which
form parts of different systems, e.g. between lineages, terri-
torial segments and age sets, or between castes, classes and
power blocks.
 Networks in Indian social stmaure 141
The distinction between a system of groups and a system
of interpersonal relations has been nicely posed by Evans-
Pritchard. This distinction, it appears, is a part of Nuer kinship
terminology. The Nuer used the word buth to refer to relations
between lineages viewed as groups. The word mar is, by con-
trast, used to refer to kinship relations between persons belong-
ing either to the same agnatic lineage or to different ones. 5
It should be recognized in this connection that, when one
talks of the 'relations between groups in a system of groups',
one is representing things at a certain level of abstraction.
Representation of the relations between castes in a system of
castes involves one level of abstraction. On the other hand, a
representation of the relations between the system of castes and
the system of classes involves an abstraction at a higher level.
Evans-Pritchard shows a clear awareness of the problem which
this raises, and its difficulties. ·
Not only can we speak of the relations between territorial groups
as a political system, the relations between lineages as a lineage
system, the relations between age-sets as an age-set system and so
forth, but also in a society there is always some relationship between
these systems in the whole social structure, though it is not easy to
detennine what this relationship is 6
It may be pointed out that the abstract relations between
groups and systems of groups can be better understood by
mapping out the concrete relations between individuals in their
diverse roles. This may be achieved by making a shift from a
study ofgroups within a system ofgroups to a study ofsocial net-
works. What are the concrete relations which an individual has in
his capacity as Brahmin, landowner and panchayat member with
other individuals? The concept of social network paves the way
to an understanding of the linkage existing between different
institutional spheres and between different systems of groups
and categories.

It is necessary to point out that the model of social structure

which bases itself on enduring groups and categories, and their
interrelations, has been developed largely by social anthropolo-
gists engaged in the study of primitive societies. Such a model
142 Society and Politics in India

does, indeed, take one a long way in the description and analysis
of societies which are small, homogeneous and relatively static.
In the study of large, complex and changing societies, however,
this approach is faced with certain limitations.
In a complex society such as India the number of enduring
groups, classes and categories is very large, and they present
a bewildering variety of types. It may be difficult within the
compass of a single study even to enumerate such units, not
to speak of providing a coherent account of their complex
interrelations. Evans-Pritchard has been able to provide a fairly
comprehensive account of Nuer social structure while confining
himself almost wholly to three systems of groups. Clearly, it is
impossible to analyse in such an economical manner the social
structure of even a single district in India.
There is another factor which imposes limitations on the
approach which confines itself to the study of enduring groups
and systems of groups. In traditional India groups such as. vil-
lage communities, sub-castes and lineages had sharply defined
outlines. It was relatively easy to delimit their boundaries. Today
the situation is somewhat different. Boundaries between groups
tend to be blurred or broken down, there is greater circulation
of personnel, and an increasing degree of interpenetration
between different systems of groups, classes and categories. 7
This process makes it increasingly difficult to locate and define
the boundaries of groups, and hence to talk meaningfully of
groups of persons within a system of groups.
In India this partial dissolution of a rigid, segmental and
hierarchical social structure is associated with increasing social
mobility, both horizontal and vertical. It is also associated with
the transition from a status-bound ascriptive social order to
one which gives greater scope to contractual relations based on
personal choice. The allegiance of the individual to his village,
his sub-caste and his lineage has, to some extent, loosened.
Along with this, the individual is being progressively drawn
into networks of interpersonal relations which cut right across
the boundaries ofvillage, sub-caste and lineage. 8
The process outlined above may be illustrated with a concrete
example. Let us consider the case of Sripuram, a multi-caste vil-
lage in Tanjore district which has been exposed to the forces of
 Networks in Indian soda/ stroaure 143
change since the end of the nineteenth century. Sixty years ago
one's social position in Sripuram was defined largely in terms
of one's membership of the village, of a particular territorial
segment9 of it, of a sub-caste, a lineage and a household. Much
of the social life of the villagers could be understood in terms of
the relations between these diverse groups, each of which had
fairly easily determinable boundaries. Today the social contours
ofthe village are becoming blurred, its population has acquired a
shifting character, and lineages and families have become greatly
dispersed. .
Many ofthe former residents ofSripuram have left the village
and gone to Tanjore, Madras, Delhi, Bombay and Calcutta. In
each of these centres they have developed new relations, while
retaining many old ones with people who still live in the village.
Many of those who have left the village continue to influence its
social life in a number ofways. Often they return at harvest time
to receive rents, and renew leases with their tenants. Several of
them send remittances to relatives in the village every month. On
occasions of birth, marriage and death they revisit the village.
The concept of social network makes for an effective repre-
sentation of the links radiating from the village to the outside
world. These links sometimes stretch across wide territorial
gaps, and often they are made up of strands of diverse kinds.
One of the Brahmin landowners of Sripuram wanted to get
a seat for his son in an engineering college at Madras. He
approached an influential non-Brahmin friend at Tanjore who
was also his father's client, the father of the Brahmin landowner
being a lawyer in a nearby town. The non-Brahmin friend,
who is chairman of a transport undertaking at Tanjore, had
influential business associates at Madras. Some ofthese persons
were able to put the landowner from Sripuram into touch with a
member ofthe committee ofthe college to which he was seeking
admission for his son. In the contemporary world of Sripuram
the individual finds it increasingly necessary to become a part of
the kind of network described above. Sixty years ago, when the
social horizons were narrower, this necessity was far less keen.

Having sketched the conditions under which networks emerge

and become increasingly important in social life, let us consider
144 Soci.ety and Politics in India

the distinctive features ofnetworks as opposed to groups, classes

and categories. The distinction between groups and networks
is primarily one of boundaries. A group is a bounded unit. A
network, on the other hand, ramifies in every direction, and, for
all practical purposes, stretches out indefinitely. Further, a group
such as a lineage or a sub-caste has an 'objective' existence: its
boundaries are the same for the 'insider' as well as the 'outsider'.
The character of a network, on the other hand, varies from one
individual to another.IO
This distinction between groups and networks, which was first
elaborated by Barnes, 11 should not, however, be pressed too far.
Networks can be either close-knit or loose-knit. In other words,
the chain of relations emanating from a person may either lead
back to him, or it may not. In traditional India, particularly in
the South, the network of kinship and affinal relations was a
close-knit one. In fact, this network had, inevitably, to stop
short at the boundary of the sub-caste, and therefore to form
a closed circuit. Mrs Karve 12 has shown how, in many cases,
an entire sub-caste can be placed on a single genealogy. Here
we have an instance where the kinship network is, in reality,
coterminous with a bounded group, namely, the sub-caste.
Thus, in the limiting case, a close-knit network becomes a
group (or a category).
A social network can be viewed as a set of concrete
interpersonal relations linking the individual to other individuals
who are members of diverse systems of enduring groups and
categories. Here we represent the network from the viewpoint of
the actor, and there are as many networks as there are actors in a
social system. Before we pass on to a consideration ofthe network
from the viewpoint of the 'observer' (i.e. the anthropologist), let
us examine a little further the implications of the subjective
definition of a network.
The anthropologist as fieldworker begins to learn about the
way in which a society works precisely by following concrete
networks of interpersonal relations with the individual actor as
his point of departure. He sees how individuals cut through
the boundaries of household, lineage, sub-caste, village, district,
party and class so as to form interpersonal relations in the
pursuit of interests of diverse kinds. He learns to differentiate
 Networks in Indian social struaure 145
one individual from another in terms of the range and variety
of interpersonal relations, and tries to relate these differences to
factors such as generation, education, occupation and so on.
A network, even when viewed from the standpoint of a single
individual, has a dynamic character. New relations are forged,
and old ones are discarded or modified. This is particularly
true of rapidly changing societies in which individual choice
plays an important role. However, extensions of the individual's
social network may also be studied in relatively static societies.
Thus, in the field ofkinship we may observe how the individual's
network of effective interpersonal relations is extended as he
passes from birth through initiation and marriage to death.
Although the anthropologist may begin by mapping out the
concrete networks of interpersonal relations ofindividual actors,
this mapping, in itself, does not fully meet the needs of his
analysis. At best it can provide him with a broad idea of the
linkage between the groups and systems of groups in a society.
For a deeper understanding it is necessary not only to chart the
concrete networks of different individuals, but to relate these
different- networks to one another, to draw up, so to say, a
master chart, in a coherent and systematic manner. This involves
abstraction and synthesis.
It is easy to see what one means by an individual's network
of interpersonal relations, for this has a concrete character.
But can one speak, for instance, of the social network of
a village? What would such .a statement mean? The village
comprises a diversity of individuals, each with his own network
of concrete interpersonal relations. These partly overlap, and
partly cut across, and are, in fact, related to one another in
very complex ways. In some spheres, the separation of one
individual's network from that of another is quite clear. Thus,
. the network ofkinship and affinal relations of a Brahmin will not
at any point meet that of a non-Brahmin, even though they be
of the same village. In the economic sphere, on the other hand,
these networks are likely to meet at a number of points. Can one
distinguish in a systematic way between different institutional
areas in which networks are relatively close-knit or loose-knit?
It seems evident that the kinship network in India is relatively
close-knit as compared, let us say, to the economic or political
146 Society and Politics in India
network. It may be that the same forces which lead to the
extension and loosening of the economic and political networks
also lead to the shrinkage and tightening of kinship networks in
contemporary India. Territorial dispersal and mobility lead to
the extension of economic and political ties; they often also lead
to a shrinkage of the network of ejfeaive kinship relations based
upon reciprocal obligations.13
We have now been led to a point at which it is necessary to talk
in somewhat more abstract terms. From viewing the concrete
networks of interpersonal relations of a number of individual
actors we have been led to talk about networks pertaining to
different institutional areas. We can now speak about economic
networks, political networks, ritual networks, and so on. It is
evident that when we speak, say, of an economic network, we
are making an abstraction. A concrete network of interpersonal
relations cannot be wholly economic in its constitution, except
in the limiting case. Generally such relations have economic
components which have to be abstracted from their concrete
matrix, and then put together.
The economic system may be viewed as a network ofrelations
regulating the flow of goods and·services. The political system
may, likewise, be viewed as a network of relations regulating the
flow of command and decision. It must be pointed out that the
links in networks ofthis kind are unitary in character, as opposed
to concrete networks of interpersonal relations where the links
are usually composite or multi-bonded.
Economic, political and ritual networks of the kind described
above would correspond to what Marion Levy14 characterizes
as 'analytic', as opposed to 'concrete' structures. Thus, a
network of economic relations provides an understanding of
the organization of production in a society, and a network of
political relations provides an understanding of the distribution
of power. Such networks in a complex society cut across
the boundaries of communities and corporate groups and, in
fact, serve to articulate them to wider social systems. And,
once we shift from the individual actor and his network of
concrete interpersonal relations to the productive system and its
corresponding network, we move from the 'subjective' network
of the actor to the 'objective' one of the observer.
 Networks in Indian soci.al struaure 147

We have seen earlier that a crucial distinction in the study of net-

works is that between close-knit and loose-knit networks. It may
be urged that one way of understanding social change in India
would be to analyse the manner in which close-knit networks are
being transformed into loose-knit ones. Traditionally the villager
lived in a narrow world where the ties oflocality, caste, kinship
and hereditary service led back and forth between the same sets
ofpersons. Relations were multiplex in character, and the circuit
of relations had a tendency to become closed.
The situation is changing in contemporary India. New inter-
ests tend to create relationships which cut right across. the
boundaries of the old established groups. Increased mobility
has led to the physical dispersal of castes, lineages and families.
The individual cannot any longer afford to confine his relations
within a village, a caste or a kin group. He has to develop
relations with people who are spread far and wide and who have
diverse social, economic and political positions and interests.
The network of social relations emanating from the individual
does not.as easily lead back to him. The closed circuit tends to
become more and more open.
The phenomenon sketched above can perhaps be best illus-
trated from the field of politics. To take once again the instance
of Sripuram, one can get only an imperfect understanding of its
political life by confining oneselfto groups such as the panchayat,
the party or the sub-caste. What appear to be of greater
importance are the networks which link the village leaders to
politicians and influential people outside, and which cut right
across the boundaries of parties, panchayats and sub-castes.
Bailey15 has spoken of 'brokerage networks' in the context of
Orissa. Such networks are of great importance to the working
of the political process throughout the country. In and around
Sripuram they link the village leader with the district leader,
the patron with the client, the MLA with the 'vote-bank', the
party boss with the financier, and the panchayat president with
the contractor.
It is evident that some of the most radical changes taking
place in Indian society today are in the field ofpolitics. A rigidly
hierarchical and segmental social structure is being transformed
148 Society and Politics in India

into one which seeks to bring about political articulation between

people at all levels of society. The peasant in a Tanjore village
is linked directly with the Member of Parliament at Delhi. To
what extent can such a linkage be effective or successful? It is in
this context that the problem ofcommunication acquires central
importance to both the politician and the political sociologist in
India. How does the system of political communication actually
operate? How does it affect the existing systems of groups and
categories in Indian society, and how is it, in tum, affected by
them? A study of the concrete links between villagers, local
leaders, party bosses, MLAs, and MPs is indispensable to an
understanding of the channels along which communications
flow, and the barriers at which they are blocked.
In a recent paper Ithiel Pool1 6 has emphasized-the importance
of informal social channels of communication in providing nec-
essary support to the mass media in traditional societies. Political
events in the state capital are interpreted and transmitted along
social networks of various kinds whose nature requires to be
investigated. There is little doubt that such networks today
link individuals not simply on the basis of caste or occupation
or locality, but on the basis of a complex combination of these
and numerous other factors.
India has embarked on a course of planned social change and
economic development. This involves, among other things, the
transmission of certain key ideas, principles and values from
the highest to the lowest levels of society. What are the social
networks along which such ideas are transmitted? What kind
of refraction do they undergo as they pass from one level to
another? How is this refraction conditioned by the nature of the
social network along which the ideas flow?
The entire process of political mobilization in a country such
as India highlights the importance of networks of interpersonal
relations. How does the politician reach down to the voter, and
how does the latter, in tum, articulate with the former? In a
country where literacy is low and where the mass media are new
and limited in scope, networks of interpersonal relations are of
primary importance to the mobilization process.
The process of political modernization has many immediate
and far-reaching consequences for the structure of traditional
 Networks in Indian social struaure 149
society. It breaks down the barriers between groups which had
crystallized over centuries. It gives a new amplitude to individual
choice in severing old relations and forging new ones. All this
leads to the development of networks on the basis of new
interests which criss-cross the entire social fabric.

In conclusion, we have to consider briefly the existence of social

networks and the part played by them in ·traditional Indian
society. It would, of course, be far from correct to say that
networks had no existence in traditional society, or that the social
life of the individual was completely contained within systems of
enduring groups. Even in the past the village was never entirely
a closed or self-sufficient unit. Links of various kinds radiated
from it, connecting its individual members to other individuals
outside. Such links, however, played a far less important part
in the past than they do today. Even the extra-village ties of
the individual often articulated him with other groups such
as lineages or sub-castes which were themselves closed in
character.
A village community which forms part of a wider civilization
can never be entirely a closed unit. In fact, articulation of a
particular kind is a basic characteristic of a civilization as
distinguished from a primitive society. The manner in which
little and great traditions are articulated through social networks
in a primary civilization has been discussed by Redfield, Singer
and others. 17 But it cannot be denied that the nature of articu-
lation in a relatively static and compartmentalized social order is
different from one which is fluid and changing in character.
 7 The future ofthe Backward
Classes:the competing demands
ofstatus and power
The Backward Classes 1 constitute an important section of
Indian society. In all they account for more than 30% of the
total population of the country. Their condition is intimately
linked with many of the basic features of Indian social structure
and, as such, is likely to be affected by any significant change
in these. Clearly, an assessment of the future of the Backward
Classes cannot be made in isolation from social and political
forces which operate through the entire range oflndian society.
Studies in social and political projection are faced with
problems of a special nature. Predictions in the social sciences
cannot be made with the same certainty or the same degree
of probability as in the natural sciences. This is so even when
the social scientist has at his disposal the entire body of data
considered necessary for his analysis. For here the validity of
the analysis depends not only on the quality of data but also on
the values which direct one's research.
Sociologists such as Mannheim2 and Myrdal3 have shown
how different persons may arrive at different (though not nec-
essarily contradictory) conclusions even when they analyse the
same body of facts. The values and interests of the sociologist
give a direction to this research and no method has as yet been
devised which can eliminate their influence altogether. Values
play a part not only in the formulation of the problems to be
investigated but also in the selection and arrangement of data.
The problem of the Backward Classes may be posed in a vari-
ety of ways. Is it essentially a political problem or is the problem
basically educational and cultural? Should the Backward Classes
be treated as a homogeneous, undifferentiated unit or should
one view separately the problems of the Scheduled Tribes, the
 The future ofthe Backward Classes 151
Scheduled Castes and the Other Backward Classes? Finally,
does one accept the prevalent basis of classifying the Indian
population into 'backward' and 'forward' as a rational one?
One is likely to have specific views on all these questions, not
only as a sociologist but also as a member of Indian society.
There are special problems inherent in the study of one's own
society, particularly when such a study deals with issues which
are of general and fundamental importan~e. The outsider has
a certain advantage. He is able to approach the problem with
a more open mind, relatively free of preconceptions. This, of
course, is not to deny that an inside view of social problems has
advantages of its own.
If value preferences tend to distort interpretations of the past,
their role in projections is likely to be even more decisive. An
effort at projection faces a bigger challenge in the matter of
valuation. For here one has to evaluate the significance not
only of elements and forces which have established themselves
but also of those which are in a process of emergence. One has
to uncover potentialities inherent in factors whose contemporary
significance may not appear very great. Here the judgement of
the social scientist is put to a kind of test which he is normally
inclined to avoid.
It cannot be too strongly emphasized that objectivity in social
research is not merely a question of good intention or even
of professional competence. It is also a function of the social
position of the person who conducts the research. As such, a
full and objective picture of social reality can hardly be drawn
with a single stroke of the brush. Rather, the area of objectivity
can be expanded only by slow degrees as different people,
occupying different social positions and representing divergent
interests, examine the same problem from a variety of angles.
It would be well to bear these limitations in mind while making
an assessment of the analysis which follows.

Who are the Backward Classes? In India, the Backward Classes

constitute a category of people who are for the 'most part
officially listed and given special recognition in a variety of
contexts. In every complex society, of course, there are indi-
viduals who may be considered as economically or educationally
152 Sodety and Politics in India

backward; generally, such people have also a low social status.

However, backwardness as understood in the Indian context
has a number of distinctive features. Firstly, it is viewed as an
attribute not of individuals but of certain clearly defined social
segments in which· membership is generally acquired by birth;
thus, the Backward Classes may in theory include individuals
who are highly advanced both educationally and economically.
Secondly, membership of the Backward Classes entitles one to
certain advantages and concessions specifically conferred by the
Government.
There is a good deal of overlap in reality between the
Backward Classes and certain economic categories such as
the agricultural labourers. But it would be a mistake to view
the problems ofthe two as identical. The Backward Classes have
as a set of communities certain distinctive problems which derive
from the status ascribed to them in traditional Indian society. It is
these distinctive problems which will engage our attention in the
analysis which follows. The very nature of the Backward Classes
and most of their special problems can be understood only in
terms ofthe basic character oflndian society, namely, its division
into a multitude of closed status groups of unequal rank, each
associated with a variety of privileges and disabilities supported
by traditional sanctions.
It may on a first examination appear tempting to restructure
the definition of the Backward Classes and to .view their
problems principally in economic terms. This would, however,
divert attention from the specific nature of backwardness in
Indian society. In the first place, the Backward Classes are by
no means homogeneous economically. Secondly, the Backward
Classes as officially viewed are a part and parcel of modem
Indian social and political reality. They are a product of forces
which are in many ways unique and distinctive oflndian history.
Their identity does not derive solely, or even primarily, from a
common economic experience but from the very nature of the
traditional system of stratification to which economic, political
and ritual factors have contributed in various ways.
The term 'Backward Classes', which has been given currency
through official publications, is not altogether a happy one. The
word 'class' suggests not only an economic category but also
 The future ofthe Backward Classes 153
one which is relatively open. In reality the Backward Classes
are not classes at all but an aggregate of closed status groups.
One's economic position is not a determining factor in one's
membership of the Backward Classes; rather, membership is
determined generally by birth.
The Backward Classes together account for more than 30
per cent of the total population of India. They are not a
homogeneous category but consist of three broad divisions,
each having its own distinctive background and, to some extent,
its own problems of transformation. The three broad divisions
are the Sc;:heduled Tribes, the Scheduled Castes and the Other
Backward Classes. The Other Backward Classes are the least
homogeneous and the most loosely defined of the three sub-
divisions. Their problems also are in many ways different from
those of the first two and it may be misleading to consider the
three together beyond a certain point.
-An element of confusion pervades the discussion on Back-
ward Classes due to a certain misapplication of terms. In this
essay the term 'Backward Classes' will be used to refer to the
most inclusive category and the term 'Other Backward Classes'
will be reserved for that section of the broader category which
remains by exclusion of the Scheduled Tribes and Scheduled
Castes. It seems to be a common practice, not only among
scholars but also among politicians to use the term Backward
Classes as being synonymous sometimes with the Scheduled
Tribes and Scheduled Castes and at other times with a some-
what broader category. This ambiguity in terminology leads to
certain important issues and problems being obscured.4
The position of the Scheduled Tribes and Scheduled Castes
is defined in a more or less specific manner in the Indian
Constitution. Lists of these communities are drawn up by the
Central Government and can be revised only by Presidential
authority. There is, in addition, a Commissioner for Scheduled
Castes and Scheduled Tribes to look into the affairs of these
communities on a continuing basis.
The Other Backward Classes, by contrast, are a more nebu-
lous category. They are mentioned in the Constitution in only
the most general terms. There is no all-India list for the Other
Backward Classes. They are not separately enumerated in the
154 Society and Politics in India
Census and, in fact, one has to work with only a rough estimate
of their population. Their position was sought to be defined
in more specific terms by the Backward Classes Commission
set up in 1953 under the chairmanship of Kaka Kalelkar. The
Commission, which was not a standing body, could not in fact
come to any tangible or agreed conclusions.
The Scheduled Tribes or Adivasis numbered 29.9 million
in 1961 out of a total population of 439.1 million. They are
popularly believed to constitute the aboriginal element in Indian
society. They are generally concentrated in the hill and forest
areas· and until recently the political system of the different
tribes enjoyed a certain-measure of autonomy. Today, however,
it is difficult to define the tribal peoples of India in terms of
any single set of formal criteria although attempts have been
made at such a definition. 5 Elements which would have to
be taken into account in such a definition are the ecological
isolation of the tribal people, the relative autonomy of their
political and cultural systems and the antiquity of association
with their present habitat.
The difficulty of applying a set of formal criteria in defining
the Adivasis arises from the fact that tribes in India are (and
have been for some time) tribes in transition. The political
boundaries of most tribal systems had collapsed well before the
beginning of the present century. A certain amount of cultural
interchange between the tribal people and the outside world has
existed for centuries. Sections of the tribal population tend to
get absorbed into Hindu society by a process which has been
fairly widespread. In fact, it is often very difficult to say of
a particular social unit whether it is a tribe or a caste. The
complexity of the phenomenon defeats any attempt to solve
such problems by means of precisely formulated definitions.
Each case has to be examined on its own merits and in
relation to the specific social and historical factors prevalent
in an area before one can decide whether or not a particular
unit qualifies as a tribe. Lists of the Scheduled Tribes have,
in fact, been drawn up after careful consideration of a variety
of individual cases.
The tribal population is concentrated in certain geographi-
cal areas.
 The future ofthe Backward Classes 155

Speaking very broadly, they may be divided into three groups

according to their distribution, namely, the tribes living in the
Northern and North-Eastern zone in the mountain valleys and
Eastern Frontiers oflndia. There is a second group which occupies
the Central belt of the older hills and plateaus along the dividing
line between Peninsular India and the Indo-Gangetic Plains. In
addition, there are tribes scattered over the extreme corners of
South-Western India, in the hills and the converging lines of
the Ghats. 6
These tribes speak a large variety ofdialects and there are enor-
mous variations in their habits, customs and arts. Historically
one of the principal features of the tribal population has been
its ecological and social isolation. For centuries the tribal people
have been confined to hills and forests and this isolation has
left a definite impress on their social systems. It has also given
them, in spite of wide cultural variations, a common destiny in
Indian society. For, one ofthe crucial problems faced by all tribal
communities in India is the prpblem ofintegration into the wider
social, economic and political system.
The Scheduled Castes or Harijans have not had a history of
isolation comparable to that of the Scheduled Tribes. They have
been segregated rather than isolated. Thus, whereas the tribal
people are concentrated in blocks, the Harijans are scattered
through every state and practically every district. The different
distributional characteristics of the Scheduled Tribes and the
Scheduled Castes lead to certain differences in their problems
of transformation. It is easier to.implement special ,programmes
of development for the Scheduled Tribes who live in compact ·•
blocks than for the Scheduled Castes who are geographically
scattered. On the other hand, the concentration of the tribal
population in particular areas provides better scope for the
development of separatist political movements among them.
The Scheduled Castes numbered 64 .5 million at the 1961
Census, thus giving a figure which is more than twice that of
the Scheduled Tribes. They are concentrated in rural areas and
are found commonly in multi-caste villages. Although they live
in close interdependence with the higher castes, many areas of
social life have been (and continue to be) inaccessible to them.
They generally live segregated in their own settlements which
156 Sodety and Politics in India
are often a little distance away from the residential quarters of
the upper castes. They have been debarred by tradition from full
participation in many of the collective activities ofthe village and
some of these restrictions are still operative. Their economic,
social and ritual status continues to be depressed although there
are certain indications of change which will be discussed later.
The Scheduled Castes have been known in popular parlance
as the untouchables. Their social condition has been governed in
important ways by the Hindu concept of pollution. Although the
practice of untouchability has been made an offence, the stigma
of pollution has not by any means been entirely removed.
The Other Backward Classes constitute a congeries of com-
munities of rather uncertain status. Lists had earlier been
prepared by the Ministry of Education and by the State Gov-
ernments. The Backward Classes Commission under Kaka
Kalelkar reported a good deal of ambiguity in these lists. 'But
there was no authoritative list of Other Backward Classes. The
Census had assumed one list; the Ministry of Education had
prepared another; and it was left to the present Commission
to recommend an authoritative list.' 7 The recommendations of
the Commission were not, in fact, accepted as authoritative.
The State Governments have in general been allowed to use
their own criteria in drawing up lists of the Other Backward
Classes. These, in turn, are not always held to be binding by
the Supreme Court which, in the case of M.R. Balaji and Others
versus the State of Mysore, decided against the recommendation
of the Nagana Gowda Commission. ·
The Central Government have particularly since 1961, been
pressing for the adoption of economic criteria in defining the
Other Backward Classes but there has been resistance to this
from a number of State Governments. Some of the castes
included in the earlier lists of the Other Backward Classes
are fairly powerful in state politics and have therefore been
in a position to exert pressure on the State Governments to
have the old criteria retained. The Lingayats of Mysore and the
Ezhavas of Kerala provide good instances ofpowerful dominant
castes which have exerted pressure on their respective State
Governments for the retention of caste as the basis for defining
backwardness.
 The future ofthe Backward Classes 157
Although the Central Government has not insisted on the old
lists being abandoned altogether, economic and other pressures
have been exerted on the State Governments for the adoption
of economic criteria. Since 1963 they have made the use of
Central Government funds for the award of scholarships to
members of the Other Backward Classes conditional upon
the use of economic criteria for defining such classes. By
now the majority of the states have adopted this criterion
for defining backwardness for the award of scholarships to
the Backward Classes other than the Scheduled Castes and
Scheduled Tribes. The Central Government have also since
1962 decided to use the term Backward Classes to refer only to
the Scheduled Tribes and the Scheduled Castes. 7 However, the
broader usage is retained in this paper since the Other Backward
Classes continue to retain a certain amount ofsignificance at the
level of the State Government.
It is difficult to give exact figures for the Other Backward
Classes because they are not enumerated separately in the
Census and because such lists as do exist are subject to
revision from time to time. Some estimate can, however, be
made of their strength by projection of the caste returns in
the 1931 Census. Speaking in very broad terms, then, the
Other Backward Classes constitute about one-seventh of the
total population of the country, being thus approximately equal
in numerical strength to the Scheduled Castes.
The core of the Other Backward Classes consists of peasant
castes of various descriptions. The position occupied by these
castes in the wider society is rather different from that of the
Harijans. Frequently they occupy a low position in the varna
hierarchy and they have in general been devoid of traditions
of literacy. Further, since they have also lagged behind in the
pursuit of Western education, they are often poorly represented
in Government jobs and white-collar occupations in general.
In spite of this, such castes sometimes occupy a dominant
position in the economic and political systems of the village.
Not infrequently they are small landowners. When they are also
numerically preponderant, their control over a village, a group of
villages or even a district may be decisive.
Dominant castes of this kind have developed a vested interest
158 Soci.ety and Politics in India
in remaining backward. It enables them to enjoy a number of
benefits in education and employment. What is more, they
sometimes have enough political power to exert pressure on the
State Government to have their names included or retained in
the list of Backward Classes. The case of the powerful Lingayat
caste of Mysore, which had first been excluded from the list
recommended by the Nagana Gowda Commission and had later
to be accommodated, has by now become well known.
This, then, is the broad social background against which we
have to consider the issues which confront the Backward Classes
in their efforts at transformation. The issues, it will be evident,
are many and diverse. In many ways they are different for the
three broad divisions, the Scheduled Tribes, the Scheduled
Castes and the Other Backward Classes.
The tribal people have had a history ofisolation which gives in
many ways a unique character to their problems. They are faced
with the task of achieving integration without doing violence
to their rich cultural and artistic heritage. In the case of the
Scheduled Castes an important problem is to break through the
barrier of untouchability, not simply in its formal legal sense but
in its widest social application. Both have to face the challenges
ofpoverty, illiteracy and social prejudice, and this is true to some
extent of certain sections of the Other Backward Classes too.
But the latter also include a number of dominant castes who
have 'arrived' so to say, at least politically, and have now to think
ofconsolidating their power against politically weaker sections of
society which may appear to threaten their dominance.
In the discussion which follows we shall try as far as poss-
ible to consider separately some of the distinctive features of
each of the three sections of the Backward Classes keeping
in mind the more general problem of backwardness. This
should not, however, be taken to mean that each section
constitutes a homogeneous or undifferentiated unit. In fact,
there are numerous social and economic differences within
each but a detailed consideration of these would require a
far more exhaustive study than the present one. Here we
can provide only a few illustrations of these differences. The
condition of tribal people in non tribal areas, for instance, is
likely to differ greatly from their condition in an area which
 The future ofthe Backward Classes 159
has a predominantly tribal population. The Scheduled Castes
in South India occupy a somewhat different position from
.their counterparts in North India; an understanding of these
differences would require analysis of the differences between
the social systems of North and South India, including the
differential importance of the idea of pollution in the two
systems. Finally, everywhere there are great disparities ofwealth
and power within the all-too-loosely defined category of the
Other Backward Classes.

That Indian society is passing through a phase of active social

change will not be disputed by many. In a sense the impact of
change appears more striking among the lower strata of society
partly because these strata had remained relatively immobile in
the past. Changes in the status of the Backward Classes, which
may appear limited in absolute terms, tend to acquire a different
significance when viewed against the background oftraditional
society.
It is well to remember that the currents of change do not all
run in the same direction. Sometimes they run counter to one
another. There are on the one hand certain factors which tend
to blur the outlines of the traditional structure and to bridge the
gaps between the Backward Classes and the advanced sections
ofsociety. The modem educational system may be viewed as one
such factor. On the other hand, the frequent use made in politics
ofthe loyalties ofcaste and tribe tends in some measure to freeze
the traditional structure.
In spite of the apparent confusion of currents and counter-
currents, the scales seem to be decisively weighted in favour
of a lowering of many of the barriers of traditional society. In
the traditional system, society was divided and subdivided into
a large number of segments which were kept rigidly separated
from one another. New networks of interpersonal relations are
now being created in every field and these tend increasingly to
cut across the boundaries of the old, established groups. New
areas ofsocial life are being opened up in which individuals from
widely different backgrounds are able to come together on the
basis of achievement, interest and personal choice. The change
from a segmental and particularistic social order to a more fluid
160 Soci.ety and Politics in India

and universalistic one is bound.in the long run to change the very
character of the Backward Classes. But paradoxically enough, it
seems that this change cannot be effected without widespread
use being made of the particularistic loyalties of tribe, caste and
community.
Of the many changes which are taking place among the
Backward Classes, we shall consider two in particular: changes
in their style of life and in their relation to the distribution
of power.
When members of a lower caste change their style of life
and move up in the hierarchy of prestige and social esteem,
we speak of this as mobility along the axis of status. When on
the other hand, they advance themselves by securing material
benefits through office in an organ of government or a party
or a pressure group, we speak of this as mobility along the axis
of power. Our analysis of the future of the Backward Classes
will centre around the chances of their movement along these
two axes.
British rule provided the Backward Classes with increasing
opportunities to imitate with effect the styles of life of the upper
castes. These opportunities were extensively used by the Other
Backward Classes and also the Scheduled Castes and Scheduled
Tribes who set about 'Sanskritizing' themselves with varying
degrees of success. Since independence, however, the emphasis
seems to be shifting from Sanskritization to competition for
positions of office and power. It appears that at many points
the demands of prestige and power came into conflict and this
conflict is likely to make itself felt more acutely among the
Scheduled Tribes and Scheduled Castes. Present indications
seem to suggest that in the years to come the latter will tum
increasingly to politics as an avenue of social mobility.

We shall first consider the changes which the different sections

of the Backward Classes have been undergoing in their styles
of life. It may be useful to fix our point of departure in the
past in order to be able to see the future from a proper point
of vantage. It must not be thought, however, that changes in
the styles of life of the Backward Classes are likely to follow a
regular path. In fact, one of the points of the present argument
 The future ofthe Backward Classes 161
is that political forces of a certain kind tend to arrest some ofthe
changes which had been taking place more or less continuously
over the last several decades.
Traditional Indian society had a structure which was highly
segmental and hierarchical in character. The segments within
it were separated from one another by clear-cut boundaries
and marriage, commensality and many other forms of social
interaction generally stopped short at these boundaries. The
social separation of the different segments was bolstered by
the fact that Hinduism allowed within its fold the practice of
a wide variety of styles of life. Each group of castes, each caste
and sometimes even each sub-caste was allowed to cultivate its
distinctive styles of life in the matter of diet, dress, worship,
marriage, etc. Hinduism tolerated a plurality of cultures but the
price of this was the maintenance of a certain structural distance
between people traditionally associated with divergent styles of
life, e.g., between vegetarians and non-vegetarians, or between
people practising different crafts or worshipping different gods.
Social separation between different segments, each pursuing
its own style of life, could be kept intact so long as the world
was fairly static or the pace of change not very rapid. In the
traditional system, mobility - whether vertical or horizontal -
was slow and limited. The expansion of transport and communi-
cation, the spread ofeducation and new economic opportunities,
and an increasing degree of political articulation are bringing
about fundamental changes in the traditional structure. The
system now tends to become relatively more open, allowing for
greater mobility and greater variety in the combination of class,
status and power positions.
Culturally the distance was greatest between the Adivasis and
the Harijans on the one hand, and the advanced sections of
society on the other. The tribal people, who were ecologically
isolated, had developed their own traditions, habits and ways of
life. The Harijans also lived in a cultural world of their own, shut
out in many ways from the world of the 'twice born'. In the case
of the Harijans there were often specific sanctions against the
adoption of the styles of life of the upper castes. Even until
recently the 'exterior' castes in South India were disallowed
the use of sandals, umbrellas and silken cloth; they were not
162 Society and Politics in India
allowed to live in brick and tile houses; and their women could
not wear upper garments. These disabilities were often enforced
by powerful traditional sanctions.
Thus, the Adivasis because of their physical isolation and
the Harijans for other reasons were unable to fully identify
themselves with the higher strata of society or to use many of
their distinctive symbols of status. For all this, social forces have
been at work, leading to a transmission of cultural elements from
the more advanced sections of society to the more backward.
Two of the most important of such forces are Sanskritization
and Westernization. 9
Sanskritization was an important feature of traditional Indian
society where it appears to have been the principal idiom of
social mobility. Its role in contemporary Indian society was
first analysed in detail by Srinivas in his study of the Coorgs
of South India. 10 Sanskritization can be defined as a process
by which a caste or a group of people moves up the social
hierarchy by adopting styles of life associated by tradition with
the upper castes. The Sanskritic model should not be viewed as
an undifferentiated one; in addition to the Brahminical model,
people seem at various places or various times to have made use
of a Kshatriya model and perhaps even a Vaishya model. 11 But
what is of particular importance in this context is that the idiom
of Sanskritization is essentially traditional in nature.
In the past, the process of Sanskritization was slow and
gradual and it offered very limited possibilities to the lowest
sections of society. Social horizons were narrow, the economy
was relatively static and population movements were limited.
This made it difficult for a lower caste to acquire quickly
economic and political power or, having once acquired it, to shed
its traditional marks ofinferiority. There were, in addition, legal
and ritual sanctions which acted against a too radical change in
life-styles. These sanctions operated with particular force upon
the Harijans who were able to cross the barrier of untouchability
rarely, if at all.
British rule released the Backward Classes (including the
Harijans) from the grip ofmany ofthe traditional sanctions. The
new courts of law refused to recognize the rights of the upper
castes to the exclusive use of particular symbols of status. The
 The future ofthe Backward Classes 163
avenues of Sanskritization were thrown open to ever-increasing
sections of society. The first to seize the new opportunities were
those whose social position had been low in traditional society
but above the line ofuntouchability. These included many ofthe
castes hitherto classified under the Other Backward Classes.
Castes which had been fairly low in the traditional hierarchy
changed their diet, their social customs and sometimes even
their gods in favour of those of the upper castes. The decennial
Census provided them with opportunities to replace their tra-
ditional names with new and more high-sounding ones. Caste
associations were formed throughout the country and these not
only put forward claims to higher social status but also urged
their members to abandon many of the practices considered
degrading by the upper castes. Thus, Sanskritization served to
lower the barriers between sections of society which had at one
time been clearly separated. It is paradoxical that this process,
which in a way represents the distinctive idiom of traditional
society, had to await the coming of the modem age before it
could acquire its full momentum.
The increased pace of Sanskritization has been dependent
upon a number of forces, some of which are likely to extend
their influence in the years to come. These include improve-
ments in transport and communication, greater mobility in
non-traditional sectors of the economy, the spread of literacy
and education among the lower strata and (paradoxically) the
institution of a secular legal order. Each of these forces, how-
ever, has other consequences which are likely to alter the very
meaning of Sanskritization and in the long run to undermine its
significance.
The revolution in transport and communication has thrown
open pilgrim centres to people from far and near. Every year
more people visit such important centres as Mathura, Kashi,
Gaya, Tirupathi and Rameshwaram. In the past a visit to a
distant centre of pilgrimage was not only replete with hazards but
also a costly affair. Now even the moderately poor find it within
their means to undertake journeys by bus or train to places which
in the past·would have appeared remote. It seems that these
facilities will be used to greatest advantage by the better-off
sections of the Other Backward Classes in the years to come.
164 Sodety and Politics in India
The mass media have been harnessed for the diffusion
of Sanskritic values and ideas. Mythological films and radio
broadcasts of devotional programmes are becoming increas-
ingly popular in the villages. This trend in the direction of
Sanskritization is likely to persist in the rural areas, particularly
among castes which are above the line ofuntouchability. Among
the top castes there seems to be a trend away from popular
Sanskritic culture and towards new values and symbols of
status. The Other Backward Classes are still too much within
the grip of the traditional system to be able to follow the
top castes in this regard. The Harijans and the Adivasis are,
in turn, different from the Other Backward Classes; their
commitment to the traditional system was never so intense as
to prevent them altogether from seeking alternative avenues of
mobility.
The Other Backward Classes include a number of castes
which enjoy a certain measure of economic and political domi-
nance. A good example of these is provided by the Okkaligas of
Mysore who have been studied by Srinivas.1 2 Such castes or
sections of them are often the ones to benefit most from the
new economic opportunities. And they are also the ones most
likely to Sanskritize their style of life in order to set the seal of
social acceptance on their material success.
Sanskritization affects the culture of castes in the lower and
middle regions of the hierarchy in a variety of ways. To begin
with an example which may appear trivial, but is nevertheless
of considerable significance, one may consider changes in
personal names among them. These changes appear particularly
striking in South India where sharp linguistic differences exist
between typically Brahmin names and names common among
the non-Brahmin peasantry. Among the latter the most common
personal names have been typically non-Sanskritic, such as
Pazhani, Thangavelu, Nadugowda or Puttappa. Today one
encounters in increasing number such typically Sanskritic names
as Narasimhan, Parthasarathy, and Srinivas. This change, trivial
in itself, is symbolic of the penetration among the Backward
Classes of a culture and a style of life which had virtually been
a monopoly of a few upper castes.
Other, more fundamental changes are also taking place.
 The future ofthe Backward Classes 165
There are changes in occupation, diet and social practices
of various kinds. Occupations considered degrading in the
Sanskritic scale of values, such as distilling or oil-pressing
or tanning, are often forsaken and there is a tendency even
to deny any past association with them. Items of food such as
pork and the drinking of alcoholic beverages are often given
up. There is a pervasive tendency to abandon widow remarriage
and to replace bride-price by dowry. Finally, castes which had
done without Brahmin priests in the past now try to secure
their services on occasions of birth, marriage and death. The
pace of Sanskritization has been heightened not only by the
activities of caste associations but also of such organizations
as the Arya Samaj and the Sanatan Dharma Samaj. The latter
act as important agencies for the diffusion of Sanskritic styles
of life among large sections of people including the Backward
Classes and especially in the rural areas. The Arya Samaj in
particular combines Sanskritization with certain universalistic
principles which create a special appeal among the hitherto
underprivileged strata of society. ·
Sanskritization, which was the principal idiom of social
mobility in the past, appears to have special significance today
for castes in the middle and the lower-middle regions of the
hierarchy. This is partly because the top castes are reaching
forward to new social values and to new symbols of status.
Thus, in Sanskritizing their life-styles, the Backward Classes
are emulating models which certain sections of the traditional
elite are already trying to put behind. And, some of the forces
which impel the emerging elite to attach themselves to new
models operate also among the Backward Classes. One of
the most important of these is what Srinivas refers to as
Westernization.13 ·
Sanskritization can be seen not only as an idiom of mobility
but also as an important source of continuity with the past.
Its symbols and values are essentially those of the traditional
order. At a time when a modernist elite is trying to push
the country towards a secular and Westernized social order,
it is not unlikely that those who had in ·the past occupied
a fairly low social position may set themselves up as the
bastions of traditional values. Unable to cope with the process
166 Society and Politics in India
of Westernization or the pace of rapid social change, such
people may well throw in their lot with traditionalist or even
revivalist movements and parties. It may appear ironical that the
defence of the traditional order should become the burden of
the people who had in the past been denied a position ofhonour
within it.
The Sanskritization of the Backward Classes provides a
stabilizing influence in Indian society. The Western educated
elite in India is often impatient with the slow rate of change in
the country. But in order to bring about change effectively it
will have to make many compromises with some of the hitherto
backward communities which are beginning to develop a new
sense of commitment to the values of traditional society.
It would, of course, be unreal. to suggest that every section
of the Backward Classes is developing a stake in the traditional
order. What has been said above applies principally to the upper
strata among them and in particular to dominant peasant castes
such as the Okkaligas or the Ahirs. The lowest sections of the
Backward Classes are often under the influence of forces of
a very different kind. It is well to remember that there are
sharp cleavages within the Backward Classes, some of which
may have greater significance than the ones which separate
the dominant castes among them from the advanced sections
of society.
Sanskritization has never meant the same thing for the
Harijans as it has for castes above the line of pollution.
Everywhere the effective adoption of the Sanskritic style of
life has depended upon a number of preconditions. These
include a minimum of economic and political power and a
not too inferior ritual status. A caste may adopt a new name
and claim a high social status but such a. claim is not likely
to be very effective where most of its members are landless
labourers and are refused the services of the village barber or
the village washerman. Dominant castes such as the Okkaligas
develop a commitment to the Sanskritic style oflife because their
economic and political position enables them to adopt such a
style with some effect.
In the past the Scheduled Castes were prevented from
Sanskritizing their styles of life by a variety of sanctions.
 The future ofthe Backward Classes 167
They were excluded from temples, bathing ghats, wells and
other public places. A large number of civic rights, necessary
preconditions to Sanskritization or upward mobility of any kind,
were denied to them by legal and ritual sanctions. The new
courts established by the British introduced the principle of
equality before the law and by doing so removed one set of
restrictions to changes in the social life of the Harijans.
However, the removal of legal disabilities did not automati-
cally enable the Harijans to exercise their civic rights. Various
kinds of sanctions were applied to keep them in their inferior
position. They were (and still are) economically dependent on
the upper castes whom they dare not offend by pressing too far
their legal claims to equality. A Harijan tenant or agricultural
labourer who dares to behave as the equal of his master
on social or ceremonial occasions may find himself deprived.
of his source of livelihood. And here it should be pointed
out that dominant castes among the Other Backward Classes
rarely look with favour upon the Harijans' claim to equality
of status. ·
Apart from moral and economic pressures, physical violence
or the threat of it is a very effective deterrent, particularly
in a village. It is not unknown even now for Harijans to
be beaten by caste Hindus for attempting to exercise their
civic rights. The dominant caste of an area is rarely (perhaps
never) a Harijan caste. It often has the strength of organized
numbers and this can be used against Harijans who are too
eager to appropriate the traditional symbols of honour. In the
past the Scheduled Castes had accepted their civic deprivations
as a matter of course. Now that a spirit of challenge has
been kindled among them, it is likely that conflicts between
the Harijans and· dominant peasant castes will become more
pervasive.
In order to gain an understanding of the issues involved in
the conflicts between Harijans and caste Hindus it may be
useful to go back a little in time. The growing emancipation of
the Adi-Dravida untouchables from traditional disabilities had
aroused the wrath ofthe Kallas as early as the 1930s. The Kallas
are a dominant peasant caste in Tamilnadu, classified among the
Other Backward Classes. Hutton reports that:
168 Sodety and Politics in India

In December 1930 the Kallar in Ramnad propounded eight

prohibitions, the disregard of which led to the use of violence
by the Kallar against the exterior castes, whose huts were fired,
whose granaries and property were destroyed, and whose livestock
was looted.1 4
The 'eight prohibitions' related, among other things, to the
use of ornaments of gold and silver, the use of upper garments
by· women, and the use of umbrellas and sandals. Hutton
further writes:
In June 1931, the eight prohibitions not having been satisfactorily
observed by the exterior castes in question, the Kallar met together
and framed eleven prohibitions, which went still further than the
original eight, and an attempt to enforce these led to more
violence.IS
It may be mentioned that one ofthe eleven prohibitions was that,
'Their children should not read and get themselves literate or
educated'. 16
The attempt of the Harijans to change their social customs
does not proceed by way of Sanskritization alone. There are
many non-traditional elements in the life-styles of the upper
strata which the Harijans also seek to imitate. Whatever may
be the idiom adopted, the very fact of upward mobility requires
the rejection of many of the civic disabilities imposed by the
former upon the latter. The upper castes are likely to see in
this a threat not only to their social status but also to their
political and economic power. This being the case, conflicts
between Harijans and caste Hindus are likely to continue for
some time. Anthropologists who have done field work in village
India in recent years report the existence and sometimes the
intensification of such conflicts. But one important point has to
be borne in mind: the nature of these conflicts and in particular
the issues over which they arise tend to change.
Even in South India the issue of wearing upper garments or
constructing brick and tile houses is no longer a living one. In
these matters the Harijans seem clearly to have won their battle.
But other issues remain, and new ones tend to emerge as the
hitherto untouchable castes press forward in their campaign to
gain full social equality with the caste Hindus. Before taking up
 The future ofthe Backward Classes 169
some of the issues which are likely to figure in the immediate
future, it may be useful to take stock of the gains which the
Harijans have accumulated over the last few decades.
Diacritical distinctions in the matter of dress, ornament and
habitation are now rarely enforced by caste Hindus upon the
Scheduled Castes. When the latter continue to retain their
former style of life, it is more because they lack the economic
resources to acquire the symbols of upper caste society and less
because they are coerced by the latter to retain their traditional
marks of inferiority. While doing field work in Tanjore district,
I was struck by the difference in dress between Palla women
of the younger and older generations. The younger women,
particularly on festive occasions, now commonly wear blouses
and saris of synthetic fibre. This does not any longer evoke
violence from the upper castes; all that remains is a faint
attitude of mockery among the Brahmins at the extravagance
and vulgarity of the new generation of Harijans.
The adoption by Harijans of some of the upper caste symbols
of status is likely to become increasingly common. These
symbols, however, are by no means all a part of the Sanskritic
model. The T anjore experieQce referred to above seems to
be of widespread occurrence. Bailey reports a comparable
situation from Bisipara, a hill village in Orissa. In describing
the participation of Pan untouchables in an annual festival in
1959, he says: ·
Many of the men wore shirts and long trousers and shoes, certainly
as a mark of status and emancipation because the normal dress of
the villagers is a dhoti, and even sandals are worn only when the
ground gets unbearably hot in April and May. The Pan women
wore blouses and mill woveri saris, and several of the younger ones
had put on lipstick and face-powder.17
There are even now certain pockets where the old disabilities
continue to be enforced. The Report· of the Commissioner for
Schedules Castes ~nd Scheduled Tribes for 1961- 65 states:
The Scheduled Castes are not allowed to wear dhotis below the
knees, and a Scheduled Caste bridegroom cannot put on a turban
with a tu"a in some of the villages of Madhaya Pradesh. In some
areas of this State band music is not allowed to be played at the
170 Society and Politics in India
time of marriage among the Scheduled Castes, their women folk
canno! wear bangles and other ornaments made of silver and the
Scheduled Caste people are not allowed to ride a horse or use a
bullock cart as means of transport. IS
There can be little doubt, however, that the areas in which such
restrictions are enforced are shrinking at a rapid rate.
There is a certain measure of ambivalence in the attitude
of the Harijans towards the traditional status symbols of the
upper castes. There is, on the one hand, an urge to adopt
many of these symbols and, on the other, an undercurrent of
resentment against the entire traditional order. The rejection of
the traditional symbols of status is helped by the existence of
alternative styles of life towards which the Harijans are likely to
tum in increasing degree. Before discussing the significance of
some of these it may be useful to consider briefly the part played
by Sanskritization in tribal society.
It is clear that the tribal people have been isolated to a far
greater extent from the broad stream of Sanskritization than
have the Scheduled Castes or the Other Backward Classes. In
spite of this they have felt the impact of Sanskritic ideas and
values and this impact has gathered momentum over the last
several decades. This has no doubt been largely due to the
opening of the tribal area to traffic from outside. One of the
most general effects of Sanskritization in this case is that it
leads to the integration of segments of tribal society into the
wider caste structure. Historically there have been numerous
examples of this kind of integration. The Bhumij in Eastern
India, the Raj Gond in Central India and the Patella in Western
India provide instances of tribes which have been integrated into
the caste structure.
The Sanskritization of tribal communities and their integra-
tion into the caste structure cannot be understood simply as a
change in rituals or life-styles. Rather this change in life-styles
usually symbolizes a more fundamental transformation in their
productive organization. Generally, it is only as tribal people
get integrated more fully into the wider economic system that
Sanskritization begins to act in a significant way.
As a typical example of Sanskritization of tribal communities
 The future ofthe Backward Classes 171
one may, among other things, consider the Bhagat Movement
among the Oraons of Chhota Nagpur, Over the last several
decades certain sections of the Oraons have sought to mark
themselves out from the main body of their tribe by adherence
to a style of life in which vegetarianism, teetotalism and ritual
abstentions of various kinds are given a prominent position. The
connection of these codes of conduct with Sanskritic Hinduism
has been noted by many anthropologists. That such movements
are often defined in opposition to the wider Hindu society
should not be allowed to obscure the fact of their absorption
of many of the values of that society.
The Meenas of Peepulkhunt in Banswara district in Rajasthan
offer another striking example.1 9 Until recently classified as a
section of the Bhils, they are now taking in a big way to the
adoption of Rajput names and other elements of the Rajput style
of life. This leads in some cases to the repudiation of their tribal
ancestry, followed by claims to Rajput status. In concrete terms,
the consequences of this kind of 'passing' can be seen in changes
in dress, worship, rules of marriage, etc. In this area also one
encounters the development of Bhagat Movements comparable
to the ones of Chhota Nagpur.
The efforts of Harijans and Adivasis to Sanskritize their
life-styles are not always met with success. The structural
distance between these communities and the upper castes is still
too great for the former to pass successfully into the ranks of the
latter. In Indian society the rules of caste endogamy ensure the
maintenance of structural distance between groups which are
of widely different background. Even when intermarriages do
take place across castes, the barrier of untouchability is rarely
crossed. Indeed, it continues to restrict intermarriage even after
conversion to Christianity. Nor is the Adivasi, even when he is
highly educated, in a better position when he seeks intermarriage
with a caste Hindu.
Where attempts at Sanskritization fail, the consequence is
often a feeling of deep resentment among Harijans and Adivasis
against the upper castes. This resentment is born out of an
attitude which is different from what prevailed in the past. In
traditional society inequality between communities was accepted
as a fundamental value. This value has been formally rejected
172 Society and Politics in India
by the new legal and political order and where it still exists it is
challenged at every point.
The attitude of resentment against the traditional order
is most easily perceptible among the younger generation of
Harijans who have been exposed to the ideas of secularism
and democracy. In some places it has been nourished by
social movements of a fairly organized nature. In Tamilnadu
the Self-respect Movement challenged the traditional social
and ritual order and sought to emancipate the lower castes from
the domination of the Brahmins. It played an important part
in the creation of a new climate among the Backward Classes,
including the Harijans.
In Maharashtra over 2,000,000 Harijans became converts to
Buddhism. The Neo-Buddhist Movement was spearheaded by
the late Dr B.R. Ambedkar, a Harijan leader who enjoyed an
all-India reputation. Dr Ambedkar was an indomitable critic
of the traditional Hindu social order with its emphasis on
inequality, segregation and ritual pollution. The conversion to
Buddhism can be seen as an assertion of self-respect on the
part of Harijans who refused to accept the degraded position
assigned to them in Hindu society. The neo-Buddhists have
dispensed with some of the traditional rituals of Hinduism
although in other regards their style of life has not altered very
significantly.
Buddhism is not the only religion which has attracted the alien-
ated sections of Hindu society. Islam, Sikhism and Christianity
have also attracted converts from the lower strata. Christianity
plays a very active part today among many tribal communities.
Throughout tribal India a variety of Christian Missions and, in
particular, the Roman Catholic Church operate as active agents of
social change. Besides providing an alternative system ofreligious
values, the Missions have introduced many new features into
tribal society such as education and modem medical facilities.
It is no accident that many of the leaders of tribal India are
Christians or at least have been educated in Mission schools.
The Harijans also have been converted in large numbers to
Christianity particularly in South India.
The Harijans and the Adivasis - particularly the younger gen-
erations among them - seem to be groping for new symbols to
 The future ofthe Backward Classes: 173
which they might attach themselves. The symbols of traditional
Hindu society are no longer adequate and it is for this reason
that the Adivasis (and sometimes also the Harijans) endeavour
to recreate a largely imaginary past in which their life was more
pure and had not been corrupted by the priest, the moneylender
and the other evils of upper caste Hindu society. It should again
be emphasized that the kinds of symbols to which these sections
of the Harijans and the Adivasis are reaching forward are likely
to be very different from the ones which the upper layers of the
Other Backward Classes tend to adopt.
But the Harijans and the Adivasis are not entirely tied
to an imaginary past in their search for new symbols, The
modem secular social order provides alternative symbols and
values which are likely to become increasingly important in
the future. In order to understand how this is likely to come
about we have to consider the process of Westernization in
contemporary India.
We shall not attempt here to give a precise definition of
Westernization.20 Broadly speaking, the process refers to the
adoption by a community of Western elements in dress, habits,
manners and customs. An important agency ofWesternization in
this sense is the modem educational system which is associated
with new norms and values, and.new symbols of prestige. The
English language, which occupies a central position in the new
educational system, is an important symbol of status in every
sector of Indian society.
Fifty years ago Western education was virtually a monopoly
of the Brahmins and a few upper castes. The Backward Classes
made a belated start in the adoption of Western elements into
their style oflife. However, since the end ofthe First World War
demands began to be made by the leaders of these communities
for the benefits of Western education. In fact, the demand for
educational concessions was a major plank in the Backward
Classes Movement. These demands were put forward in a
particularly organized manner in South India, especially in
Madras and Mysore and also in Maharashtra.
The Backward Classes in general and particularly the Other
Backward Classes have been trying consistently to narrow the
174 Society and Politics in India
gaps in Western education between themselves and the 'forward'
classes. Mysore State offers a striking example where rapid
strides have been taken in µtis direction by the two dominant
backward castes, the Lingayats and the Okkaligas. Such gains
which may appear rather limited in absolute terms have, of
course, to be measured against the past, and the rather low
fJVerage rate of literacy and education in the country. Further,
whatever may be the results achieved so far, there is every
indication of a continuous rise in the demand for education by
all sections of the Backward Classes.
There is no gainsaying the fact that literacy and education
among the Scheduled Tribes and the Scheduled Castes are
still very low. But once again what has been achieved must be
measured against the background of traditional society which
denied almost wholly the benefits of education to these sections
of society. Viewed in the light of the past, the progress of
some of these communities (e.g. the Lushais of Assam or
the Mahars of Maharashtra) appears remarkable. The benefits
of education are most likely to spread with increasing speed
among the Scheduled Tribes and Castes. For, not only are
demands being generated from within but the -Government
is investing increasingly larger sums ·of money to meet these
demands.
It is difficult to provide accurate figures for Harijan and
Adivasi intake into schools and colleges. Isaacs gives certain
estimates which present a general picture of changes that have
come about since Independence.
In the past fifteen years, while the total school population has
more than doubled, the figures for ex-Untouchables has swelled
eightfold or tenfold, to almost six million, and now, as already
indicated, there are more than four million ex-Untouchable
children in primary schools (which are now largely free to
all), and something close to a million and a half in middle
and high schools (in which the Scheduled Castes are exempt
from fees and in many places receive stipends for support),
and about fifty-six thousand in institutions of higher leam-
ing.21
Over the last two decades there has been a phenomenal rise
in the number of scholarships awarded by the Government
 The future ofthe Backward Classes: 175
of India to members of these communities. In 1944-5 only
114 post-matric scholarships were awarded to members of
the Scheduled Castes. The figure had risen to 10,034 by
1954-5 and to 60,165 by 1963-4. Corresponding figures for
the Scheduled Tribes are, 84 for 1948-9, 2,356 for 1954-5 and
11,670 for 1963-4.22 There can be little doubt that these figures
will continue to rise (though certainly not at the same rates) over
the next ten years.
There are indications that the Harijans and the Adivasis are
making increasingly effective use of the facilities provided by
the Government to better their economic and social positions.
Formerly many of the posts in the higher services reserved for
members of these communities could not be filled for want of
suitable qualified candidates. This is no longer the case today
and is not likely to be so in the future. To take a crucial
example, whereas only five of the sixteen posts reserved for
the Scheduled Castes in the IAS were actually filled in 1957,
since 1962 all the posts reserved for them have been actually
filled. 23 The position with regard to the Scheduled Tribes is
similar.
Modem education acts in a very special way as a solvent of
the barriers between different communities. The modem school
is an effective area of desegregation and this is of particular
importance from the viewpoint of the Scheduled Castes. The
school brings together in increasing numbers children from
castes which are widely separated from one another. Even
in the orthodox South, Brahmin, non-Brahmin ·and Harijan
children come together in the school. In the villages ofTanjore
district,. for instance, it is a new experience for Harijan children
to sit with the children of their Brahmin masters in the same
room and study and play together. This early experience, even
when it is short-lived, creates a new sense of confidence
among the Harijans which is almost entirely absent in the
older generation.
The differences in attitude between the generations and the
future implications of these differences were brought home to
me vividly in the course of my fieldwork in a Tanjore village.
In this village, the Brahmins live separately in their exclusive
area of residence called the agraharam. The Harijans live apart
176 Society and Politics in India
on the fringe of the village or a little away from it. Even today
Harijan men almost never enter the agraharam. When a Harijan
tenant has to deliver grain t<? the Brahmin landowner, he stands
at the head of the agraharam and calls out until somebody (often
a non-Brahmin) is sent to bring the grain. The village school
is situated in the same agraharam, although at one end. It is
attended by a number of Harijan children whose movements
within the agraharam are now hardly noticed. While children
freely come and go, a Harijan elder still considers it an act
of daring - perhaps even a little impious - to enter the
agraharam.
It has been noted earlier that there were in the past many
differences in culture between the Scheduled Castes and the
higher strata. These differences were symbolized in different
styles of speech, including vocabulary, diction and accent. Such
differences tend to be ironed out today as boys and girls from
every caste are brought up together in the atmosphere of a
common school.
In Madras State the 'mid-day meals scheme' has been a sub-
ject of much discussion and controversy. The scheme provides
mid-day meals for all school-going children throughout the
State. The children are generally served together irrespective
of caste. This sometimes leads to abstention from the meals by
boys and girls of the upper castes, particularly the Brahmins.
But for all this, the experience is a ·new one and in view of
the central importance of commensality in the Hindu scheme
of values, it is likely to have considerable significance for
the future.
Western education creates a hunger for white'-collar occupa-
tions among the younger generation of Adivasis and Harijans.
Those who succeed in getting jobs as teachers, account-
ants or clerks tend to be cut off from their communities.
Their position is, in fact, replete with uncertainty. Stand-
ing between two worlds, they are often unable to gain a
foothold in either. Although their number so far is limited,
it is bound to increase and this increase will render more
acute the social problems of transition from the most back-
ward sections of society to the new middle class. As Isaacs
has put it:
 The future ofthe Backward Classes: 177

it is the educated ones who are bearing the brunt of the painful
experience of change. They have to acquire a whole new identity
for themselves, a whole new way of relating to a society that
is not yet quite ready to welcome them on any real basis of
equality.24
The tremendous urge for white-collar jobs among the Sched-
uled Tribes and Castes cannot be explained solely by motives
of economic gain. In every range of Indian society a very high
value is attached to non-manual work and a white-collar job is
universally viewed as a passport to respectability. This view is
sometimes held all the more keenly by the Backward Classes
who have until recently been almost wholly excluded from such
occupations. Now that white-collar jobs are almost within reach,
no price appears to be too high for them.
It is largely for this reason that boys from the Scheduled
Tribes and Castes are not attracted in very large numbers
to the craft training schools of the Government. A Harijan
boy who has been through high school would normally prefer
a clerical job even where higher earnings are offered by
skilled manual work. Among people who have been tied to
social degradation for generations, the appeal of respectability
is particularly urgent.
Isaacs has provided a sensitive description of the ways in
which educated Harijans use the anonymity of city life to
discard their traditional identities. Middle-class jobs enable
them to adopt many elements of the Western style of life which
is differentiated less by caste than by income, occupation and
education. This kind of transition is of course achieved by only
a few and then too it is never wholly effective.
. Westernization as an alternative means of acquiring status has
made inroads also into tribal society. Martin Orans, who has
made an intensive study of tribal life in Bihar, writes: 'The
aspiring Santai who was well acquainted with Jamshedpur
might adopt Western clothing, an automobile or motorcycle,
and a puk/ea, house ... rather than vegetarianism, teetotalism, and
a sacred thread'.2s
Thus, the values and status symbols of the upper castes tend
to be progressively internalized by the Backward Classes. The
178 Society and Politics in India
competition for higher social status spreads and becomes inten-
sified. This competition is waged sometimes in the traditional
idiom, e.g. when a lower caste tries to Sanskritize itself and
sometimes in a more modem idiom, e.g. when demands are
made for higher education or Government jobs. In the past it
was generally accepted that different life-styles were appropriate
to different sections of society. Now the social aspirations of
the Backward Classes tend to be pitched at an increasingly
higher level.
It is clear that the competition for status cannot bring the same
kind of success to all sections of society. None the less, this
competition is an important feature of modem Indian society
and its pace is likely to quicken. In the past, social mobility
was slow and gradual because the symbols of high social status
were often not open even in theory to the lower strata of society.
The value system today offers increasingly greater choice to the
individual although it is evident that the channels through which
this choice can be exercised are not equally open to all.
In the past the unit of social mobility was generally not the
individual but a caste or sub-caste. A caste had a more or less
homogeneous culture and it was difficult for an individual to
change his style of life effectively in the absence of correspond-
ing changes in his caste or kin group. Today there is increasing
scope for individual mobility. Western education, the expansion
of caste-free occupations and the possibilities of geographical
mobility - all enable the individual to change his style of life
on his own. And, ·the criterion of individual achievement is
likely to play an increasingly important part in the emerging
social order.
An increase in the tempo of individual mobility is likely to
bring about a change in the very structure of caste. Closed
status groups based upon birth are likely to yield increasingly to
relatively open status groups based upon education, income and
occupation. There is indeed a noticeable trend in this direction
in large urban centres where the ties of caste are being slowly
but gradually supplanted by those developed in the school, the
office or the club. However, these trends are as yet largely
confined to a small upper layer of urban Indian society. It may
take more than a generation for Harijans and Adivasis to gain
 The future ofthe Backward Classes: 179
full acceptance in these new status groups based on factors other
than caste.
The rapidity with which members of the Backward Classes
pass into the newly emerging status groups will depend to
a large extent on two factors. It will depend firstly on the
extent to which facilities of education and employment are
made available to them and secondly on the weakening of the
traditional attitudes regarding purity and pollution among the
upper castes. There are indications that both these tendencies
are beginning to operate but it is unlikely that significant results
will be achieved within a short time. And, the rural areas are
likely to lag considerably behind the urban centres in the
emergence of status groups based primarily upon education,
occupation and income.

So far we have been concerned largely with the movement

of the Backward Classes along one particular axis, that of
social prestige or status. The internalization of the values,
idioms and symbols of the wider society, whether through
Sanskrit:iution or Westernization, has a certain unifying effect.
It tends to pull down the walls which in the past segregated the
different sections of society. This it does by replacing clearly
differentiated styles of life by ones which are more general and
standardized. The different cultures of the multitude of castes
and communities tend more and more to be replaced by a single
culture in which the same aspirations, values and symbols are
shared by an ever-widening circle ofpeople.
The upward movement of members of the Backward Classes
is not always a smooth or easy process. Many of the status
symbols of the upper strata of society are inaccessible to the
Backward Classes if not in principle at least in reality. The
channels of mobility whether in the status system or in the
system of production are still very restricted in Indian society.
On the other hand, the new political system has thrown open
many possibilities ofadvancement to people from the Scheduled
Tribes, the Schedules Castes and the Other Backward Classes.
Today if a Harijan cannot find a place in a higher status group,
he can still hope to become an influential political leader. The
adoption of a democratic political order in a highly particularistic
180 Society and Politics in India
society ensures that people from every major section of it can
aspire to positions of power and authority. For Harijans and
Adivasis political representation is guaranteed by the principle
of reservation.
The success of the Backward Classes in the competition
for power requires a certain assertion of particularistic ties. A
Harijan must assert that he is a Harijan if he is to mobilize
the support necessary for his political advancement. And it
is here that the demands of power and status come into
conflict. Whereas the Backward Classes are prompted to merge
their identity with the higher strata to enhance their status,
considerations of power and material advantage lead them to
define their identity in opposition to the advanced sections of
society. This is the dilemma of backwardness. A low caste
would like to acquire a high-sounding title, claim Kshatriya
status and assume the symbols of high social status; at the
same time it would insist on its right to be officially classified
as backward.
It would be of interest to examine the conflicting demands of
prestige and power from the viewpoint of the Backward Classes.
How have these demands been adjusting themselves and what
shape are they likely to take in the future? Frustrated in their
efforts to gain social acceptance, the Backward Classes are likely
to tum increasingly to political action.
The new political system adopted since independence offers
vast possibilities to large minority blocks such as the Scheduled
Tribes, the Scheduled Castes and the other Backward Classes.
It cannot be said that these possibilities have as yet been fully
exploited. The process by which the different sections ofsociety
get progressively inducted into the arena ofpolitics is a slow and
gradual one. In India politics was in the early stages_ confined
largely to a few upper castes. They were the ones to take the lead
in forming political associations in order to bargain for power
with the British.
But politics has not remained confined to a small coterie of
upper castes. It has made its way into ever-widening areas of
society. The four general elections since Independence have
brought home to every section of the Backward Classes the
importance of organized politics. There are many indications
 The future ofthe Backward Classes: 181
that they are going to make increasing use of political action to
bargain for a better position in society.
In its formal aspects at least changes in the political order have
been much more radical than.changes in other spheres oflndian
society. And changes in political form do have an impact on the
intensity ofpolitical participation. As Myron Weiner writes, 'But
it is also true that India has become tremendously politicised.
Politics has become the avenue for personal advancement in
a society in which commercial activities offer little status and
administrative posts are relatively few in number'. 26
There is a clear trend towards increasing participation in the
political 'process by sections of society which have hitherto been
excluded from positions of power. It is difficult, however, to
assess how far there is as yet a real articulation of interests.
There is widespread poverty, ignorance and illiteracy among
the Backward Classes, particularly the Scheduled Castes and
Tribes. Most members of these communities have as yet only
a dim awareness of the nature and strategy oforganized politics.
It is not unlikely that those who riow represent their interests do
so in a narrow and short-sighted manner.
However, even the most backward sections of society have
by now had the experience of four general elections. Their
awareness ofpolitical parties, movements, machines and election
campaigns has grown steadily over the last fifteen years. By now
the experience of being courted during elections by eminent
leaders has become familiar even to the hitherto 'exterior' castes.
This must in course of time create in them a new awareness of
their strength in the political arena.
Recent field studies by social anthropologists often report a
change of mood among the Harijans particularly of the younger
generation. Bailey27 indicates how even in a remote hill village
in Orissa the Harijan Pans played an active and even aggressive
role in the 1957 elections. In Tanjore district where I did
field work in 1961-2 the Congress and its ally, the Dravida
Kazhagam, drew active support from many Harijan youths for
the election campaign.
Participation in election campaigns and contact with party
bosses tend to create among the younger Harijans a sense of
impatience towards the slow and uncertain process of social
182 Sodety and Politics in India
mobility through imitation of the life-styles of the upper castes.
Young Harijans who are inducted into political machines may
feel that they can dispense with Sanskritization. Indeed, for
a person who aspires to be a political leader among Harijans
and Adivasis, a repudiation of Sanskritization may have a high
symbolic value. In Tamilnadu, where the Dravida Kazhagam has
had a strong appeal among the Backward Classes, it is not rare
for young Harijans to reject explicitly the values and symbols of
the upper castes.
Whereas a generation ago an ambitious Harijan might have
tried to acquire social respectability by changing his style of
life, his counterpart today is more likely to try to build political
connections. And young Harijans with ambition, drive and
initiative are in demand all round. They are sought by every
political party; for, they are not only valuable as vote-banks, but
can also be put up as candidates for reserved seats from the Zilla
Parishad right up to Parliament.
The scramble for power among linguistic, religious and caste
groups has become an important feature of post-independence
India. Virtually every region, community and caste claims special
benefits for itself. And pressure is exerted at every level for the
satisfaction of these claims. The Backward Classes have not
lagged behind in this regard. In fact, the policy of protective
discrimination pursued by the Government has encouraged
them to put forward their demands with special vigour.
The institution of Panchayati Raj is likely to quicken the pace
of politicization. Provisions have in general been made for the
reservation of seats for the Scheduled Tribes and the Scheduled
Castes at all levels of the three-tiered structure. The village
panchayat is likely to become in many areas an arena for conflicts
between Harijan members and those of the 'clean' castes. The
latter are rarely prepared to treat on equal terms those who
had been until recently excluded from many of the important
spheres of social life. So conflict is likely to multiply in the
near future. And, perhaps conflict is a necessary condition for
the articulation of the interests of the hitherto underprivileged
sections of society.
In this context it is well to remind ourselves that there are deep
cleavages within the Backward Classes. It is likely that the most
 The future ofthe Backward Classes: 183
stubborn opposition against the attempts of Harijans to improve
their social position will come from those who are immediately
above them rather than from the top castes. It is a matter of
common observation that competition for status is often most
acute between segments which are structurally adjacent to otie
another. This is because such segments operate largely within
the same social universe. It is for this reason that castes which
are just above the line ofpollution are more likely to be jealous of
their privileges in relation to the Harijans than the Westernized
upper strata of society.
Conflicts between Harijans and caste Hindus over civic rights
are likely to play an important part for some time to come. The
new generation of Harijans is no longer in a mood to accept
with resignation the civic disabilities imposed upon them by the
upper castes. Their contacts with politicians and officials have
given them a growing awareness of their rights as citizens of
a democratic society and they are rapidly becoming jealous of
these rights. A show of strength on their part is a likely source
of violence in the rural areas where, in spite of a superficial
acceptance of democratic values, the structure remains by and
large inegalitarian.
In 1957 severe rioting took place in Ramanathapuram district
in Tamilnadu between Harijans and Thevars, a locally dominant
caste included among the Other Backward Classes. The riots
centred around the Immanuel murder case in which a young
Harijan was stabbed to death, allegedly for offering rude
remarks to a prominent Thevar politician. The setting of the
incident was provided by a by-election in which Harijans and
Thevars supported rival candidates.
The Ramanathapuram riots provide only.a striking example of
a phenomenon which is of widespread occurrence and probably
on the increase. It would perhaps be unreal to expect the
Harijans to become integrated into the wider social system
with rights of full and equal participation without any conflict
or violence. But the extent to which such violence can be
contained will depend in large measure on the effectiveness of
our machinery oflaw and order.
A special correspondent of The Mail wrote during the third
general elections: ·
184 Society and Politics in India

Shrewd observers draw attention to the new trend in the political

scene in the State, with emphasis shifting from the outdated
Brahmin versus non-Brahmin to Harijan versus non-Harijan as a
reaction to the numerous concessions made to the Harijans by the
Government. 28
These remarks, made with regard to Tamilnadu, seem to
indicate a trend which has general significance.
As Harijans in increasing numbers enter the arena ofpolitics,
conflicts between them and the upper castes are bound to multi-
ply. These conflicts will in all likelihood manifest themselves
most clearly at the level of the village or the small local com-
munity. There are now many village panchayats where Harijans
have a large representation. In rare cases the domination of the
panchayat by Harijans may lead to the withdrawal of the 'twice
born' from participation.
Where Harijans are more or less evenly matched in numerical
strength with caste Hindus, a certain amount of tension or even
violence is likely to become a part of the system. During my
stay in Tanjore district in 1961-2 several cases of violent
conflict between Harijans and non-Brahmins (particularly of
the dominant Kallar caste) came to notice. There was one
incident in a neighbouring village which culminated in the
murder of a Kallar landowner by the Harijan president of the
village panchayat. During the legal proceedings that followed, the
Kallar tried to storm the court room at Tanjore in an attempt to
do violence to the offending Harijan.
Tribal politics differs in many ways from the politics of the
Harijans or the Other Backward Classes. There are certain
special problems in their case which arise partly out of their
geographical isolation and partly out of their ethnic identity
which is even more sharply defined than that of the Harijans.
We noted earlier how the process of Sanskritization was in the
past responsible for the weakening of the cultural identity of
tribal peoples in the different parts of the country. But here also
the process of cultural assimilation is countered by a variety of
forces (principally political in nature) which seek to reassert the
identity of the tribal population.
The shift in emphasis from a ritual idiom of mobility to one
 The future ofthe Backward Classes: 185
which relies more on organized politics has been noted by
various observers of tribal society. Discussing the trend of
modem leadership in Chota Nagpur, Dr Vidyarthi writes:
The most significant and obvious fact is the switch over from reli-
gious leadership. to political leadership. This change corresponds
to a difference in aims and a difference in methods. The modem
leaders are representing the tribal people to an outside political
world. They are not concerned primarily with raising the social
status ofthe tribals by making ritual practices more closely resemble
those of the Hindus. Their chief task is to improve the material
conditions of the tribals and to obtain government funds and
services. Their activities, then, are more within the realm of the
civil and political and less within the realm of the religious and the
social.29
The search for a new identity by leaders oftribal communities
is understandable. There are important sections of people
outside the tribal world who have warned against the loss of
their social and cultural identity. In the case of the Adivasis,
assimilation (as opposed to integration) has been viewed as
a genuine threat by no less a person than the late Prime
Minister. The attitude of 'progressive' Indians towards Harijans
is fundamentally different in this regard and less ambivalent.
Few voices would be raised in protest against the assimilation
of Harijans into the wider social system.
Close observers of the Adivasis have, in consequence, noted
the development of a spirit of 'tribalism' among them. This is
due to a variety of factors, among which the special treatment
policy of the Government and the work of the Christian
Missions are of importance. Today the spirit of tribalism
can be given expression through the processes of organized
politics. The demand for a tribal homeland and the growth of a
political party to put it forward indicate the politicization oftribal
society. Professor N.K. Bose, until recently Anthropological
Adviser to the Government, writes: 'Tribal communities for-
merly tried to better their condition by either identification with
the Hindus or with the ruling class through Westernization and
Christianity. Now power can be derived by political organisation
into parties'. 30
Tribal groups may, of course, press their demands through
186 Sodety and Politics in India

parties committed to unity and integration or they may form

separatist parties to represent their exclusive interests. Separatist
demands have been particularly strong in the NEFA and Assam.
The Jharkhand Party had also made similar demands. On the
other hand, the Congress has shown an extremely high degree
of resilience in meeting the special demands of communities of
every kind. The recent merger of the Jharkhand Party with the
Congress in Bihar seems to suggest that the party in power has
many devices for softening the tone of separatist demands.
The kind of compromises which the ruling party is prepared
to make is likely to deflect it, to some extent, from its main aims
and policies. The Congress is likely to continue its tolerance of
tribal exclusiveness partly as a measure of political expediency;
an uncompromising attitude is more likely to precipitate a crisis
than to overcome it. It seems probable that the special treatment
policy will slacken the pace of development for some time to
come by creating an atmosphere of apathy and complacency. It
is difficult, however, to foresee any radical change in this policy
in the immediate future. And the policy remaining broadly the
same, the chances of a crisis being precipitated by separatist
forces appear rather limited.
It is most unlikely that the 'communal' element will be
eliminated from Indian politics for some time to come. N.K.
Bose writes, 'under the climate created by special treatment,
communal consciousness has been inordinately accentuated'.31
But the significance of this development can be evaluated in
different ways. The growth of 'communal consciousness' need
not be viewed as necessarily an unhealthy or disruptive force. It
may, on the contrary, be a precondition to the integration of the
tribals into the wider body politic. For the measure of integration
lies not so much in a passive acceptance of the status quo as in
the adoption of a body of common political rules through which
divergent interests are organized and articulated.
Thus, what has been viewed by some as an increase in
communal consciousness may be one step forward in the
politicization of Indian society. The years to come are likely
to witness a fuller participation in the political process by larger
sections ofthe tribal population, and a more _effective articulation
between tribal leaders and the masses. The wider political order
 The future ofthe Backward Classes: 187
can be brought close to the tribal masses only when their leaders
at every level learn to put forward demands on their own behalf.
In a heterogeneous society where particularistic ties are of such
importance, it is unreal to expect political integration to come
about without large concessions to special demands.
It may be interesting in this context to re-examine the old
controversy regarding isolation, assimilation and integration as
three alternative lines along which tribal communities might
develop. The case for isolation seems to have been abandoned as
both unreal and undesirable. The tribal people everywhere are
being drawn increasingly into wider social, economic and politi-
cal networks. In addition to market forces of diverse kinds, the
developmental activities of the Government are bound to make
steady inroads into the world of the Adivasis. And the Adivasi
leaders are rapidly coming to realize that they can gain more for
themselves and their communities by coming to terms with the
Government and the ruling party than by remaining isolated.
Increasing articulation with the wider social, economic and
political system is likely to lead in the long run to the disappear-
ance ofmany ofthe distinctive features oftribal life. The process
of modernization has a certain standardizing effect and it may
be too costly for the Government to keep alive in a hot-house
atmosphere the picturesque or artistic elements oftribal culture.
The search for an identity on the part of tribal leaders does not
always involve a genuine revival of the traditional heritage; more
often it is part of an attempt to come to terms with political
forces; ·which in the long run are bound to destroy the very
bases of this identity.
The Harijans illustrate even more clearly than the Adivasis
the mechanism of integration through the political process.
At the height of the Depressed Classes Movement under Dr
Ambedkar it had at times appeared as if they might move
further and further away from the wider social system. In
the years since independence the Congress has successfully
absorbed their demands and integrated the bulk of the Harijan
leadership into the structure of the party.
The Congress Party seems to have succeeded in winning over
the loyalties of the Harijans by making numerous concessions
to them. Although such loyalties may not go very deep, it is
188 Society and Politics in India

doubtful whether any other party can compete with the Congress
in this regard. Bailey has recorded in some detail the nature
of Harijan support for the Congress in Bisipara; which he "
attributes to 'the ideological pull (or perhaps one should call
it "enlightened self-interest") of Congress policy towards the
Harijans'. 34 I found confirmation of Bailey's observations in a
Tanjore village more than a thousand miles away.
The integration of the Harijans into the wider body politic
is, by comparison with the.Adivasis, relatively easy. Unlike the
latter, they are not geographically concentrated and there is
hardly any question in their case of the demand for a separate
homeland. But numerically they are far more important than the
tribals and, for historical reasons, the problem of untouchability
has caught the imagination of the Western educated elite in the
country. For this reason the Congress has been compelled to
accommodate the demands of the Harijans at every level and
these demands are likely to become progressively organized both
within and outside the party.
There is a curious dilemma in the position adopted by the
Government and the ruling party towards the Harijans. On
the one hand, the leaders among the latter are impatient of
the limited and tardy nature of the concessions granted to
their community. On the other, there is a growing resentment
among castes of the middle region against the 'soft' attitude
of the Government towards the Harijans. The Congress, in
gaining the support ofHarijans through accommodation of their
demands, is likely to create increasing disaffection in the ranks of
the upper castes.

The arguments presented above suggest that the Harijans (as

well as other under-privileged groups) are likely in the near
future to seize their civic rights with increasing success and
perhaps also get increasing representation in political bodies
of various kinds. At this point it is possible only to raise the
question as to how effective this representation is likely to be.
In the last analysis it may be argued that the fundamental
problem facing the Harijans (and also, to some extent, the
others) is a material one. It is a problem oflandlessness, poverty
and unemployment. How well can the Harijans or the Adivasis
 The future ofthe Backward Classes: 189
make use of their position of growing political strength to solve
or even tackle some of these basic economic problems?
It is admitted on all hands that the economic transformation
of the bulk of the Scheduled Tribes and the Scheduled Castes
has been taking place at a creeping pace if at all. The Harijans
and Adivasis by and large continue to be poor, indebted and
landless. Governmental measures to ameliorate their economic
conditions have had, at best, a moral or symbolic effect.
Given the magnitude of the problem, it is doubtful whether
any substantial change can be brought about solely through ,
measures such as the allotment to a few Harijans of bits and
pieces of land or the reservation of a certain number of jobs
in the Government for them. The same, of course, applies in
a large measure to the tribals.
In the pre-British economic system Harijans had almost
wholly been in the position of agricultural labourers. In certain
areas their position was no better than that of serfs tied to the
soil. British rule and developments since independence have
progressively emancipated them from their former servile status.
But the change in legal status has rarely been accompanied by
any real change in economic position. Most Harijans continue to
be landless agricultural labourers. In some ways their economic
position has been rendered more insecure than in the past. In the
past, traditional obligations assured the Harijan labourer of some
source of employment and sustenance. Today with the collapse
of traditional obligations, even this is threatened.
The material position of the tribals is not very much better
than that of the Harijans. Here, however, the issues and prob-
lems are a little different. There is, to begin with, the problem
of wasteful techniques of cultivation. Attempts are being made
by Government to control these but the problem itself is
particularly acute for only a small section of the tribal people.
Land alienation and money-lending are the two most impor-
tant economic problems which confront the tribals (and indi-
rectly the Government). The two problems are closely related,
since the money-lenders are also the ones to whom land is most
frequently alienated. They have been sought to be remedied
by legislative as well as executive action but the success so
far has been very limited. Such action sometimes has the
190 Sodety and Politics in India

consequence of raising higher the barriers between tribals and

non-tribals without bringing about any substantial improvement
in the economic condition of the former.
When tribal people do become a part of the wider social order
they generally occupy within it the lowest rungs of the economic
ladder. They join as landless labourers, often without security
of employment, working sometimes on the farms of other
people and sometimes eking out a bare existence through the
sale of baskets, mats or jungle produce. In such cases their
economic problems are not significantly different from those
of the Scheduled Castes.
It is rather difficult to visualize any radical transformation in
the economic position of the Scheduled Tribes and Castes for
some time to come. If such a transformation does come about
it is unlikely that it will have its sources within tribal or Harijan
society. Any significant change in their economic position will
require major changes in the external system, in particular the
agrarian class structure of the country as a whole. The Harijans
and Adivasis are almost everywhere prevented by their insecure
position from initiating any kind of major economic change. Nor
can governmental intervention by itself bring about any quick
and substantial improvement, the magnitude of the problem
being what it is.
Changes in the economic position of the Other Backward
Classes are, if anything, even more difficult to assess. For one
thing, they appear to be a rather assorted category, including at
one end castes which are similar in position to the Harijans and
at the other end powerful dominant castes like the Lingayats and
Okkaligas. Further, the Other Backward Classes do not form a
sharply defined category in the social system, and in the absence
of numerical data it is impossible to consider trends ofeconomic
change among them separately from such trends in the whole of
Indian society.
The Central Government's policy to do away with the
'communal' definition of the Other Backward Classes and
its refusal to draw up an all-India list of these communities
is bound to affect their position in important ways. These
communities had developed a separate identity partly as a
consequence of policies pursued by the State Governments
 The future ofthe Backward Classes: 191
of associating them with certain concessions and advantages.
When these are given to individuals on the basis of income, and
not to whole communities, the boundaries between the latter and
the so-called advanced sections ofsociety will cease to have their
former social and political significance.
It is not within the scope of this paper to consider the
prospects of change in the agrarian class structure of the
country. Here it can only be reiterated that the future of the
Backward Classes is intimately related to what happens to the
economy as a whole. In this paper we have tried to consider
the problems of the Backward Classes qua Backward Classes,
as understood in the Constitution and recognized for specific
purposes by the Central and State Governments. Although it
is true that the Backward Classes overlap in large measure with
certain economic categories, as defined in this country, they are
not 'classes' at all but groups of communities.
The distinctive features of the Backward Classes are in large
measure related to the very structure of Indian society. When
this structure undergoes change, the Backward Classes are likely
to lose much of their identity. There is ample evidence to show
that Indian society is undergoing profound changes. The tradi-
tional segmental structure is losing many of its characters and a
greater interpenetration of segments is coming about. Networks
of interpersonal relations cutting across the boundaries of caste
and community are becoming an increasingly important feature
of the new economic and political systems. Finally, ritual values
which played a decisive role in keeping alive the concept of
structural distance are becoming progressively weakened.
It is true that the forces of modernization are as yet largely
confined to a limited sector of Indian society. Their effects are
felt primarily among the Western educated middle classes, par-
ticularly in the urban centres. The Backward Classes are largely
concentrated in the rural areas and the impact of Westernization
·on them has been less widespread than on the upper castes.
It is therefore not unlikely that the Backward Classes and, in
particular, the Scheduled Tribes and the Scheduled Castes will
retain their identity for some years to come.
 8 Equality as a right and as a policy
In an important essay on affirmative action in the United States
Professor Ronald Dworkin has urged the need to acknowledge
the distinction between equality as a right and equality as a
policy, a distinction that, according to him, 'political theory has
virtually ignored'.1 I would like to devote the present lecture
to an examination of this distinction for, although the subject
bristles with difficulties, something may be gained by bringing
to it the approach of comparative sociology. Problems that are
very similar to those brought -to light by the DeFunis case (and
the Bakke case) in America arise frequently and persistently in
India in a very different social and cultural setting.
For all its inherent difficulties, one gets a sense of the
distinction between equality as a right and as a policy while
considering the structure of the Indian Constitution. Due to
a variety of historical reasons the concern for equality figures
prominently in it. But the equality provisions in the Constitution
are not all of the same kind or to be found in the same place.
The principal provisions are divided between Part III, entitled
Fundamental Rights, and Part IV, entitled Directive Principles
of State Policy. Hence a student of the equality provisions in the
Constitution of India must sooner or later stumble upon the very
distinction that Professor Dworkin has urged us to make.
The equality provisions in the part of Fundamental Rights
relate on the one hand to equality before the law and, on the
other, to equality of opportunity. Here equality is conceived
in terms of individual rights and in terms of what has been
described as the 'anti-discrimination' principle. 2 All Funda-
mental Rights, including the right to equality, are enforceable
by the courts. As against these, the Directive Principles of State
Policy are not enforceable by the courts although they are of
great social and political significance. They provide a framework
for a policy of greater equality overall in the distribution of
 Equality as a right and as a policy 193
resources and in the relations between the different members
of society.
There are S'everal reasons why the equality provisions in the
Indian Constitution appear stronger and more extensive than
in the American. The American Constitution is nearly two
hundred years older than the Indian. When it was being written
equality was a novel social ideal and the possibility of changing
the structure of society through systematic social policy was still
to receive serious attention. The Indian Constitution was written
in a different age and political climate. When it was being written
there was no longer any novelty in the idea itself of the right to·
equality, whereas a great deal had come to be expected from a
policy of greater social equality.
Considering the time at which the American Constitution
was written, one would not reasonably expect it to incorporate
a framework of policy for removing the social and economic
disparities between, say, men and women or whites and blacks.
An expectation of that kind would appear normal and natural to
those who fashioned the Constitution of India in the middle of
the present century. The persistence of gross disparities between
castes and communities, and between the sexes was associated
with backwardness and, in the case of the former, also with
national disunity, and it was believed that the remedy for all
this lay not only in the creation of individual rights but also in
the formulation of an active social policy.
The makers of the Indian Constitution believed that the
individual right to equality and the social policy of equality
were complementary and would reinforce each other. That
belief is widely shared in contemporary India, and it has to
be understood in the background of the continuing hold of
the traditional hierarchy over a large part of Indian society.
Not only are resources unequally distributed but social customs
are set in a hierarchical mould. The right to equality, whether
equal protection of the laws or equality of opportunity, would
have little security without important changes in the structure of
society.
Similar arguments have been made by others elsewhere. More
than fifty years ago Tawney had pointed out in the context of
British society that equality of opportunity meant little in the
194 Society and Politics in India

absence of what he described as 'practical equality'3 The idea of

practical equality found expression in the welfare state in Britain
and other West European countries. In India the problem was
of a different magnitude, hence the decision to inscribe the
commitment to equality as a policy in the Constitution.
Not everyone, of course, believes that in order to secure
equality as a right, it is necessary to promote equality as a
policy. Those who are described as libertarians are opposed to
equality as a policy although they may be strong champions of
the equal rights of individuals. They argue that it is impossible
to pursue systematically a policy of equality without continuous
and arbitrary interference by the state in the free activities of
individuals. Their view is that equality as a policy is subversive
of equality as a right, and that one can have either the one or
the other but not both.4 This kind of argument is not commonly
encountered in India where a correspondence between the two
is widely assumed, and those who are opposed to equality as a
policy are likely to be opposed also to equality as a right, although
such opposition would not normally be articulated in public.
The Indian experience of the last thirty-five years raises
doubts about the presumption of harmony between equality
as a right and equality as a policy. Almost immediately after
the new Constitution was adopted, two major instruments of
the policy for greater equality, agrarian reform on the one hand
and benign quotas in education on the other, came up against
the provisions on Fundamental Rights. These provisions had to
be realigned by the First Amendment to the Constitution so as
to accommodate policies designed to reduce disparities between
classes and disparities between castes. 5 There has been, since
then, some limitation of individual rights, including the right to
equal opportunity, in the interest of policies designed to bring
about greater equality overall. Individual rights do not have the
same depth and firmness in India, the same anchorage in its
social structure, that they do in the United States.
In the United States as compared with India there is a
much stronger emphasis on equality as a right and perhaps
less public support for equality as a policy. In the American
tradition equality has been inseparably linked with individual-
ism. 6 Equality is valued to the extent that it is brought into being
 Equality as a right and as apolicy 195
by individuals through their own unaided effort: it is something
to be won for themselves by individuals, not imposed upon them
by an external agency, particularly the state. If Indians accept
equality they are less troubled by having to accept it under state
auspices.
The American approach to equality through individual rights
was shaped by those historical circumstances which made
America a land of opportunity. In no other country were
opportunities thrown open on such a large scale to individuals
with ability, drive and initiative. But the tremendous dynamism
of American society, with its high rates of geographical and
occupational mobility, concealed the massive impediments to
equality faced by blacks and other minorities, and by women.
Far from there being equality of opportunity in any meaningful
sense, in their case there was not even the formal guarantee
of equality before the law. Whole segments of society were
kept out of the domain within which equality as a right was
acknowledged. These segments have developed their own social
identities which cannot now be erased by a mere alteration in
the structure of legal rights. It is in this context that the need
for equality as a policy came to be acknowledged in America,
despite the general lack ofpublic sympathy for state intervention
in social arrangements. 7
Dworkin's essay is important because it forces us to consider
the tension between equality as a right ·and as a policy. He,
clearly, does not wish to surrender equality as a right but seeks
to define and protect its essential core. At the same time, he
believes that greater equality overall is desirable in itselfand that
a policy to achieve it is both legitimate and feasible. He maintains
that preferential policies in education and employment do not
necessarily violate individual rights although they may cause
hardships to individuals. His view is that the rights said to be
violated by policies for greater equality overall are more often
imaginary than real.
It is obvious that a policy of greater equality overall, though
desirable in itself, is not without its cost. Experience has shown
in the United States, and even more abundantly in India, that
such a policy puts some, or even many, individuals at a
disadvantage. But it can be argued, according to Dworkin,
196 Sodety and Politics in India
'that in certain circumstances a policy which puts many indi-
viduals at a disadvantage is nevertheless justified because it
makes the community as a whole better ofP. 8 What is actually
involved in putting 'many individuals at a disadvantage'? Does
it involve putting restrictions on their rights and, if so, what
kinds of rights?
The right to equality has been conceived in a number of ways.
Dworkin has sought to make a distinction between the right to
equal treatment and the right to treatment as an equal. The
right to equal treatment is 'the right to an equal distribution of
some opportunity or resource or burden', whereas the right to
treatment as an equal is the right 'to be treated with the same
respect and concern as anyone else'. 9 The second, according
to Dworkin, is fundamental whereas the first is derivative. It
can certainly be argued that the right to equal treatment can
be more easily sacrificed than the right to treatment as an equal;
and, further, that every sacrifice of the first does not involve a
sacrifice of the second.
The right to treatment as an equal, as Dworkin defines it, is a
very important one, but we must acknowledge that in saving it
we do not save everything. A policy of reverse discrimination,
like other policies for bringing about greater equality overall,
requires individuals to forego certain things in regard to which
they have normal expectations if not established rights. It would
be idle to pretend that there is no hardship here, or, even, that
it is easy to balance this hardship to the individual againsnhe
benefit to the community as a whole. But the hardship is not all:
hardship may be imposed with or without due concern for the
loss suffered by those on whom it is imposed; and in societies
where very little concern has been shown for those on whom
the most severe hardships have been imposed, that is no mean
consideration.

I would now like to consider the substance of equality as a right

in contemporary India so as to see where it is re-inforced and
where it is threatened by policies for greater social equality. This
is a very large question and I will have to confine myself to only
a few aspects of it, but those will have to be examined in relation.
to the structure of Indian society and its cultural traditions.
 Equality as a right and as a policy 197
The point of departure in most discussions of the subject is
that the right to equality is an individual right. 10 Whether we
talk of equality before the law, or of the equal protection of the
laws, or of equality of opportunity, the primary referent is the
individual. Historians, political theorists and sociologists have
shown how in the West the right to equality grew with the growth
ofindividual autonomy.11 In India, where the bias until recently
has been for the collectivity, the rights of the individual have
still to contend with age-old traditions, customs and habits. The
Indian experience shows that it is one thing to create the right
to equality but quite another to make that right secure: it would
avail very little to create a right if nothing could be done to give
it security.
Traditional Indian society was hierarchical to an unusual
degree. 12 Although the hierarchical conception of society was
not unknown in the West, the elaborateness, the stability and
the continuity it acquired in India is without historical parallel.
In a hierarchical society the legal order is structured in terms
of the privileges and disabilities of groups rather than the
rights of individuals. There was little concern in traditional
Indian society over either the equal distribution of benefits
and burdens (the equal treatment of individuals) or the equal
consideration of interests (their treatment as equals). Whole
segments ofsociety were excluded from positions of respect and
responsibility without consideration of individual interests.
The most striking inequalities in traditional Indian society,
supported by both classical and customary law, were those based
on caste and on gender. Though the disabilities of the inferior
castes and ofwomen were numerous and pervasive, they did not
operate in the same way in the two cases. As a recent observer
has noted, 'A paradoxical feature of inequality based on sex was
that women suffered greater disabilities the higher they were in
the caste hierarchy.' 13 Despite regional variations and historical
changes, the system retained the same basic structure for about
two thousand years and down to the middle of the last century.
Caste and family governed the activities of individuals to a
very large extent, and the subordination of the individual to
the family was more marked for women than for men, and
more for women of superior than of inferior castes. Among
198 Soci.ety and Politics in India
the Hindus, who constitute the great majority of Indians, caste
was an inescapable part of civil life. Every individual was born
into a particular caste and expulsion from caste ·deprived him
automatically of the right to property and other rights, and
amounted virtually to civil death. The first major step in the
emancipation of the individual from caste was taken with the
passage of the Caste Disabilities Removal Act in 1850 exactly a
hundred years before the adoption of the Constitution of India.
The Act of 1850 merely opened the way for enlarging the
rights of the individual; it did not do away with all the disabilities
of caste. As I have already indicated, these disabilities were
manifold and had accumulated over centuries. It would be
unreasonable to expect all the accumulated disabilities to disap-
pear as a result of a single act oflegislation or oflegislation alone,
without corresponding changes in the economy and society. A
large number of them remained untouched by law for a hundred
years, and many survived in practice even after they were legally
abolished by the new Constitution in 1950.
Among the many disabilities inherited from the past and
continuing into the present, I shall dwell mainly on those
that arise from untouchability. The classical legal literature,
the Dharmashastra, gives detailed accounts of the disabilities
suffered by the lowest castes, described variously as Chandalas,
Svapachas and Antyajas. While parts of the account are probably
fanciful, broadly the same structure of disabilities is revealed in
ethnographic descriptions oflndian society in the late nineteenth
and early twentieth centuries. However, it is difficult not only
to define untouchability but also to specify the number of its
real or potential victims. The official category of Scheduled
Castes, corresponding broadly to those earlier described as
untouchables, accounts for 15.75 per cent of the present
population. 14
The disabilities imposed on the untouchables included resi-
dential segregation; confinement to polluting and menial occu-
pations; arid denial of access to temples, wells and other civic
amenities. These disabilities were maintained to a large extent
through a set of religious values in which the ideas ofpurity and
pollution occupied a central place. But physical coercion was
also used to keep untouchables as well as others in their place.
 Equality as a right and as a polity 199
With the erosion of traditional ideas of purity and pollution, the
use of force has probably increased, since the maintenance of
social disabilities is of some economic advantage to landowners
and other privileged members of the community. However, the
line of untouchability was never very clear, and disabilities were
suffered by many inferior castes in addition to those listed as
the Scheduled Castes. To add further to the complication,
disabilities were commonly imposed by superior untouchable
castes on inferior ones.1 5
It is in this background that we have to consider the very
generous provisions made for equality as a right in the Indian
Constitution of 1950. The three principal rights to equality are:
(1) equality before the law, including equal protection of the
laws (Article 14); (2) prohibition of discrimination on grounds
of religion, race, caste, sex or place of birth (Article 15); and
(3) equality of opportunity in matters of public employment
(Article 16). It was recognized that these three articles provided
necessary but not sufficient conditions for securing the right to
equality. A further step in that direction was Article 17 according
to which untouchability was abolished, and the 'enforcement of
any disability arising out of "Untouchability"' made an offence.
The Constitutional abolition ofuntouchability underlined the
right to equality within the framework of a certain policy. As a
recent commentator has said,
The frame of Article 17 suggests that the rights there conferred
are conferred not merely for the benefit of individual(s) but as a
matter ofpublic policy because enforcement ofany disability arising
out of untouchability forbidden by that article is made an offence
punishable in accordance with law.16
In accordance with the provisions of Article 17, Parliament
enacted the Untouchability (Offences) Act in 1955. The scope
of the Act was extended and its provisions made more stringent
by a major amendment in 1976 which also changed its title to
the Protection of Civil Rights Act. These may be viewed as
affirmative steps taken by the state to give greater security to
equality as a right.
The provisions of the Protection of Civil Rights Act relate
to disabilities that may be broadly classified as religious, social
200 Society and Politics in India
and economic. It has often been pointed out that the practice of
untouchability had a religious basis, with its roots in the Hindu
ideas of purity and pollution. Since some people were consid-
ered to be in a permanent state of pollution, they were denied
access to temples; denial of access to places ofpublic worship to
such people is now punishable by law. Similarly, denial ofaccess
to wells, bathing places, cremation grounds, shops, restaurants
and other public facilities is also punishable according to the
provisions of the Act. There were in the past two kinds of
restrictions on the choice of occupation by untouchables: they
were not allowed to enter superior occupations and they were
forced to perform certain kinds of tasks such as scavenging,
flaying, tanning, etc. The imposition of either kind of restriction
is now punishable by law.
The Protection of Civil Rights Act has under Section ISA
made provisions for a machinery to ensure against the violation
ofcivil rights. These include the supply oflegal aid to the victims
of untouchability; the appointment of officers for initiating and
supervising prosecutions under the Act; and the setting up of
special courts for the trial of offences. It also has provisions for
monitoring the operation of the Act by agencies specially set up
for the purpose. Finally, the Union Government is required to
place before both Houses of Parliament Annual Reports on the
working of the Act. A perusal of these Reports shows how many
kinds of initiative by the state are considered necessary to give
support to its provisions. These range from the rehabilitation of
scavengers to cash awards for marriages of non-untouchables
with untouchables.
Article 17 and the Protection of Civil Rights Act can do some-
thing to change the social atmosphere in which untouchables
live, but such measures cannot by themselves secure fully their
right to equality. The majority of untouchables are ignorant and
illiterate, and unaware of the provisions of the Protection of
Civil Rights Act and their implications. Most of them live in
small rural communities where the distinction between private
and public is not always clearly drawn. They are generally poor
and often abjectly dependent on the caste Hindu landowners
for whom they work. The machinery of the law does not
easily reach into a rural community and, where it does, it is ·
 Equality as a right and as a polity 201
heavily compromised in favour of the established interests of
the dominant landowning castes.
A case study of the untouchables in India clearly shows that
disabilities cannot be abolished or the right to equality estab-
lished at a single stroke. Nobody would seriously claim, thirty-six
years after the Constitution was adopted, that untouchables
enjoy in the average Indian village the same rights as the other
members of the community. But if there is still such a long way
to go in securing the right to equality, this is not because of a
policy ofequality, but in spite ofit. That policy has helped a little
in securing the right, but hardly enough.
It would be a mistake to believe that overcoming disabilities
and securing the right to equality are slow and painful processes
for the untouchables alone and that for all others they are
smooth and easy. What is true for the untouchables is true,
though in a rather different way, also for women. It is also true
in more or less the same way, though not to the same extent,
for various inferior castes and other disadvantaged groups in
addition to the untouchables. It is for the benefit of all these
various groups - untouchables and others - that a policy of
greater equality overall has been made a part of the Directive
Principles of State Policy.

I have tried to show that there are historical circumstances in

which it is difficult to secure the right to the equal protection
of the laws without some affirmative action by the State. Is it
possible to secure the right to equal opportunity under those
circumstances without any affirmative action? Even if we argue
that it is not, we must keep in sight the fact that the implications
of affirmative action for securing the right to equality are not the
same in the two cases. Hence, those who support affirmative
action for the removal of disabilities need not support it for the
equalization of life chances.
Equality of opportunity is an important component of the
right to equality in most contemporary societies. But it must
be emphasized that the scope of equality of opportunity varies
greatly from one society to another. It has a limited scope in
many societies, including some in which there is a considerable
measure of substantive equality. Many tribal societies, including
202 Society and Politics in India

those in the Islamic world, are egalitarian in a certain sense, yet

the concept of equality of opportunity is not fully applicable to
them. Equality of opportunity becomes a value where a high
value is placed on the individual, especially on individual
achievement. It is possible for societies to value equality while
disregarding individuals and their achievements.1 7
Traditional Indian society emphasized neither equality of
condition (as many tribal societies do) nor individual achieve-
ment (as most industrial societies do). It was considered right,
proper and desirable there for the individual to remain in the
station of life into which he was born, and this is to some extent
true of all predominantly agrarian civilizations. Competition,
individual achievement and equality of opportunity have a spe-
cific institutional setting, being associated both historically and
sociologically with a particular occupational structure. Equality
of opportunity is an issue in the office or the factory in a way in
which it is not in the cottage or the farm. 18
In India a modem occupational structure has emerged but
it does not have the same reach and scope as in an advanced
industrial society. The majority of people continue to live
by agriculture and related activities, although the traditional
occupational structure has lost much of its coherence. Only a
minority of people have employment in settings where equality
of opportunity can be reasonably expected to work. Not only
does equality of opportunity have a different significance here
as compared to, say, the United States, but the problem of
ensuring it is also different in nature. Employment policy in
India tends to be directed towards creating more jobs in those
sectors where individual mobility is relatively free from the
constraints of tradition.
In India the constraints on individual mobility are enormous,
but even the constraints generally regarded as traditional are not
all ofthe same kind and do not all operate in the same way. In the
traditional village the organization ofwork according to caste left
little room for individual mobility between occupations, whether
within a lifetime or between the generations. An oilpresser could
not become a carpenter, or a washerman a blacksmith, on
account ofcustomary sanctions against the change ofoccupation
which operated for centuries and until quite recent times. These
 Equality as a right and as apolicy 203
sanctions no longer enjoy the support of law but it would still
be regarded as unusual for people to wish to change their
occupation within the context of their own village. To change
their occupation they have to move to a different context, and
this is by no means an easy thing to do.
A certain amount of movement was possible even in the
traditional context, but it was limited in various ways. Those
who were classified as untouchables were confined to menial and
polluting tasks, and they were strictly excluded from every kind
of superior non-manual occupation. It is difficult to exaggerate
the cumulative effect on the untouchables of the twin processes
of confinement and exclusion. They have put them at a severe
disadvantage not only in material terms but also in terms of
motivation. That kind of disadvantage can scarcely be cancelled
by the mere removal of the legal constraints on occupational
mobility.
Although the new occupations in the office and the factory
are in principle caste free, the association between caste and
occupation is carried over from the traditional to the modern
occupational setting. 19 The various castes are not represented
in the new occupations according to their proportions in the
population. The inferior castes in general, and the Scheduled
Castes in particular, are very thinly represented in the superior
administrative and managerial occupations. In a society which
has adopted the principle of equality of opportunity but whose
members remain acutely conscious of caste distinctions, these
disparities are now a source of anxiety and concern. This
concern finds expression in the policy to redress such dispar-
ities between groups, but the policy in turn impinges on the
equal rights of individuals, particularly their right to equality
of opportunity.
There are several reasons why untouchable and other inferior
castes are so thinly represented in the higher occupations. While
Article 15 prohibits discrimination on grounds ofcaste and Arti-
cle 16 guarantees equality of opportunity in public employment,
there is, in fact, widespread prejudice against the inferior castes
in general and the untouchables in particular. This prejudice is
not easy to measure and by its nature it is difficult to establish in
the individual case. It operates more actively at the lower levels
204 Society and Politics in India

of employment where recruitment, tenure and promotion are to

a large .extent personalized than perhaps at the higher levels
where these processes are organized in a more impersonal way.
But there is reason to believe that some candidates are at every
level rejected on account of prejudice even when they have the
necessary qualifications.
There is, in addition, a marked decline in the number of
qualified candidates as we move down from the upper to the
lower levels of the caste hierarchy. This is a result of poor
material conditions, and more of past than of current social
prejudice, or of a kind of prejudice for which it is impossible
to assign individual responsibility. It has been an objective
of social policy to expand the pool of qualified candidates
among the disadvantaged sections of society through freeships,
scholarships and other educational inputs. This objective finds
its justification in the argument that equality of opportunity
depends 'not merely on the absence of disabilities, but on the
presence of abilities'. 20
Although the pool ofqualified candidates among the untouch-
ables and other inferior castes has expanded somewhat, it
has now become apparent that this expansion is a slow and
long-drawn process. Even while it takes its course, special
measures may be adopted for recruiting more members of
these disadvantaged groups into higher occupations by relaxing
the conditions of appointment in their favour. Apolicy ofpositive
discrimination has, in fact, been extensively used for increasing
the numbers of persons belonging to inferior castes in public
employment. It is at this point that the tension between equality
as a policy and as a right is most clearly revealed, for it
is difficult to create and maintain on a large scale special
opportunities for some without taking something away from the
equal opportunities of all.
Article 16 of the Constitution on equality of opportunity
in public employment contains a clause by which the state
is enabled to make special provisions 'for the reservation of
appointments or ·posts in favour of any backward class of
citizens'. Further, Article 335 stipulates that the claims of
members of the Scheduled Castes and Scheduled Tribes are to
be 'taken into consideration, consistently with the maintenance
 Equality as a right and as a policy 205
of efficiency of administration, in the making ofappointments to
services and posts in connection with the affairs of the Union or
of a State'. Article 15, which prohibits discrimination on grounds
of religion, race, caste, sex or place of birth, incorporated
through the First Amendment a clause by which the state is
enabled to make special provisions 'for the advancement of any
socially and educationally backward class of citizens or for the
Scheduled Castes and the Scheduled Tribes'.
It should be clear that positive discrimination in India differs
in scale, if not in principle, from affirmative action in the United
States. There is a great deal of disagreement in India, leading
sometimes to violence, about how the policy should be made
to work and how far it should be extended, but the principle
behind the policy is widely accepted as an important part of
the commitment to equality. In the United States there appears
to be far greater ambivalence about accepting the principle
itself. This, no doubt, is partly because of the distinctively
American conception of equality which, as Professor Fiss has
pointed out, is based on the 'anti-discrimination' rather than the
'group-disadvantaging' principle. 21 In India, where collective
identities are so much more marked, disparities between groups
figure prominently where equality is a consideration.
American courts, even while accepting affirmative action, have
been hostile to racial or ethnic quotas. 22 In India quotas or
reservations are accepted by all branches of government as a
necessary part of a policy of greater equality overall. It must
be pointed out that, as far as employment is concerned, the
Constitution has only enabling provisions for reservation, but
caste quotas in employment have been created extensively by
legislative and executive action, and the courts have in general
accepted them. 23 Caste quotas in admission to educational
institutions, particularly to medical and engineering colleges,
have also been accepted as a part of the programme of positive
discrimination.
Caste quotas in education and employment were introduced
during British rule and, in some parts of the country, all
positions in certain sectors of employment and education were
reserved according to caste and community. These quotas were
designed more in the interest of political balance than of social
206 Sodety and Politics in India
equality. Comprehensive caste quotas have been struck down by
the courts as being against the Constitutional provisions for the
right to equality. 24 Caste quotas are now admitted only if they
can be shown to be required by a policy of equality, not by
a policy of balance. This means that there can be quotas for
inferior but not superior castes.
Qµotas in education and employment are everywhere allowed
in favour of the untouchables - and also the tribals - because of
their marginal position in traditional Indian society. Such quotas
have also been extended, though not uniformly for the different
parts of the country, to various other castes and communities
on the ground that they too are disadvantaged or backward. It
has, indeed, been argued that such groups have developed a
vested interest in remaining 'backward', and political pressure
is often used by them to get themselves so classified. It is at
this point that a policy of greater equality in the long run yields
to calculations of immediate political advantage.

As I have said, the Constitution has enabling provisions for

reservation on the basis of caste in education and employment
where such reservation serves the policy of greater social
equality. Further, reservation has re~eived the support of the
legislature, the executive and the judiciary although the support
has not been equally enthusiastic or equally consistent in the
three cases. There has also been public opposition to caste
quotas in education and employment. The grounds ofopposition
are various and it is not easy to separate them from each other.
Some - perhaps the silent majority- are opposed to the very idea
of greater social equality. Others may favour greater equality but
feel that this cannot or should not be brought into being by
state intervention. Yet others may feel that, while there should
be a policy of greater social equality, positive discrimination in
general and job reservation in particular are not good policies.
It is a remarkable fact that in India, where a hierarchical way
of life was created very early, and maintained and justified for
centuries, there is no public defence of hierarchy today. To
be sure, people practise inequality extensively and justify their
practice in private, but anyone who speaks against equality _in
public is bound to lose his audience. In independent India the
 Equality as a right and as a policy 207
language of equality has caught the imagination of not only
politicians and professors but also judges and civil servants.
Granted that it is difficult to determine the extent of genuine
concern for equality, it is likely that those who are in favour of
it will be, on the whole, inclined to favour an active policy of
greater social equality. Those who speak the language ofequality
cannot but be struck by the wide disparities in social conditions
between Brahmans and Harijans, landowners and landless, or
men and women. They have seen some of these disparities
being reduced in their own lifetime and are likely to believe
that much more can be done by concerted and co-ordinated
effort. These beliefs and sentiments are commonly held by those
who are closest to the centres of planning and policy-making.
It is true that there has been a certain loss of faith in planning
and policy-making in India, but that has not been replaced by
an increase of faith in the individual's capacity to change his
circumstances by his own unaided effort.
Those who are in favour of a policy of equality are not all
equally in favour of positive discrimination. Some of them feel
that positive discrimination contributes little if anything to social
equality and has serious negative consequences, while others
feel that it is at best a weak instrument of social transformation.
These latter argue that a policy of greater equality overall will
be much better served by agrarian reform, or the eradication of
.poverty, or the removal of illiteracy. They maintain that a great
deal more can and should be done for the disadvantaged, but
that the disadvantaged ought to be defined by rational economic
criteria and not by caste.
Even those who feel that all untouchables deserve special
attention as against, say, all landless labourers, may not wish to
put the main emphasis on reservation in public employment or
higher education. In other words, it may be possible to conceive
ofpositive discrimination in somewhat broader terms than those
of reservation, by linking affirmative programmes more closely
with the removal ofdisabilities. An example would be the supply
ofdrinking water, on a preferential basis, to untouchable hamlets .
in the rural areas. Other facilities, such as primary education,
primary health care, house sites, etc., may also be made available
to untouchables and tribals on a preferential basis. There is
208 Society and Politics in India
clearly some discrimination in diverting scarce resources in
favour of such groups and to the possible detriment of very poor
or destitute individuals not ~elonging to them, but its negative
effect may be offset by the wide diffusion of basic facilities to
millions of people who have strong claims on them.
Those who favour reservation in the narrow sense would
certainly concede that there are costs in it, but they would
argue that its social costs are in the aggregate outweighed by
its benefits in the form of greater equality overall. It is difficult to
measure these costs and benefits, and they are perhaps perceived
differently in different societies, depending upon their traditions
and circumstances. If I might hazard a quick generalization, I
would say that in the United States the costs are weighed largely
in terms of individual rights, whereas in India they tend to be
weighed more in terms of national unity.
In an article on the Bakke case, Professor Dworkin adverted
in passing to the argument, which he dismissed, that affirmative
action might lead to the balkanization of the United States. 25
Problems of 'balkanization', 'separatism' and disunity do not
present themselves with the same urgency before every nation
or in every phase in a nation's life. America in the 1980s
can perhaps take a somewhat distant view of the threat of
balkanization. In contemporary India, on the other hand, what
is called 'national integration' is a major policy preoccupation.
To the extent that caste quotas appear as a threat to national
integration, they are bound to be counted on the debit side.
Whether caste quotas become a threat to national unity or
not will depend upon the scale on which they are adopted.
Moreover, the political auspices under which they are intro-
duced and extended also affect people's perception of their
desirability. Caste and communal quotas were first introduced,
and in some areas on a very comprehensive scale, by the
colonial administration in India, and they were perceived by the
nationalists, whether rightly or wrongly, as divisive in intention.
Immediately after independence, when the new Constitution
was being written, caste quotas acquired a bad odour, and some
of the existing ones were struck down by the courts.
What is turning once more into a source of anxiety is not
simply the fact of reservation but its scale. While caste quotas
 Equality as a right and as a policy 209
in selected occupations in favour of untouchables and tribals
might be tolerated if not supported by many, the extension of
such quotas to large numbers of other castes and communities
acquires a different political complexion. It is one thing to
offer special concessions to groups because they are socially
stigmatized but quite another to have to yield such concessions
to them because they are politically powerful. The culmination
of that process is the creation of preferential policies by powerful
majorities for their own advantage. This has happened in India
in regard to language and locality, but on a smaller scale than in
some other countries such as Malaysia. 26
While defending reverse discrimination in America, Professor
Dworkin argued that DeFunis (or Bakke) had nQ Constitutional
right to a legal (or medical) education of a certain quality
that could be said to be violated by a policy of preferential
admissions. He would go further and argue that DeFunis did not
have even the right to equal treatment in the matter ofallocation
ofplaces in law school, although he did have a right to treatment
as an equal. Dworkin would make a distinction between benefits
like legal.(or medical) education which can be provided to only a
few and those like elementary education which can be provided
to all and to which, therefore, the right to equal treatment might
be admitted.
DeFunis could seek remedy in the courts only under the
equal protection clause introduced into the United States
Constitution by the Fourteenth Amendment. As Dworkin has
pointed out, that Constitution 'does not condemn racial classi-
fication directly',27 nor does it provide specifically for equality of
opportunity.
The Indian Constitution differs from the United States in
both respects. It directly prohibits discrimination 'on grounds
of religion, race, caste, sex or place of birth' (Article 15); and
it has a clause that guarantees equality of opportunity (Article
16) in addition to the one that guarantees equal protection of
the laws (Article 14). On the latter, it has been pointed out
that 'the Constitution envisages the right to equal opportunity
as an independent right and not as a subsidiary right flowing
from Article 14 or as an incident or amplification of it'. 28
It is true that equality of opportunity is guaranteed only in
210 Society and Politics in India
public employment and that the prohibition of discrimination
is qualified in a number of ways, but the provisions are in both
cases explicit.
The Indian Constitution, unlike the United States, was
designed in full awareness of the negative as well as the
positive implications of preferential policies in education and
employment. The subject was debated extensively in the Con-
stituent Assembly and elsewhere in the country. Dr Ambedkar,
regarded widely as the architect of the Indian Constitution,
pleaded for some quotas as well as for equality of opportunity,
and maintained that the former could, up to a point, be
accommodated within the latter. He was a strong advocate of
quotas for the Scheduled Castes and Scheduled Tribes, but
warned that extensive quotas would 'eat up' the very principle
of equality of opportunity.29
In India thousands of qualified candidates are every year
rejected for admission to medical and engineering college and
appointment to government service because they have to make
room for other candidates, apparently less qualified than they,
in whose favour caste quotas operate. Can we say that the
rejected candidate has no ground for complaint, even when he
is apparently better qualified, because he has no constitutional
right to the place he has been denied and that the right to equal
treatment does not apply in such matters? To be sure, he has still
the right to treatment as an equal, and that right may be saved
by treating his interest with concern and sympathy even while
rejecting him. But what would it mean, in such circumstances,
to treat someone's interest with concern and sympathy?
It may certainly be argued that those rejected are only appar-
ently better qualified and that candidates from inferior castes,
though they perform badly in written tests and interviews, are
in a deeper sense better qualified because of the range of their
social experience, their sensitivity to social issues, their capacity
to withstand adversity, and so on. While this may possibly be
true in the aggregate, it is impossible to establish the facts in
the case of each individual candidate accepted or rejected, for
in the circumstances in which such selections are made, the
law of large numbers prevails. At any rate, no serious attempt
is made to weigh and assess these additional criteria as against
 Equality as a right and as a policy 211
the ones ordinarily in use. Hence, the individual candidate who
is rejected, despite better scores, inevitably feels that he has
been passed over to make room for another on account of caste
and not some definable individual ability that he lacks and the
other has.
Where caste quotas come to be extensively used, definable
individual ability will tend to be at a discount. This may not
be a bad thing for countries that have long suffered from
the Protestant ethic, but its implications will be otherwise for
countries like India which have a different cultural tradition.
There, in view of the long tradition of the subordination
of the individual to family, caste and community, individual
ability needs to be nurtured rather than restrained. It is freely
acknowledged that in public sector employment in India ability
and efficiency count for rather less than they should. While
it would be absurd to attribute this solely to reservation,
caste quotas certainly contribute to the general devaluation of
individual ability.
The urge to smooth out disparities speedily leads the Gov-
ernment -to act at times in ways that not only devalue individual
ability but also discredit its own procedures. Tests of ability for
recruitment or promotion to particular posts are devised, and
then waived or relaxed in favour of candidates belonging to the
Backward Classes. A judge might on occasion seek to rescue the
Government from its embarrassment.
The case of N.M. Thomas provides an interesting insight.
Lower division clerks in a particular branch of the Kerala
government were required to pass within a fixed time certain
departmental tests in order to be eligible for promotion as
upper division clerks. The limit of time for passing the tests
had been extended by successive orders of the Government in
favour of candidates belonging to the Scheduled Castes and
Scheduled Tribes while they were already enjoying the benefits
of promotion. Moreover, as many as thirty-four out of a total of
fifty-one vacancies had been filled in this manner. The Kerala
High Court struck down the rule under which the orders were
passed. On appeal the Supreme Court reversed the judgement
on the ground that the extensions were necessary to give fuller
representation to the Scheduled Castes and Scheduled Tribes
212 Society and Politics in India
among upper division clerks. Making light of the qualifications
expected from junior functionaries of the Government, Justice
Krishna Iyer observed, 'After all, here is a pen-pushing clerk,
not a Magistrate, accounts officer, forest officer, sub-registrar,
space-scientist or top administrator or one on whose initiative
the wheels of a department speed up or slow down'. 30 Then,
again, 'And, after all, we are dealing with clerical posts in
the Registration Department where alert quilldriving and a
smattering of special knowledge will make for smoother tum-out
of duties'.31
Where individual ability is widely discounted, individual rights
cannot remain secure. It would be going too far to say that the
individual had no rights in traditional Indian society, but we must
at the same time acknowledge that such rights as he had were
very different in kind from those that the Constitution has sought
to establish. There has been some drift away from the security
originally given to individual rights by successive amendments
to the Constitution. The Twenty-fifth Amendment, enacted in
1972, is regarded as a turning point:
'Prior to the 25th Amendment, the notion was that Fundamental
Rights prevailed over the directive Principles of State Policy.... The
25th Amendment introduced a radical change in the concept, in the
interest of promotion of social justice and egalitarianism envisaged
under the Constitution'.32
In other words, within a quarter-century of being established,
equality as a right began to yield to equality as a policy.
Although the Twenty-fifth Amendment was occasioned by
the need to protect state laws designed for agrarian reform, the
insecurity ofindividual rights is most clearly revealed in the con-
text ofpositive discrimination, for here the individual has to yield
not only before the·state but also before caste and community.
Positive discrimination and the group-disadvantaging principle
on which it is based, bring to our attention the problem of dis-
parities between groups as against that of inequalities between
individuals. It can be shown that increased opportunities for
individual mobility do not lead at once or automatically to the
reduction of disparities between castes or between races. 33
Strong advocates of caste quotas have in that light argued that
 Equality as a right and as a policy 213
equality between castes must first be established before there
can be equality between individuals. 34
It may well be the case that no individual has a Constitutional
right to a place in a medical or an engineering college, or in
a particular service of the Government, but a comprehensive
system ofcaste quotas raises another kind ofquestion. Do castes
and communities have claims to particular shares ofsuch places?
What kinds of claims are these, and how are they to be weighed
against the claims of individuals? These are not speculative
questions because caste quotas are adjusted and re-adjusted
through intense political bargaining in a language whose implicit
assumption is that castes and not just individuals have claims.
The language of bargaining over caste quotas reflects an
important part of the Indian social reality. In the traditional
Indian village rights and obligations were to an extent defined
by caste. The barber, the washerman, the potter, the carpenter
and the blacksmith provided services not merely as individuals
but as members of their respective castes. If a particular barber
or potter was unavailable it was the obligation of the local caste
group to provide a substitute, and each of these groups had
its own rights in the village and its total produce which it
protected from encroachment by individuals from other groups.
The system of quotas introduced by the colonial administration
reinforced, against countervailing tendencies, the belief that
every caste or group of castes could claim a share of what was
at the disposal of the State.
I would not like to leave the impression, while concluding,
that nothing has changed in India, for I do believe that a great
many things have changed. There is widespread antipathy to
the excesses of hierarchy and some genuine sympathy for
equality. But collective identities have withstood this change of
orientation from hierarchy to equality without being substantially
weakened. For that reason, although the concern for equality
may be both genuine and strong, that concern has a somewhat
different orientation in India as compared with the West.
In the distinctively Indian orientation more emphasis is likely
to be given to equality as a policy than to equality as a
right. The Twenty-fifth and the Forty-second Amendments
appear to be signposts in that direction. 35 Indians who regard
214 Society and Politics in India
themselves as forward-looking have been prepared to accept
some restriction of individual rights in the interest of a more
equitable distribution of property. It is not yet clear how far
they will agree to such restriction in the interest ofgreater parity
between castes.
 9 Individualism and equality

The relationship between individualism and equality is a large

subject which social anthropologists have only recently begun
to explore systematically. It has, however, been discussed in a
variety ofcontexts for a century and a half by scholars in different
intellectual disciplines: philosophers, political theorists, lawyers,
historians, economists, and others. Many of the social theorists
of the nineteenth century were struck by the simultaneous
emergence of the desire for equality and the appreciation of the
individual, and they sought to establish a relationship between
the two. Alexis de Tocqueville was one of the first to argue that
individualism and equality were both new values and that they
were inseparably linked in •their origin and development. The
intimate connection between the two was also stressed, though
in other ways, by the jurist Henry Maine and the historian Jacob
Burckhardt.
Individualism, equality, and their relationship have so far been
discussed almost entirely within the context of Western culture.
It is desirable to extend the discussion to cover not only those
societies in which these values were first clearly articulated but
also others to which they have spread and in which they have
found some room for themselves. The social anthropologist can
bring to bear, perhaps a little more fully than others, a compara-
tive perspective on the subject. He is trained not only to look for
differences among cultures but also to appreciate and respect
these differences. It is necessary to ensure, however, that we do
not, out of a false sense of appreciation of or respect for other
cultures, stress the differences beyond their true proportions.
While cultures undoubtedly differ, and differ in important ways,
they are in the modem world also closely interconnected. This
is particularly relevant in the context of equality, for as an ideal
and a value it has acquired a certain appeal in every part of the
modem world.
216 Society and Politics in India
A second important requirement of the approach of social
anthropology or comparative sociology is to study ideas and
values in their concrete social setting, to take into account,
in the language of Durkheim, both collective representations
and social morphology and to relate them to each other. We
need to examine not only the strength and consistency of the
ideal of equality in a particular society but also the institutional
arrangements and the interpersonal relations through which
that ideal is expressed. It is equally necessary, though more
difficult, to do this for individualism, which has so far received
less attention from the sociologist or social anthropologist than
from the historian of ideas (see Dumont, 1983).
Both individualism and equality are large, vague, and ill-
defined ideas. I will not try at this point to define them even
in a preliminary way, but it is necessary to note that each ofthem
can be and has been conceived in more than one sense. Equality
and individualism can be shown to reinforce each other when
conceived in certain ways; when conceived in other ways, they
can be shown to limit or, in the extreme case, even to exclude
each other. Those who argue that equality and individualism are,
as it were, two sides of the same coin generally leave it unsaid
that by stressing other aspects of equality or of individualism
than the ones they stress, it is possible to formulate a very
different kind of relationship between the two.
Many of those who have written about individualism have
acknowledged that it has been conceived in more than one way.
However, they have often adopted a partisan attitude, arguing
that one of the available conceptions - the one to which they
subscribe - is the true conception and some other conception,
which may be equally popular, a false one (see Hayek, 1980
(1946); Durkheim, 1969 (1898)). On closer examination it
might well turn out that a conception of individualism rejected
as false spells out some of the implications and presuppositions
left unstated in the one accepted as true.
Broadly the same thing may be said about equality, of which
more than one conception usually prevails in the same society.
The philosopher or historian of ideas is apt to pick out a
particular conception of equality and to build his argument
from the premise that it is widely endorsed in a given society,
 Individualism and equality 217
ignoring the fact that there are other conceptions of equality,
not unrelated to the first, to which some or many members of
the same society may be indifferent or even hostile.
Thus, defining individualism is one thing, and demonstrating
that it has a central or unique place in the collective conscious-
ness of a given society is another. We need not take at face value
the kind of statement which says that the individual counts for
nothing in some societies and for everything in others. 1 When
we say that individualism is a value in a given society we mean
that its people hold the individual to be as important as, or more
important than, clan, caste, estate, race or nation, and that they
act in ways which enable us to infer that they assign him such
significance. With something as large as individualism, however,
it becomes evident that even in the same society people hold
divergent beliefs and act in divergent ways. At any rate, as Lukes
(1973) has cogently demonstrated, in countries like Britain,
France and Germany, which are recognized as the natural
habitat of individualism, some misgiving about over-emphasis
on the individual has always been e"l)ressed by Catholics,
Conservatives, Socialists, and various others.
One might think that there would be agreement at least in
contemporary Western societies on the value to be placed on
equality as a principle of social life, but that would be far from
the truth. Tocqueville pointed out some of the negative social
consequences of equality even while drawing attention to the
providential nature of its progress. Voices have been raised
against the pursuit of equality, certainly in Britain and also in
other Western countries, throughout the nineteenth century and
in the twentieth. It is true that these voices have been muted or
muffled at times and in pla~es, but they have begun to make
themselves heard again in the last ten or fifteen years on both
sides of the Atlantic (see Letwin, 1983a). It might have caused
Tocqueville some surprise to have learnt that the strongest
arguments against equality were to come in the last decades of
the twentieth century from America.
It is obvious that the commitment to equality - as also the
value assigned to the individual - varies within the same society
between classes, between racial, religious, and linguistic groups,
and betwt!en men and women. While we cannot take all
218 Sodety and Politics in India
these variations into account in every study, we have to be
cautious about characterizing a whole civilization in terms of
a single value or a single pair of values and about making such
characterizations the basis of a radical opposition between whole
civilizations.
The difficulties of identifying the dominant values of a l~e
and complex society are well known to the social anthropologist.
We might take up a nonnative text, such as a constitution, but we
know that the values which people actually live by are at variance,
and sometimes widely at variance, with their constitution. We
might take up representative thinkers, but their selection bristles
with difficulties. Limiting ourselves only to England in the
second half of the nineteenth century, whom should we regard
as a representative thinker on equality? Matthew Arnold (1903),
who castigated his countrymen for their 'religion of inequality'?
Or T.H. Huxley (1890), who, in his essay 'On the natural
inequality of man', argued that they yielded too readily to the
rhetoric of equality? Clearly, we have to adopt an approach which
seeks continually to relate ideas, values and norms to the social
arrangements in and through which people live and work.
Our understanding of many ofthese issues has been advanced
considerably by the writings of Louis Dumont (1977; 1983).
The point of departure of Dumont's work is the contrast,
at the level of values, between societies governed by holism
and hierarchy on the one hand, and by individualism and
equality on the other (Dumont, 1972). Dumont's work has
had a marked impact on Indian studies, since a great deal of
what he has written in his books and in the journal founded
by him, Contributions to Indian Sociology, has been about India.
Despite the force and subtlety of this work, it has certain
limitations. It has concentrated on the structure of traditional
Indian society and on holism and hierarchy in it, leaving
largely unconsidered many of the issues that are of concern
in India today. Whatever might have been the emphasis of
traditional Indian culture, both equality and the individual are
central concerns in the contemporary constitutional and legal
systems; and it is impossible to understand what is happening
in India today without taking into account Constitution, law, and
politics.
 Individualism and equality 219
My own interest in these questions derives less from a
preoccupation with speculative theory than from a concern
with the realities of society and politics in contemporary India.
More specifically, it has grown out of a consideration of the
problems of the Backward Classes and the presuppositions and
implications of the policy of positive discrimination. Positive
discrimination challenges certain common assumptions about
equality, for it sets out to achieve equality by not only taking
collective identities into account but assigning them a certain
pre-eminence over individual identities. And these questions
about equality and about individuals and collectivities have a
certain comparative significance, for they now arise not only in
India but also elsewhere, notably in the United States.

The assumption of a relationship between individualism and

equality has been since the nineteenth century a part of the
collective wisdom of a large section ofthe Western intelligentsia.
This is not to say that everybody accepts that assumption; it is
nevertheless widely acknowledged. The work of Tocqueville
(1956 (1835-40)), which was a landmark in this regard, has
influenced generations ofwriters on the subject. 'Individualism',
he wrote, 'is a novel expression, to which a novel idea has
given birth' (vol. 2, p. 98). And he added, 'individualism is of
democratic origin, and it threatens to spread in the same ratio
as the equality of condition'. Tocqueville himself was too close
to the values ofan earlier age and had too finely tuned a mind to
view the growth and spread of individualism - or of the passion
for equality - with unmixed pleasure.
In the work of Burckhardt (1954 (1860)) we are taken
back to an earlier period of European history for tracing the
origins ofindividualism, that of the Renaissance in fifteenth- and
sixteenth-century Italy. What Burckhardt emphasized was not
so much 'equality of condition' in Tocqueville's sense as a
new appreciation of quality, excellence and distinction wherever
found in individual human beings, irrespective of birth or
estate. He too was struck by the relationship between this
new individualism and what he called the 'equality of classes'
(see especially pp. 265-72), but it must be pointed out, as I
have done elsewhere (Beteille, 1984), that his ideas about social
220 Society and Politics in India
equality are vague and inconsistent.
Here a word of caution needs to be introduced, because how
far back we can go in tracing the origins of individualism or of
equality will depend on how strictly or broadly we define these
terms. Historians have noted the presence of individualism,
social mobility, and the spirit of equality in pre-modem times
(Macfarlane, 1978), and anthropologists have recorded evidence
of all three in the simplest of tribal societies (Malinowski, 1926,
Beteille, 1984). It is well to remember what Maitland said in
his monumental work on English law in the Middle Ages: 'we
may have seen enough to give us pause before we assent to any
grand dogma which would make "communalism" older than
"individualism". The apparent communalism of old law covers
an individualism which has deep and ancient roots (Pollock and
Maitland 1968 (1895), p. 688).
Dogmas aside, we cannot fail to notice certain basic differ-
ences in legal ideas and concepts between past and present
societies. In some societies the caste, the clan, or the family
overshadows the individual to a large extent. In others the
individual is given a kind of pre-eminence as the bearer of rights
and capacities, as citizen, as voter, as earner, and as consumer.
There are specific social and legal arrangements under which
the autonomy of the individual and respect for him are fostered
and promoted. Maitland's senior contemporary, Maine (1950
(1861); 1914 (1875)), showed how the emancipation of the
individual from the family and other such groups was historically
linked with the progress of equality, including equality between
men and women.
These and similar ideas have been brought together in
Dumont's impressive body of work. The guiding thread of
this work is the contrast between hierarchical and egalitarian
societies, between homo hierarchicus and homo aequalis. For
Dumont, hierarchy is the social expression of a fundamental
value which he calls 'holism', and equality is the expression
of its opposite, individualism. His work on India (1972), which
provides a detailed analysis of hierarchy, promised a subsequent
and complementary work on Europe, to be devoted to equality.
His work on Europe (1977), though it bore the title Homo
aequalis, 2 turned out to be more on individualism than on
 Individualism and equality 221
equality. There need not be any objection to this, provided
we accept the premise that individualism entails equality; but
it is the premise itself that I would like to submit to examination
and criticism.
Dumont's contrast between homo hierarchicus and homo aequalis
may be seen as a reformulation ofTocqueville's contrast between
aristocratic and democratic societies (Beteille, 1979), but there
are important differences. Dumont's interest is in contrasting
two different societies, separated in space and having .distinct
historical traditions. Although his work shows both imagination
and insight, it lacks the historical spirit which animates the
work of Tocqueville. While Tocqueville pushed hard the
contrast between democratic society and aristocratic society,
his fundamental concern was to show how the one grew out
of the other. Dumont's preoccupation is with difference, whereas
Tocqueville takes pains to establish continuity.
A study of the relations, whether of opposition or agreement,
between hierarchy, equality, the individual, and the group can-
not be made without a historical awareness of the changes taking
place throughout the modem world. Basic legal and political
ideas are no longer insulated within their original matrix in the
way in which they were in the Middle Ages or earlier. Societies
and cultures which were different and separate until a century
or two ago have become mutually implicated as never before.
The contemporary social significance of ideas and values has
to be considered separately from their historical origin, however
important the latter might be. It cannot be argued - certainly not
in the modem world - that ideas and values that have been
current for merely four or five generations are necessarily less
significant or effective than those which held the field for a
millennium or more.
Nor is it a question of the passage from one part of the
world to another of only ideas and values. A whole range
of new institutions - offices, factories, trade unions, political
parties, schools, colleges, universities - has arisen in societies
in which until very recently hierarchical values and collective
identities prevailed. These institutions cannot even begin to
work unless some attention is paid to the claims of individuals
for equal treatment. Conversely, collective identities have shown
222 Society and Politics in India

remarkable durability in societies in which both equality and

individualism have long been established as fundamental values.
Equality, individual rights, and the claims ofcollectivities are all
part of the language of contemporary politics, and this language
has now acquired a certain universal currency.

The difficulty of arriving at a clear judgement on these issues

becomes apparent as soon as we begin to look into a living
society in the contemporary world. The most striking feature
of Indian society today is the co-existence of divergent, even
contradictory, beliefs and values. Hierarchical values are in
evidence everywhere; yet people proclaim loudly, and not
always insincerely, that equality should be placed above every
other consideration. Individuals compete with each other and
claim their dues as individuals in a growing number of fields:
yet loyalty to caste, tribe, sect, clan, lineage, and family have a
continuing, and in some fields an increasing, hold over people.
A convenient place to begin an enquiry into these fundamental
concepts is the Constitution of India. What the Constitution
actually signifies for the different sections of Indian society -
educated and uneducated, urban and rural, men and women,
Hindus, Muslims, and Christians, the high- and the low-born
- is a separate question that cannot be examined here. All
that we need to acknowledge at this point is that it signifies
something for at least some sections of Indian society. If there
is an overall design in the Constitution, that design may be said
to put equality in the place of hierarchy and the individual in the
place of caste. Hierarchical values are repudiated, and the com-
mitment to equality is strongly asserted; but the repudiation of
collective identities of the kind on which the traditional hierarchy
rested is not as clear as the repudiation of hierarchy itself.
In the Indian Constitution the individual is the principal,
though not the sole, bearer of rights and responsibilities, and
citizenship is an unmediated relationship between the individual
and the state. This has to be contrasted with what prevailed in
the past, when the civil status of the individual derived from
his caste or his family. Until 1850 expulsion from caste, at
least for Hindus, who constitute the vast majority of the
population, amounted virtually to civil death. 3 The rights of
 Individualism and equality 223
individual citizens are equal rights, and there is in that sense
a convergence in the idea of citizenship of the values placed on
equality and those placed on the individual.
The right to equality is a part of the Fundamental Rights made
available to all citizens by the Constitution (See Tripathi, 1972).
The three most important among these relate to equality before
the law (Article 14), the prohibition ofdiscrimination on grounds
of religion, race, caste, sex or place of birth (Article 15), and
equality ofopportunity in matters of public employment (Article
16). All these articles bring out the centrality of individual rights,
treating religion, race, caste, sex, etc., as possible impediments
to their full exercise. To settle the issue more fully, Article 17
abolishes untouchability, while the remaining article on equality
in the section on Fundamental Rights attends to an issue ofonly
historical interest, the abolition of titles.
The Constitution, however, does not give attention to the
individual alone. His rights are to some extent qualified and
limited by other considerations. Chief among these, from the
point of view of equality, are the special claims of certain
tribes, castes, and other communities together designated as
the Backward Classes.
Part XVI of the Constitution is devoted to 'Special Provisions
relating to Certain Classes'. Some of these provisions are very
specific, others general or even vague. The most specific
are those relating to political reservations: seats are reserved
in the lower house of Parliament and in the corresponding
houses of the state legislatures for the Scheduled Castes and
Scheduled Tribes, roughly in proportion to their strength in the
population. The constitutional provisions relating to reservation
in the services of the government are vague and general but
none the less important. They have invariably been interpreted
in executive orders and judicial decisions to mean that the
Scheduled Castes and Scheduled Tribes have special claims
in regard to public employment on a par with their claims to
seats in the legislatures.
The Scheduled Castes and Scheduled Tribes, which together
make up 23 .5 per cent of the population, comprise only the most
clearly defined part of the Backward Classes. The Constitution
mentions other backward classes but does not specify what they
224 · Society and Politics in India

are or what provisions are to be made for their benefit. A

large number of states have, however, made special provisions
for them which are in some regards comparable to those in
existence for the Scheduled Castes and Scheduled Tribes.
The courts have upheld some of these provisions, and there
has been a fair amount of political support for them. Indeed,
parties in office have until recently tended to assume that
by making or extending such provisions they can enhance
their electoral appeal. 4 There is also public opposition to the
extension of caste quotas in employment and education to the
Other Backward Classes. 5 It is difficult to infer solely on the
strength of political support for .or opposition to reservation
how strongly Indians belonging to different sections of society
regard as legitimate the respective claims of individuals and of
collectivities. There clearly is some tension if not contradiction
here between two principles: equal opportunities for all on the
basis of citizenship and special opportunities for some on the
basis of caste or community (see Galanter, 1983; Beteille n.d).
The courts have repeatedly pointed to this, and there is growing
awareness of it in public life. 6
I ought to point out that caste or, rather, communal quotas
in employment and education were introduced into some parts
of India during British rule and were already in existence when
the Constitution of 1950 came into effect. These quotas were
bound to come into conflict with the rather strong provisions
for equality of opportunity in the new Constitution, and they
did. Article 16, guaranteeing equality of opportunity in public
employment, had a clause enabling special provisions to be made
in favour ofthe Scheduled Castes and Scheduled Tribes. Article
15, prohibiting all forms of discrimination, had no such safety
clause. Almost immediately after the Constitution came into
force, caste quotas in admission to medical college in the state
of Madras were challenged. The Supreme Court struck down
the standing order of the government of Madras on quotas
on the ground that it was ultra vires of the Constitution
(Galanter, 1983).
The Constitution had to be amended in 1951 to make
reservation in both education and employment legally valid.
This did not mean that caste quotas could no longer be
 Individualism and equality 225
challenged in the courts. They have been challenged over
and over again, but now for violating the spirit rather than
the letter of the Constitution (Galanter, 1983). The courts
have responded differently to reservations for the Scheduled
Castes and Scheduled Tribes and for the Other Backward
Classes. Till around 1962 they were generally inclined to ask
for what may be described as strict scrutiny in regard to caste
quotas for the Other Backward Classes. Since then they have
been somewhat more accommodating towards programmes of
reservation operated by the state governments.
The political movement for selection on the basis of caste
quotas as against open competition has definitely gained strength,
at least in North India since 1977, although more recently there
has been indication of a reaction against it (Government of
India, 1981). The recent agitation over reservations, both for
and against, has revived suspicions that, despite professions to
the contrary, what Indians really care about is caste; and caste,
one hardly needs to point out, is the antitheses of the individual
and of equality.
It is in this light that we must examine the question, often
raised, about the significance to be attached to the commitment
to equality in the Constitution of India. The commitment
is no doubt vigorously expressed, but is it seriously meant?
Indians themselves often say that the talk about equality in the
Constitution, in the law courts, or on political platforms is mainly
rhetoric - that there is no genuine commitment to it at any level
of Indian society. One may point out, in support of this kind of
scepticism, that the Constitution itself and the laws sustain the
basic units of a hierarchical society by acknowledging in a large
number of cases the special claims of castes and communities.
When we survey the politico-jural domain in India since
independence, we are struck by two things: continued assertions
ofthe commitment to equality and continued concessions to col-
lective identities. Two related questions arise out of this: (1) Is
equality possible without the assignment ofsocial pre-eminence
to individual rights? (2) Is equality possible where collective
identities appear so compelling? These are both important
questions, but they cannot be seriously considered so long as
we adhere to the dogma of the inseparability of equality and
226 Society and Politics in India
individualism. We shall see in what follows that the relationship
between the two is a complex one: they reinforce each other in
some respects, but in other respects their paths diverge.

I would like to argue that, no matter how deeply loyal he may be

to his caste or community, the educated Indian today also values
Article 15 of the Constitution, which prohibits discrimination
on grounds of religion, race, caste, etc., and Article 16, which
guarantees equality of opportunity in public employment irre-
spective of those considerations. He does so not only because
of his direct exposure to Western or modem values but also
because of the compulsions of the social arrangements by which
he has to live and work. As more and more Indians make use of
the new educational system and the new occupational system,
more and more of them begin to appreciate that, no matter
what the proper considerations might be in the village or the
home, considerations of caste and community ought not to
prevail, under normal conditions, in the school, the college,
the university, the office, or the factory. I am not saying that
they believe that these considerations do not in fact prevail, only
that, ideally, they ought not to prevail.
Evidence of a shift in people's perceptions of right and wrong
ways of doing things may be found in the recent literature on
factory employment in various parts of the country. People know,
of course, that jobs are secured in all sorts of ways, among them
by using influence, including the ties of kinship and caste, al)d
even bribery, but there is a sense in which they also feel that the
best way of obtaining a job is 'on merit'. Holmstrom (1984, p.
210) writes about selection on merit on the basis ofhis fieldwork
among factory workers in Bangalore: 'Workers and managers
talk as if it is the morally superior way to find work: workers
who got their jobs on "merit" are proud of it, those who used
influence are on the defensive'. And again, ' "merit" is an ideal
standard against which to measure the other ways people get
jobs'. All this represents a shift in the never stable balance
'between values ofhierarchy and equality, or between fixed roles
and competitive individualism' (p. 283).
It is important to note the institutional accompaniments of
equality of opportunity. Equality of opportunity - which rep-
 Individualism and equality 227
resents equality in one sense and not in every sense - acquires
its full significance in the modem occupational and educational
systems. Equality of opportunity would signify very little in a
tnoal community or even a peasant community, although such
a community might in other respects be markedly 'egalitarian'.
Indians today can no longer regard the modem educational
system or the modem occupational system in the same way
in which their forebears thought of traditional learning or of
traditional. crafts and trades. That every individual - or at least
every male - should have,·irrespective of caste or community,
an opportunity to go to school or enter government service is
now widely, though not universally, acknowledged as a principle.
Why this does not happen and what can be done to make it
happen in practice are questions that have to be considered
separately.
The idea of treating individuals on merit is in some sense a
new one, but we must not exaggerate its novelty. No matter
how disguised, it may be recognized in every major religious
tradition, Hinduism not excepted. It is true that the old social
practice emphasized the distinctions of caste and community,
but the requirements of a new social practice might give new
life to the moral claims of individuals as individuals. This
can be clearly seen in the writings of a large number of
nineteenth-century Indian intellectuals, who began to take a
keen interest in equality (Ganguli, 1975).
The claim that in public life individuals be treated on merit,
irrespective of outward differences, was made early in modem
India in relation to race. 7 The British brought with them a new
social theory based on equality, but colonial social practice was
often markedly at variance with this theory. Indians in privileged
positions in the traditional order, who discriminated extensively
against their fellow Indians on the basis of caste, were irked
by being discriminated against on the basis of race. They
experienced or imagined such discrimination in every sphere
of public life: in the civil service, in the law courts, and in
virtually every profession. A new Indian intelligentsia, not very
different in ·its social composition from the old, came into being
in the nineteenth century and established an important place
for itself in society. It became a matter of principle with this
228 Sodety and Politics in India
intelligentsia to demand equality of opportunity, for Indians and
Europeans alike, in the civil service and in the professions. Entry
into the Indian Civil Service, which represented the pinnacle of
prestige and power, became a key issue: the Indians did their
best to secure equality of opportunity, and the British did their
worst to subvert it. 8 It would hardly appear reasonable on the
part of these Indians to seek to repudiate the distinctions of
race if at the same time they sought to uphold the distinctions
of caste. Precisely this point was made in an essay on self-rule
and alien rule by one of the leading intellectuals of nineteenth
century India, Bankimchandra, who also published a tract on
equality (Bankimchandra, n.d. (1382 Bengali calendar)).
The most conspicuous impediment to equality of opportunity
is the presence of legal disabilities based on race, caste and
gender. In India the disabilities of the inferior castes and of
women were very many and of very long standing (Sivaramayya,
1983). These legal disabilities were gradually whittled down over
a hundred-year period and finally abolished under the new Con-
stitution of 1950. However, a great deal ofdiscrimination against
the lower castes and against women, enjoying the sanction of
custom, if not law, survived the legal abolition of disabilities.
This too was an impediment to equality of opportunity and seen
as such. In the villages, where traditional occupations prevailed,
the question of equality of opportunity hardly arose. It did arise
for those who sought to leave the village and its traditional
occupational setup to enter the new world of schools, colleges,
universities, offices, and factories, and more and more people
wanted to do precisely that.
It is not surprising in this light that B.R. Ambedkar, aprincipal
architect of the present Constitution and himself an untouchable
and a lawyer, should have argued strongly in the Constituent
Assembly for making the individual the basic unit in the new
scheme of things (Constituent Assembly Debates 1950, p. 39).
In arguing for the individual, Ambedkar was arguing for equality
and against the constraints of caste and community. There were
some in the Constituent Assembly who advocated making the
village the basic unit, but he resolutely opposed them on the
ground that the village was the seat of caste and faction and
would allow little room for individuals to develop their abilities.
 Individualism and equality 229
Ambedkar was not the only one who saw the link between
equality and the primacy of the individual. Others also recog-
nized individual mobility as a good thing in itselfand as a solvent
of the traditional barriers of caste and community. But how far
could one count on individual mobility alone to transform the
very structure of a hierarchical society? It has to be stressed that
what Ambedkar and others like him wanted was a change not
merely in the positions of some individuals but in the structure
ofsociety itself. Asociety which permits some individual mobility
may at the same time tolerate large social disparities.
The plain fact is that, despite equality of opportunity and
despite individual mobility, there are large social disparities
between upper castes and lower castes in India, between
whites and blacks in the United States, and, to some extent,
between men and women everywhere. What is obvious to us
today and was not as obvious a hundred years ago is that by
ensuring equality of opportunity between individuals we do not
immediately or automatically reduce disparities between groups.
Formal equality of opportunity combined with free competition
between individuals may, in fact, lead to an increase rather than a
decrease in such disparities. It may no doubt be argued that what
leads to the reduction ofinequalities between individuals will, in
the long run, lead also to the reduction of disparities between
groups, but there is little reason to believe that this will happen
on its own in the short run. And in the modem world, members
of disadvantaged groups - blacks, untouchables, perhaps also
women - have lost the taste for enjoying the gift of equality
posthumously.
It would thus appear that certain collective identities may have
to be acknowledged or even emphasized as a part of the concern
for equality. This·is seen in the United States in the context of
affirmative discrimination in favour of blacks and certain other
minorities. It is true that the American courts have in general
treated racial classifications as suspect, but such classifications
have also been admitted in a number ofimportant cases in which
the issue has been the removal or reduction of social disparities
(see Rossum, 1980). Fiss (1977) has criticized the American
courts for adhering to too narrow a conception of equality, one
that takes into account only discrimination against individuals,
230 Society and Politics in India

when a more adequate conception would take into account also

the disadvantages of groups.
In India affirmative or positive discrimination has been written
into the Constitution itself, both in Part IV on Directive Prin-
ciples of State Policy and in Part XVI on Special Provisions
relating to Certain Classes. The 'classes' in question comprise
a variety of groups with pre-existing collective identities, mainly
the tribals and the untouchables. Ambedkar, who spoke strongly
in· support of the individual in the Constituent Assembly,
pleaded also for the recognition of the special claims of c~rtain
groups. What was at issue was not simply equality as a right
available to all individuals but also equality as a policy aimed
at bringing about certain changes in the structure of society (see
Dworkin, 1977).
It is far from my intention to argue that the claims to equality
of individuals and of collectivities are nicely balanced in the
Constitution of India or that they can be nicely balanced. Law
and politics in India have, in fact, been bedevilled by these
conflicting claims ever since the Constitution came into effect.
It cannot be too strongly emphasized that the Constitution treats
the provisions in favour ofthe Scheduled Castes and Scheduled
Tribes as special provisions. Ainbedkar himself argued in the
Constituent Assembly that these special provisions should not
be allowed to 'eat up' the general provisions for equality of
opportunity for all individuals alike (Constituent Assembly
Debates, 1950, pp. 701-2). These special provisions continue to
be in
force, and it cannot be argued that, because they take collective
identities into account and perhaps even strengthen them, they
are by definition hostile to the spirit of equality.

I myself have misgivings about the extent to which the ends of

equality can be served by stressing collective identities of the
kind based on race and caste (Beteille, 1981), but it is not
because I believe that those ends are secured by a single-minded
pursuit of individualism. I would now like to turn the coin
over and examine, on its other side, the disjunction between
individualism and equality.
Some of the most penetrating insights into the relationship
between individualism and equality we owe to the writings of
 Individualism and equality 231
SiIDinel (1971, 1950 (1917)). Simmel saw quite clearly that
individualism might lead to either an appreciation of human
equality or a preoccupation with the inequality of man, and
he distinguished between what he called the 'individualism
of equality' and the 'individualism of inequality' (1971, pp.
251-93). It is-not enough merely to distinguish the two kinds
of individualism, for one might go on to argue that they are
different and separate and are mistakenly given the same name
(see, for example, Hayek, 1980 (1946); Durkheim, 1969 (1898)).
It is a part of SiIDinel's argument, unlike, for instance, that
of Durkheim on the same subject, that the two kinds of
individualism are historically related and products of the same
social process.
According to SiIDinel, the individualism ofequality came into
its own in Europe in the eighteenth century. This kind of
individualism, as Durkheim too had pointed out, found its most
complete expression in the moral philosophy ofKant: 'treat each
man as an end in himself, and never as a means only' (Williams,
1962, p. 17). What was important in this conception was the
claim of moral autonomy for each individual and his freedom
from the constraints of estate, guild and church. And every
individual had, as a moral agent, an equal claim to autonomy
and respect. While Simmel was undoubtedly right in drawing
attention to the eighteenth century contribution, these ideas
have much broader appeal. They were echoed in the debates
over citizenship, equal protection of the laws, and equality
of opportunity in the Constituent Assembly in mid-twentieth
century India.
This coupling of individualism and equality was inherently
unstable and bound to come apart as soon as the stress shifted
from what was common to all individuals to what was unique
to each individual. SiIDinel argues that this is what happened in
the West in the passage from the eighteenth to the nineteenth
century. We need not take this periodization at face value. The
stress on the uniqueness, the incomparability, ofindividuals was
already present in the civilization of Renaissance Italy, and it
went with a belief in natural distinctions among men no less than
the belief, stressed by Burckhardt, in human equality.
SiIDinel traces the passage from the individualism of equality
232 Society and Politics in India
to that of inequality in the following terms: 'First there had been
the thorough liberation of the individual from the rusty chains
of guild, birth right, and church. Now, the individual that had
thus become independent also wished to distinguish himself
from other individuals' (Simmel, 1950 (1917), p. 78). To use
the well-worn language of textbook sociology, inequality based
on ascription was, after an interval, replaced by inequality based
on achievement. 'Yet,' as Simmel puts it, 'this new inequality was
posited from within'.
The nineteenth century saw not only the emergence of a
new idea of human inequality but its systematization in a new
scientific theory. Evolutionary biology achieved pre-eminence
around the middle of that century, particularly in Britain, and
it influenced anthropology, sociology, psychology, economics,
and virtually every science whose subject was man. It provided
a new framework for explaining and justifying inequality in the
animal kingdom as well as the social world. In England Galton
(1950 (1869)) and that great popularizer of evolutionary biology,
Huxley (1890), both presented strong arguments in support of
the inequality of man. Huxley even appealed to religion in
opposing the case for equality, arguing (p. 13) that the only
kind of equality admitted by either Judaism or Christianity is
'the equality of men before God - but that is an equality either
of insignificance of or imperfection'.
Every student of anthropology is familiar with the argument,
supported by nineteenth century biology, that the different
human races are unequally endowed. That argument came
into its own in the heyday of colonial rule and slavery. It
did not end, however, with the demise of either slavery or
colonial rule but tends to reappear with unfailing regularity, to
the surprise of many well-meaning anthropologists. One reason
it dies hard is that it is not a separate argument but part of
the general argument about natural inequality (Beteille, 1980).
Human beings are unequally endowed; these inequalities are
more marked.between individuals of different race, but they are
also present between individuals of the same race. The natural
inequality of the sexes is another part of the same general
argument.
The argument about natural inequality is more than a scien-
 Individualism and equality 233
tific argument; it has also an important nonnative component.
Huxley and others like him argued against egalitarianism not
only because they thought it scientifically untenable but also
because it appeared morally flabby. They had great faith in men
of unusual ability, natural geniuses (hereditary or otherwise),
and regarded the idea of levelling as contrary to both nature
and morality. Individualists believe in equality of opportunity,
no doubt, but they also believe in competition and inequality of
reward, and they are strong opponents oflevelling. Ifit is argued
that Stalin too was an opponent of levelling, or urf.llJni/(Jt)ka
(Ossowski, 1963, pp. 112-13), it will only strengthen my point
that 'individualism' appears in many guises to many people.
The moral argument against levelling that has probably the
widest appeal is the one which relates it to envy. 'I have no
respect for the passion for equality', wrote Justice Holmes
(quoted in Hayek, 1960, p. 85), 'which seems to me merely
idealizing envy.' Nozick (1974, p. 239) has recently sought to
establish a relationship between equality, self-esteem, and envy
in somewhat more abstract terms: where there is scarcity of a
desirable object or quality,'the envious man prefers neither one
having it, to the others having it and his not having it'. But
individuals are, in fact, unequal, such individualists would argue,
in their qualities as well as their possessions, and to yield to the
passion for equality would be to reduce humanity to its lowest
coinmon denominator. ·
A society whose individual members are reduced to the lowest
common denominator can only stagnate; it cannot progress.
Whatever affinity there might have been in the past between
individualism and equality, individualism is in the contemporary
world linked more with the idea of progress than with that
.of equality. It has always been a central part of the capitalist
creed that the ultimate source of all progress is the individual.
Different individuals cannot all move forward at the same pace.
This being so, inequality is built into the dynamic structure of
every progressive society, and the suppression of it must sooner
or later bring to a halt the engines of progress (Hayek, 1960,
especially chapter 3).
Individualists who espouse the idea of natural inequality thus
have various arguments against levelling. They say, first, that
234 Society and Politics in India
it is impossible or unworkable. In the economic field they
draw attention to entrepreneurship and in the educational
field to intelligence and try to show how, in the end, men
of ability come out on top in either case despite all the
obstacles placed· in their way, since it is impossible to sup-
press natural talent. It may also be argued against levelling
that it can be maintained only through the arbitrary use of
coercive force. It is not worth paying the price of this, the
argument goes, to maintain a kind of equality that in no
way corresponds to the natural scheme of things. What is
true of levelling in the totalitarian state is true of it, though
in a milder and less visible form, also in the welfare state.
There are, of course, many who would maintain that levelling
is both impossible to achieve and futile to attempt. We thus
come back to a central dogma of libertarian capitalism -
that equality as a policy (though not necessarily as a right)
is bad because it does violence to liberty as well as effi-
ciency.

Dumont's contribution to our understanding of both hierar-

chy and individualism, despite its importance, has ignored
almost completely what may be called, after Simmel, the
individualism of inequality. This is surprising, because it is
that form of individualism which is today most conspicuous in
Dumont's own chosen field of enquiry, the field of economic
doctrine.
What is original about Dumont's work is not that it has
established for the first time the affinity between individualism
and equality, for that affinity, as we have seen, had been
noted by many before him. Its originality lies in the way in
which it has established the affinity within a particular field
of ideas, the field of economic doctrine. There Dumont has
shown how from Mandeville to Marx the rise of economic
doctrine brought into focus the twin concerns for the indi-
vidual and for equality. Now, while there is much to be
said about the affinity between the two in past economic
doctrine, it is the disjunction between them that strikes us
in the economic doctrine of the present. Among contem-
porary economists the strongest enthusiasts for individualism
 Individualism and equality 235
are precisely those who are either indifferent or hostile to
equality.
Contemporary economists who are strong individualists may
not reject equality as such, but they take a narrow view of
it as being co-extensive with equality of opportunity. More-
over, they define equality of opportunity itself in a narrow
way, as 'formal' rather than 'fair' equality of opportunity (see
Rawls, 1973). Now, one may choose to define equality in
such a way as to deprive it of all content and to make the
very idea of it vacuous. It has indeed been argued that the
term 'equality' is misplaced in phrases such as 'equality of
opportunity' or 'equality before the law', for what is at issue
is a principle not of equality but of universality (Letwin,
1983b, p. 46).
Equality of opportunity can be made into a principle of
universality and nothing more by being delinked completely
from all considerations of distribution. The crucial question
about equality today turns on the way in which we view the
relationship between equality of opportunity and distributive
equality. There are those who would seek a balance between
the two, and there are those who would define the first
so as to exclude the second (Beteille, 1986). In a recent
book, two British intellectuals, one of them a member of the
Conservative cabinet, have said, 'Equality of results is itself
the enemy of equality of opportunity' Ooseph and Sumption,
1979, p. 28). It would be quixotic to describe this position as
egalitarian, although the authors certainly take their stand on
individualism.
The most far-reaching argument about the incompatibility
of egalitarianism and individualism has been presented, among
contemporary economists, by F.A. Hayek, although similar
points have been made, if anything even more forcefully,
by Milton Friedman. In an early essay entitled 'Individual-
ism: true and false'. Hayek drew attention to the threat to
true individualism presented by socialism, with its preoccu-
pation with equality and distributive justice. Men are born
unequal, they start from unequal positions: the state should
not intervene to alter all this even for the sake of equality of
opportunity:
236 Society and Politics in India

From .the point of individualism there would not appear to exist

even any justification for making all individuals start on the same
level by preventing them from profiting by advantages which they
have in no way earned, such as being born to parents who are
more intelligent or more conscientious than the average (Hayek,
1980 (1946, p. 31).
Hayek stresses the point that liberty and the rule of law
are bound to produce inequality. We must both treat individ-
uals alike and recognize their disparate abilities, accepting the
inequality of result that follows: 'From the fact that people are
very different it follows that, if we treat them equally, the result
must be inequality in their actual position'. And further, 'The
equality before the law which freedom requires leads to material
inequality' (Hayek, 1960, p. 87). Hayek is strongly opposed
to the imposition of any preconceived pattern of distribution
and, indeed, to any attempt to create such a pattern through
conscious social intervention. He does not object to 'equality as
such' if it emerges as a by-product of unfettered competition
under the rule of law. But this concession to equality does
not amount to very much where it is accompanied, as it is in
Hayek's work and that of others, by a rooted conviction that
natural talents are unequally distributed.
Hayek is an individualist, not an egalitarian. He believes in
equality of opportunity in so far as it is compatible with or
required by the rule of law and no farther. As an individualist,
he is opposed to nationalism and, above all, to socialism, but
it is interesting that his antipathy to the nation and the state
does not extend to every kind of group. True individualism,
he says:
recognizes the family as a legitimate unit as much as the individual;
and the same is true with respect to other groups, such as linguistic
or religious communities, which by their common efforts may
succeed for long periods in preserving for their members material
or moral standards different from those ofthe rest of the population
(1980 (1946), p. 31).
Thus, 'true individualism' may compromise with the family and
up to a point also with ethnicity and caste but never with
state power. This point of view of one of the most influential
 Individualism and equality 237
economists ofour time is markedly at odds with Dumont's views
about both individualism and equality.
Friedman's position is in many ways similar to Hayek's. He
is a strong proponent of equality of opportunity but deeply
mistrustful of distributive equality. In his opinion, taking from
some to give to others in the name ofsocial justice is detrimental
to the freedom and autonomy of the individual. The ends of
distributive justice and those of individual freedom necessarily
collide, and one must choose: 'One cannot be both an egalitar-
ian, in this sense, and a liberal' (Friedman, 1962, p. 193).
It is quite clear that neither Friedman nor Hayek would like
to pass for an egalitarian. They are both strong individualists,
Friedman unequivocally so, and both might have been consid-
ered egalitarians in an earlier age, say, in the age of Mandeville.
An egalitarian today is not simply someone who believes in or
advocates equality of opportunity as admissible under the rule
of law. He must show some concern for distribution and, in
addition, some commibnent to the principle of redress - 'the
principle that undeserved inequalities call for redress' (Rawls
(1973, p. 100).
The concept of equality as understood today - and this is
true both of the United States and of India - incorporates a
component of distributive equality, even at some cost to formal
equality of opportunity. Here it is important to keep in mind
the distinction, to which Dworkin has drawn attention, between
equality as a right and equality as a policy. The principle of
redress, which has become a part of eq~ality as a policy, is
opposed by strong individualists on two grounds. The first,
which is the more general ground, is that it involves some
expansion of state power at the expense ofindividual autonomy.
The second, which relates specifically to positive discrimination
or affirmative action, is that it sacrifices the claims ofindividuals
to those ofcollectivities based on race, caste, and ethnic identity
(Glazer, 1975; Goldman, 1979).
We have to make a distinction between those who 'do not
object to equality as such' provided it is available without cost
and those who are committed to equality as a value. As I have
more than once suggested, equality means many things to many
people, but we can hardly describe equality as a value where
238 Sodety and Politics in India
there is no readiness to forego or sacrifice anything for a fuller
realization ofit. This is not to say that what is sacrificed - liberty,
for instance, or efficiency, Of progress - is itself of no value, for
in that case we would not call it a sacrifice. It is doubtful that
equality as a value in the specific sense has become established
as a universal feature of any modem society, and there is reason
to suggest that the strongest impediments to its establishment
today are presented by those whom we may legitimately describe
as individualists.
If we are to take equality seriously, we must enlarge the
concept of equality. As soon as we do that we recognize that
the different components of the concept do not harmonize well
with each other. We recognize also that the pursuit of equality
limits the attainment of other ends such as those of efficiency,
liberty, and even the self-realization of the individual. This
should not lead us to assert either that the 'central argument
for Equality is a muddle' (Lucas, 1965, p. 299) or that 'equality
of opportunity is a false ideal' (Coleman, 1973, p. 135). It is the
task of social theory to recognize the diversity of human ends
and to understand and interpret the ways in which each society
seeks, according to its own historical circumstances, to reach
a balance between the different ends·it values. Social equality
and individual autonomy may not be as inseparably linked as
they were once believed to be, but it is difficult to see how any
modem society can discard or neglect.either of these two ends.

On 'Individualism and equality': reply to Dumont

'Individualism and equality', along with comments by a number
of authors and my reply to them, was published in Cu"ent
Anthropology in April 1986. About a year later Professor Dumont
sent a separate comment to the journal in which he criticized me for
misrepresenting his view on the relationship between individualism
and equality. He also explained his concept of value and his
approach to the comparative study of values. Below is my reply
to Dumont which was published, together with his comment, in
Cu"ent Anthropology in December 1987.
 Individualism and equality 239
In the Preface to From Mandeville to Mar.r, Dumont recalls how
at the end of Homo hierarchicus he had outlined his new intellec-
tual project, a work on the modem equalitarian type of society.
'I added, somewhat lightheartedly,' he said, 'that this might be
done in a book called Homo aequalis (Dumont, 1977, p. vii). On
reflection he seems to have thought better of this, for he goes on
to say in the very next sentence: 'I have now come to a soberer
assessment of the situation'. This, one might reason, is why he
gave the book the title it bears, for From Mandeville to Mar.r
sounds distinctly soberer than Homo aequalis. That, however,
was not the end of the matter, for he published the same book
in French under the very title he had lightheartedly chosen,
Homo aequalis. The dispassionate reader will perhaps concede
that this alternation between sobriety and lightheartedness on
Dumont's part might give rise to some misunderstanding among
his readers. Why should he choose two completely different
titles for the same book published simultaneously in Chicago
and Paris? Why should he write a book on individualism, call the
book Homo aequalis, and then appear to disclaim responsibility
for the view that individualism entails equality (see the first two
sentences of his second paragraph)?
For all that he may have written in fine print, Dumont does
not like to miss an opportunity to indulge his taste for symmetry.
Homo hierarchicus, Homo aequalis; holism versus individualism;
hierarchy versus equality; India as against the West; others as
opposed to ourselves. Take a system, stand it on its head, and
you get another system. Take a model, tum it around on itself,
and you get another model. This craving for symmetry is far
more than Dumont's personal weakness; it is a disease of a whole
intellectual climate.
One can well understand that after finishing his book on
India Dumont should wish to ·write a book on Europe that
would be symmetrical with it. One cannot reproach him for
having failed to achieve his purpose. That he did fail must
have been apparent to him before it became apparent to others.
He neither acknowledged nor explained his failure but simply
confused the issue by presenting the book as Homo aequalis to
one set of readers and as From Mandeville to Mar.r to another.
Dumont has sought to defend himself - against Srinivas
240 Sodety and Politics in India

- by saying: 'Beteille has chosen to give the terms at issue

a sense different from mine'. That is indeed the case, for
my main concern was to examine the relationship between
individualism and equality ·and not to present an exegesis of
Dumont's work. In examining that relationship I referred to
the views ofseveral scholars - Tocqueville, Simmel, Hayek, and
others - and not just Dumont. Dumont ignores all views that are
not in agreement with his own. To take one example: Hayek is
a major contemporary theorist who has written on identically the
same subject but from a different angle. One will look in vain in
Dumont for any reference to such points of view.
Since I am told by Dumont that I have given my terms a sense
different from his, I would like to know what exactly is the sense
he has given the term 'equality'. It is not simply that he has ignored
the large and controversial literature on equality; he has not even
bothered to define what he means by it. The astonishing thing in
a book called Homo aequalis is that it has so little to say about
equality. The whole subject is disposed of in the first few pages,
and even there all that we get is truisms to the effect that equality
is the opposite ofhierarchy, that modem societies are egalitarian,
that other societies might value equality but not in the same way,
and so on. There is nothing remotely comparable here to the
discussion of hierarchy in Homo hierarchicus. Homo aequalis is
not a book on equality, and a book on individualism cannot
be passed off as a book on equality. because the relationship
between the two is not what Dumont asserts it to be. There is
something fictitious about the symmetryofthe design ofwhich the
book is a part.
There is a large literature on the subject of equality. Anyone
familiar with this literature must know that it is a difficult and
ambiguous concept, signifying different things to different per-
sons. One who asserts so categorically the primacy of equality as
a value in modem societies might take the trouble to examine its
principal meanings to see where they converge and where they
diverge. It is an essential part of my argument that the concept of
equality is inherently ambiguous. Dumont has nothing to say on
this, either in his book or in his rejoinder to my essay. All that he
does is to reassert that equality is the paramount value in modem
as against pre-modem societies. To strengthen his argument he
 Individualism and equality 241
tells us that in Germany at present university professors do not
have two votes as they had in England in the past. Can one be
more banal?
Dumont concedes 'that the identification of paramount values
is of course open to debate and is not supposed to be always
easy'. I am not sure that he realises how much he is giving
away by making that concession - or perhaps he feels that he
has covered himself adequately with phrases like 'supposed to
be' and 'always'? In any case, I would be the last to demand
as a condition for discussing paramount values that such values
should be easy to identify. But where a task is difficult, or even
'supposed to be difficult', should we not discuss these difficulties
candidly? Should we not try to look for a method that will
make the task less difficult? Where has Dumont discussed
the difficulties of formulating a method· for identifying the
paramount values of one's own society? His whole approach is
different. He merely asserts the paramountcy of a certain value
and, when challenged on the point, says that the subject is not
an easy one and is open to debate. Why take as self-evident a
subject on which scholarly opinion is known to be divided?
It has been argued that by denying the hierarchy of values I
deny value itself. This is tendentious. I do not believe that there
is any society in which all values are of equal significance, and I
doubt if any sociologist believes in such nonsense. But that does
not oblige me to agree that all values can be arranged in a single
hierarchy of the kind specified by Dumont. Why should I agree
to this when I find that, on being challenged, Dumont himself
turns out to be shaky about 'the identification of paramount
values'?
I am not impressed by Dumont's fascination for 'hierarchy',
which is of a piece with his craving for symmetry. Values must fit
nicely together. Something that does not fit has no significance
and is therefore not a value. As I have stressed elsewhere,
hierarchy is a theological rather than a sociological concept,
and Dumont argues about value in a manner that is close to
the theological.
Dumont invests what he calls the modem ideology with a
unity which those who live in the modem world are not always
able to find in it. Where he sees unity and harmony, others find
242 Soci.ety and Politics in India
contention and strife not only among individuals and groups but
also among values, including fundamental if not 'paramount'
values. This difference between Dumont and others is partly a
matter of faith and conviction and partly a matter of perspective.
It is easier for the anthropologist and the historian of ideas who
views a system at a distance to discover unity in it than for the
sociologist who views it at close quarters. Those who live in
the modem world can see the difference between a formal
endorsement of the principle of equality and a commitment
to equality as a value even at the cost of other values. Dumont
accuses me of denying value itself because I refuse to accept
his construction of the hierarchy of values. I view the matter
differently. For me something is a value for people only if they
are prepared to forego some other thing to secure it.
Not only the modem ideology as a whole but the idea of
equality itself lacks the unity that Dumont would like to impute
to it. I start with the idea of equality and am struck, like most
modem writers on the subject, by the variety of meanings it has
been made to bear. My paper considers the various meanings
of equality in order to bring to light the tensions between them.
These issues, which I consider to be central to an understanding
of equality as a value, have all been swept under the carpet by
Dumont both in his book and in this rejoinder. How is it possible
to discuss the significance ofequality as a value without so much
as mentioning the various meanings given to the term by both the
proponents and the opponents of equality?
Let us take first that aspect of equality on which there is
likely to be least disagreement: equality before the law. We
can agree that equality should mean at least this and that
equality in this sense is generally valued today even by those
who may be opposed to it in other senses. Furthermore, regard
to equality in this sense is distinctive of modem societies such
that even where women and racial or ethnic minorities were
denied equality before the law, that denial could not survive very
long against the force of the general principle. Finally, equality
in this sense is very closely linked with and perhaps inseparable
from individualism. I have said all this in my paper, and on this
I agree with Dumont even though he might disagree with me.
Then there is equality of opportunity, but here we are
 Individualism and equality 243

already on shifting sands. The idea of careers open to talent

had a certain significance as a pledge of• the commitment to
equality in the transition from feudalism to capitalism, but it
is doubtful that it has the same significance today. It is easy
to see that what Rawls (1973, p. 100) describes as a 'callous
meritocratic society' can be at the same time fully committed
to equality before the law and formal equality of opportunity.
Could we describe equality as the paramount value of a callous
meritocratic society? The question is not easy to answer, and it
seems not to have occurred to Dumont to ask it.
Despite formal equality - equality before the law and careers
open to talent - large inequalities in fact continue to exist in
modem societies: I should imagine that the test of equality as
a value would lie in the orientations of people towards these
inequalities. Some accept them as inevitable or even desirable
whereas others maintain that they should be reduced even at
some cost to individual members of society. This difference is
an extremely important part of what Dumont calls the modem
ideology. I see very little justification for confusing these two
very different orientations on the ground that they can both be
related to a positive evaluation of equality in some sense.
If it be agreed that there is a genuine difference of orientation
between those who believe that the inequalities that survive the
institution of formal equality are inevitable and should be left
undisturbed and those who maintain that something should be
done for the redress of such inequalities, then it should be clear
than individualism is associated more with the first than with the
second orientation. This was the basic argument of my essay. I
would have valued Dumont's observations on it, for I regard his
work on individualism very highly, but his zeal for vindicating
himself has prevented him from making any such observation.
The question I have raised is a large one, but it may be
pursued into Dumont's chosen field of investigation, that of
economic doctrine. How were individualism and equality related
as values in the economic doctrine of the eighteenth century?
How are they related in the economic doctrine oftoday? I believe
that Dumont's answer to the first question is substantially true
but that the generalization he draws from it about the whole of
modem ideology is false. Had he paused to consider the second
244 Society and Politics in India
question he might have hesitated to make the generalization,
but then he would not have been able to accomplish in its
full symmetry the task he set for himself at the end of Homo
hierarchicus.
Leaving aside for the moment the question of paramount
value, contemporary economic doctrine is not uniformly egali-
tarian. There is a part of it that is indifferent if not hostile to
substantive as against formal equality. We need not enquire here
whether that part is a large or a small one. What is striking is
that individualism is a marked feature of precisely that part
of contemporary economic doctrine that is most opposed to
substantive equality. There may be as much to learn on this
score from Hayek and Friedman as there is from Bernard de
Mandeville.
Dumont points out that he uses individualism in a sense
different from mine. Here he is on much firmer ground,
because, in contrast to the case of equality, he has taken
some pains to explain what he means by individualism. But
he has been too quick to take offence; my essay was meant
not to denigrate his work but to raise certain questions set
aside by him. The distinction between the two meanings of
individualism - one derived from Kant and the other from
the utilitarians - was pointed out by Durkheim (1969 (1898))
nearly a hundred years ago. What I tried to do in my essay was
to link the second meaning of individualism with the defence of
inequality in certain forms of modem ideology.
In the very last paragraph of his comments Dumont seems
to admit that modem ideology may not after all be made
of whole .cloth, but I would like to repudiate the percep-
tion he seeks to attribute to me 'that modem ideology has
become more complicated through Germany's acculturation
from the end of the 18th century onwards'. My essay had
nothing to do with 'Germany's acculturation' but sought to
explore tensions within modem ideology that may be perceived
in France, England and the United States as well as in
Germany. Dumont is welcome to his own theory about the
peculiarities of the Germans: why should he wish to saddle
me with it?
The argument about 'Germany's acculturation' strikes me
 Individualism and equality 245
as unconvincing. The rise of national .socialism in twentieth-
century Germany is too complex an issue to be settled in terms
of the late emergence of individualism. Individualism may have
emerged late in Germany, but that can hardly be said about
Italy. We have Burckhardt's classic study (1954 (1860)), in
which the development of the individual is placed at the heart
of the ideology of the Renaissance in Italy. Why did fascism
find such a congenial home in Italy in the present century? I am
sure that some ad hoc argument can be found within Dumont's
scheme to explain that as well.
I find the argument that the modernization of Germany has
been marked by 'a synthesis of individualism and holism' rather
lame. What, within Dumont's conceptual framework, would
such a synthesis amount to? First we are told about the radical
heterogeneity of 'individualism' and 'holism' and then, when
something perculiar, or apparently peculiar, turns up, we fall
back upon their 'synthesis'. I should have thought that within
the kind of framework on which Dumont insists there can
be no synthesis of individualism and holism but only their
co-existence. As far as I am concerned, the co-existence of
contradictory values is a common feature of all societies, and
there is nothing peculiar to Germany in this.
Dumont prescribes strict conditions for the study of values
but himself overlooks these very conditions. In his conceptual
universe values do not merely co-exist, they have to be valorized
inter se. He has taken great pains to explain that the central
truth about value (in the singular, and not in the plural) is
not the content of individual values but their arrangement in a
hierarchy (Dumont, 1980b). And has he not castigated me for
denying value itself because of my failure to acknowledge the
hierarchy of values? How does one hierarchize such radically
heterogeneous values as individualism and holism? If I have
understood Dumont's argument about value - and it is not an
easy argument to understand - Germany has values, but not
Value. This seems to be the implication of his stance, although
elsewhere he might be profuse in expressing his warmth and
sympathy for German society and culture.
Dumont says that he has not disparaged India but, in contrast,
'vindicated India in the very aspects that made her looked down
246 Society and Politics in India
upon by many in the West'. I do not wish to question his good
faith, but it seems to me that what he has tried to vindicate is
the world that Indians have left behind, not the one that they
are trying to create. He should not feel too surprised, therefore,
if some Indians do not find his attempts at vindicating India
entirely to their taste. Dumont's own sympathies lie all with
traditional India and hardly at all with modern India. I can see
that he has, out of his deep concern with traditional India, tried
to forge a method for the study of his own society, but it is that
very method that has stood as an obstacle to his understanding
of contemporary India.
Contemporary India is even less amenable than modern
Germany to Dumont's method of study. That method places
not merely values but values in their hierarchical order - i.e.
Value - at its heart and centre. Over and over again Dumont
has contrasted Indian ideology with modern ideology, and I
would say that it is that basic contrast that animates his work.
At the same time, he recognizes that India also exists in the
contemporary world and, in that sense, partakes of modernity.
Surely, hierarchy and holism no longer exist in their complete
and pure forms in contemporary India, and some room has
had to be made there for the basic modern values that are
'compelling for everyone in the present world'. What exactly is
the hierarchical arrangement of all this in contemporary India?
Is such a hierarchical arrangement possible? No matter what
he may say in extenuation, Dumont's conception of value, if
he takes it seriously, must oblige him to regard modern India
as lacking in coherence and unity. ·
Dumont's lack of ease with modern India is writ large in his
work, although it does not shine as brightly as his enthusiasm
for traditional India, which is partly an India of his own
construction. In this construction traditional India is made of
whole cloth. Modern Europe, at least Western Europe - and
with strong reservations about Germany - is also made ofwhole
cloth, though of a somewhat inferior texture. Modern India, in
Dumont's construction, is not made of whole cloth; it is a thing
of shreds and patches.
When Dumont writes about the unique configuration of
values characteristic of modern ideology, India, clearly, is far
 Individualism and equality 247

from his thoughts. India is the exemplar ofholism and hierarchy,

which are at the opposite pole from modem ideology, although,
when pressed to the point, one has to admit that India also
'partakes of modernity'. This makes of modem India a kind
of anomaly that is easily forgotten when basic and fundamental
questions relating to Value or the hierarchy of values are at
issue. Dumont's two basic contrasts are between India and
the West and between pre-modem and modem ideologies.
Modem India does not serve well to illuminate either of these
contrasts.
The anomalpus position of modem India accounts for a
peculiar feature of Dumont's comparative method. He moves
easily and without warning from the contrast between India and
the West to that between pre-modem and modem ideologies.
For him the two contrasts are for all practical purposes the same.
This can be made possible only by bracketing·or discounting
modem India. Dumont cannot but acknowledge the existence
of modem India on the plane of facts, and his few essays on the
subject bear witness to that, but he does not fully admit to its
existence on the plane ofValue. I do not ascribe this to bad faith
but see it as a necessary outcome ofwhat he calls his comparative
method.
Dumont's comparative method is severely constrained by
the symmetry of his own construction. By conscious choice
that method assigns priority to contrast over comparison, to
difference over similarity, and to discontinuity over continuity.
It deals with the varieties of society not as they actually exist but
in so far as they exemplify certain constellations ofvalue in their
pure form. From the viewpoint of this method not all societies
are of the same value: societies are valorized in the very acts of
comparison and contrast.
I am unable to understand what Dumont means by attributing
to me the rejection of the comparative method. If I made my
comparisons with an open mind about both similarities and dif-
.ferences, both continuities and discontinuities, does that mean
that I 'ignore comparison' or 'eschew...the idea of comparison'?
It is absurd to describe my rejection of a rigid and artificial
scheme of contrasts as a denial of the comparative method. I
believe that it is an abuse of that method to highlight difference
 248 Society and Politics in India
and discontinuity by suppressing evidence of similarity and
continuity.
A comparison of Homo hierarchicus with its counterpart From
Mandeville to Mar.r will show that the two books differ widely
not only in subject matter but, what is more germane to my
argument, in type of data and method of analysis. The book
on India is markedly anthropological, and that too in a specific
sense ofthe term: it relies mainly on contemporary ethnographic
data and on insights derived from a reading of classical texts.
The passage from the one to the other, spanning a period of
2,000 years, is smooth, easy, and almost effortless. To be sure,
Dumont does not deny that India has a history, but historical
analysis is quite marginal to the techniques by which he has
identified holism and hierarchy in India. In that sense these
values and Indian civilization itself are made to appear timeless
and ahistorical.
Far from being marginal, the historical method is at the heart
and centre of Dumont's analysis in From Mandeville to Mar.r.
A historical sequence is clearly established, and the data are
carefully arranged to accord with that sequence. From Mandeville
to Mar.r pictures a world on the move, not one in which the
operation of time stands suspended. If the book on India is an
exercise in anthropology, this one is an exercise in the history of
·ideas. In the first book the secondary material is drawn mainly
from studies by ethnographers, in the second from studies of
a quite different kind. It is not as if anthropologists had not
made ethnographic studies of European communities, but one
will look in vain for accounts of their work in Dumont's study
of 'modem ideology'. I need not remind readers of Current
Anthropology that there are dozens of ethnographic studies made
by anthropologists in every part of Europe. What makes these
studies inappropriate as raw material for a work devoted to
European ideology? What makes similar studies conducted in
India by anthropologists similarly trained and using similar tech-
niques appropriate as raw material for a work on the civilization
of India? In what way can the two constructions be regarded as
comparable? I may not have achieved great success in my use of
the comparative method, but I have not used two different scales
for weighing the data on the societies I have compared.
 Individualism and equality 249
Dumont has been less than candid in facing up to the prob-
lems involved in comparing his own society with societies that
are distant in space, time, or moral texture. If the anthropologist
is, in Levi-Strauss's (1963, p. 378) picturesque phrase, 'the
astronomer of the social sciences', how does the astronomer
come to terms with the historian and the political economist?
The difficulties are formidable, and they cannot be disposed
of with cliches such as 'the reversal of perspectives' or 'the
approach to a civilization through its own values'. Where the
same subject is being studied, and must be studied, by people in
different existential situations, very little progress can be made
without candour about one's own views and consideration for the
views of others.
It has been suggested that I value tradition too little and
modernity too much. That is a harsh judgement against which
I do not wish to defend myself here. I cannot question someone
else's right to his own preference, although I know only too well
how generations of Western sympathizers of India have used
their admiration for Indian tradition to mask their allergy to
the India of their experience. As to modernity, there is very
little in that today - whether in India or in the the West - to
make the imagination soar. But I belong to the modem world,
and I would be untrue to my vocation as a sociologist to disown
the world to which I belong.
 10 Individual and person as subjeas
for sociology
Some thirty inches from my nose
The frontier of my Person goes,
And all the untill'd air between
Is private pagus or demesne.
Stranger, unless with bedroom eyes
I beckon you to fraternize,
Beware of rudely crossing it:
I have no gun but I can spit.
W.H.Auden

The terms 'individu~' and 'person' evoke and express a great

variety of ideas, and it is not possible at the outset to assign
a single, unvarying meaning to either of them. Some might
find it safer to take only one of the two concepts - and each
is broad enough - but my first objective is to draw attention
to the confusion of tongues that marks the discussion. I will
make my own observations about the changing position of the
individual in contemporary Indian society, but they will come
later. For the present, one might add to 'individual' and 'person'
the concept of 'self for that also has figured prominently in
sociological discussion of a certain kind, particularly in the
United States: 1 the same concept, though in a different form,
has been a perennial, not to say an obsessive, concern in certain
traditional systems of Hindu philosophy. 2 '
Individual and person are not neutral or uncharged scientific
terms; they represent important values on which there are sharp
and passionate disagreements, expressed as well as unexpressed.
Hence, even when their meanings are discussed in apparently
scientific terms, there is often an undercurrent of polemic.
 Individual and person as subjeas for sociology 251
Some scholars who are deeply attached to these values adopt
a possessive attitude towards them, arguing that they are
exclusive to their own social and cultural tradition. Others
who have inherited a different tradition then bring forward
a mass of evidence - not always germane to the issue - to
demonstrate that they have an equal, if not a better, claim
to the same values. A very common feature of contemporary
discussions on individual, person and self is for scholars to talk
past each other.
The variety of referents of each of these terms is not neces-
sarily the outcome of partisanship, but it enables partisanship
to flourish. While the terms in use and their meanings are
various, their usage is not carefully specified. The disciplines
which use the concepts of individual, person and self are many:
metaphysics, legal and political philosophy, history ofideas, psy-
chology, sociology and cultural anthropology. Scholars working
in a given discipline have generally felt free to kidnap terms,
concepts and arguments from other disciplines, leading more
often to confusion than to clarity. Standardization of terminology
is, in any case, very difficult, but partisanship compounds the
difficulty.
Although authors have their preferences, it is a common ten-
dency, at least among sociologists, to use individual and person,
or person and self, interchangeably. This is governed partly by
the conventions of the language being used. In English one can
make 'individuality' and 'individualism' out of 'individual', but
corresponding derivations cannot be constructed as easily out
of 'person'; 'personeity' sounds archaic and not quite right,
although 'selfhood' or even 'personhood' is frequently used,
at least in philosophy. Durkheim wrote most typically about
the individual, although he also used 'person' and 'personal';
Mauss wrote about the person, but also used 'the selr to cover
broadly the same range of meanings.
Let me illustrate with a passing example the kind of termi-
nological problem that frequently arises. Individual and group
are generally opposed in the sociological literature, and that
opposition was an important part of Durkheim's argument in his
first book3 as well as in his later writings. Now, while sociologists
might use individual and person interchangeably, and contrast
252 Society and Politics in India
both with the group or collectivity, in jurisprudence the legal
person can be either an individual or a group. It is well known
that in India, for instance, a group such as the Hindu undivided
family was, and continues to be, regarded as a legal person.
There are innumerable examples in archaic societies of groups
of almost every conceivable kind being treated as legal or moral
persons. This should not lead us to overlook the simple fact that
the individual might also be treated as a legal person in such
societies, or, conversely, that 'English law recognizes the legal
personality of groups and not only individuals.
Sociologists differ in their conception of the individual or
person not only because they deal with facts from different
societies, or with different facts from the same society, but also
because they differ in their philosophical and moral orientations.
The philosopher uses methods and techniques of analysis that
are both different from those of the sociologist and highly
specialized. Durkheim pointed out at the beginning of the
present century that 'the philosophical analysis of the idea
of person [was] far ahead of the sociological analysis'. 4 This
is perhaps still the case today, and the sociologist must watch
very carefully for the snares and pitfalls he will encounter
in his attempt to catch up with the results of philosophical
analysis.
Sociological accounts of the individual and the person have
taken shape largely within the orbit of contemporary Western
thought. Philosophical accounts of the person and the self have
a place also in other intellectual traditions with whose technical
apparatus those working within the framework of modern social
science are not always fully familiar. Students of comparative
sociology generally devote themselves to such topics as social
structure, social differentiation and social stratification. These
are large topics in themselves, and few who deal with them
acquire mastery over the relevant literature for the full range
of human societies. Not only is this literature highly uneven,
but for any given society or civilization there is no necessary
correspondence, in terms of quality, richness and reliability,
between the sociological literature and the philosophical one.
Even a very competent comparative sociologist is rarely at home
in philosophical cultures that are distant from his own in space
 Individual and person as subjeas for sociology 253
and time. In dealing with them he yields sooner or later to the
temptation of applying his common sense to the solution of
problems that are simple only in appearance.
A civilization may choose, in so far as such choices are made
by civilizations, to confine its preoccupation with the self or
the person to the world of ideas - metaphysics, ontology and
epistemology - without giving it much scope for expression in
the economic and political domains. This should not lead to
the settled conclusion, to which Mauss appears to have been
led in regard to India and China, that the concept then loses
its social efficacy for ever. 5 There are other possibilities to
consider, including the possibility that an ideal, insulated for
centuries from the everyday concerns of society, may be brought
back to life by the challenge of new historical conditions. The
present may relate itself to the past not only by replication but
also dialectically.
Since a variety of disciplines is engaged in the study of
individual, person and self, the approaches to the study are
many and diverse. Even within sociology, as Niklaas Luhmann
has recently pointed out, there are two rather different currents
which treat the individual as an emerging unit. In one of these,
best represented by G .H. Mead and his successors in the
United States, the individual is viewed as emerging 'from social
encounters', while in the other he is viewed as emerging 'from
history'. 6 The first is psychological in its orientation, while the
second is historical and comparative. It is with the second that
I shall be mainly concerned here.
In comparative and historical studies of the individual one
notices close interconnections among concept, theory and
method. As I shall be concerned mainly with such studies,
I would like to make a few observations on theory and
method since they bear closely on the selection of the facts
presented for comparison and contrast. What is striking about
the comparative study of the individual by sociologists and
social anthropologists is the correspondence between theory
and method; the correspondence is not only very close, it
is of the particular kind in which method is constrained by
theory. Now, there are two kinds of theory about the place of
the individual in human society and culture. According to the
254 Society and Politics in India
first, the individual or person is present everywhere although
his social role might vary enormously from one case to another.
According to the second theory, 'the individual as a value' or 'the
category of the person' is a unique feature of modem Western
civilization or ofWestern civilization as such, being absent or very
weakly developed in non-Western civilizations as well as in tribal
societies. Adherents of the first theory make their comparisons in
such a manner that human beings always and everywhere come
out as basically similar. Adherents of the second theory arrange
their comparisons with an eye to the uniqueness of a particular
society, usually their own; in its extreme form it is a method of
contrast rather than comparison.
It seems to me that one of the two approaches, the one which
emphasizes difference and contrast is the one that prevails today
in studies concerned with the individual in Indian society. Since
I am not wholly in sympathy with that approach, I would like
to spend a little time in exploring the roots of the difference
between the two approaches. I will begin with a brief consid-
eration of the place assigned by anthropologists of different
persuasions to the individual in primitive societies. This is
necessary in so far as one of the most influential views about
the individual in Indian society are linked directly to theories of
the individual in 'archaic' societies, a term used loosely to cover
both primitive societies and the ancient civilizations of China
and India.
If we look among anthropologists of an earlier generation,
we will find the two theories of the individual (or the person)
in the writings of Mauss (1872-1950) on the one hand and
Malinowski (1884-1942) on the other. These two anthropolo-
gists are exemplary not only for their own work, but also as
exponents of two distinct traditions in the study of man. Mauss
and Malinowski were almost exact contemporaries and they both
enjoyed great influence. Malinowski's reputation was perhaps
the greater in his own lifetime, but it declined sharply after he
died, undermined to some extent by his own students. 7 The
reputation of Mauss, by contrast, remains undiminished, not
least because of the steadfast loyalty of his pupils and disciples. 8
It is remarkable all the same that when the individual or person
is discussed today so much respectful attention should be paid to
 Individual and person as subjeas for sociology 255
Mauss and so little - indeed, hardly any - to Malinowski, even
by anthropologists in Anglo-Saxon countries. 9
Mauss's essay on the person is well known among anthropolo-
gists. It was delivered as the Huxley Memorial Lecture before
the Royal Anthropological Institute, 10 and it was the subject of
a recent Oxford symposium. Two separate English translations
of it have been published within the last ten years. 11 Instead
of trying to summarize the essay, I shall begin by drawing
attention to two of its important features. The first relates to
its theoretical line of descent and the second to its ethnographic
base of support. The central theoretical argument of the essay
has retained a remarkable sameness in French sociology from
Durkheim who was Mauss's teacher to Mauss's own pupils and
successors. That argument was first formulated at length by
Durkheim in The Division ofLabour in Society; 12 it was repeated
in a somewhat different form in a polemical essay published by
him five years later; 13 and it can be picked up at various points in
his The Elementary Forms ofthe Religious Life. Echoes of all these
works can be clearly heard in the essay by Mauss.
Durkheim's argument in his very first book, The Division
of Labour in Society was that, contrary to the views of .both
utilitarians and contract theorists, the individual was not the
starting point of social evolution; he was its end product.
The nineteenth-century view of individualism as an emergent
property was grounded in evolutionary theory, most notably in
the work of Herbert Spencer. It is true that Durkheim took
issue with Spencer in all his major writings that dealt with
the individual, but the alternative theory that he offered of the
emergence of the individual was also an evolutionary theory.
Although his style is far less assertive than Durkheim's, Mauss's
theory of the person is also set firmly within an evolutionary
framework. This is as one would expect, for he presented his
own essay as a sample of the work of the French school of
sociology whose adherents derived both their theory and their
method from Durkheim. It is this evolutionary framework that
encouraged Mauss to proceed at 'inordinate speed' from the
Zuni and the Kwakiutl to contemporary Europe without casting
more than a glance at the 'very great and ancient societies' of
India and China.14
256 Society and Politics in India
Now, as Evans-Pritchard often pointed out, the kind of
evolutionary theory to which Durkheim and Mauss subscribed
(Mauss more circumspectly than Durkheim) relied on a char-
acteristic artifice of method, which may be called the artifice
ofinversion. No matter what the institution under consideration
might be - religion, marriage, family or whatever - the end point
is taken as being given in the form current in one's own society;
the starting point is then constructed deductively by an inversion
of the features known in advance to have prevailed in the end. 15
Both Durkheim and Mauss placed a very high value on the
individual in their own society on moral as well as scientific
grounds; and, if the individual was to count for everything in
the most advanced society, as both of them believed and hoped,
then it stood to reason that he should count for nothing in the
most primitive ones. Although in an evolutionary series societies
are placed on a continuum, from the most primitive to the most
advanced, Durkheim, and to some extent also Mauss, frequently
uses his typology to contrast the most advanced society in which
the individual is valued above everything with all other societies
in which he is oflittle or no value.
Durkheim's contrast between the rigidly structured societies
of the past where 'the collective consciousness completely
envelops our total consciousness' so that 'our individuality is
zero' 16 with the more fluid and flexible modem societies where
the collective consciousness withdraws to make room for the
morally autonomous and responsible individual is well known.
In making the contrast he frequently uses examples from both
primitive tribes and ancient civilizations to illustrate what he calls
the 'alveolar' system of society. Mauss follows him in this when,
for example, he says in the opening pages ofhis celebrated work
on The Gift;
For it is groups, and not individuals, which carry on exchange,
make contracts, and are bound by obligations; the persons repre-
sented in the contract are moral persons - clans, tribes and families;
the groups, or chiefs as intermediaries for the groups, confront and
oppose each other .17
Durkheim was a master of the artifice of inversion, and he
presented his argument about the evolutionary primacy of the
 Individual and person as subjeas for sociology 257
group over the individual with superb dialectical skill. What he
sought particularly to impress upon his readers was that he had
facts whereas ·his adversaries - whether utilitarians or contract
theories - had only their opinions. Now, it is well known that
Durkheim's ethnography, in The Elementary Forms of the Religious
Life and even more in The Division ofLabour in Society, is highly
defective. Between The Division of Labour and the essay on the
person by Mauss there was an interval of nearly half a century
which saw a tremendous advance in knowledge about simple
societies through intensive fieldwork in Oceania, Africa and
elsewhere. This research not only presented new facts about
primitive societies, it created a new image of primitive man
- quarrelsome, amicable, generous, mean, aggressive, timid,
treacherous or steadfast - much as human beings are anywhere
in the world. 18 Intensive fieldwork appeared to have an in-built
bias for the individual both in the collection of data and in their
presentation.
Malinowski was the pioneer of the new ethnography, and
he felt himself to be more than a match for any armchair
anthropologist with pretensions to general theories of evolution.
He knew how to choose his targets, and he hit hard. He took up
the cudgels on behalf of the 'individualism' of primitive man, and
he set about to overwhelm the sceptics with a profusion of facts
fresh from the field. The idea of primitive men and women as
mere actors in a play of masks, as represented in Durkheim's
The Elementary Forms and even more pointedly in Mauss's essay
on the person, was precisely the one he set out to demolish. 'The
opposition of primitive "group sentiment", "joint personality"
and "clan absorption" to civilized individualism and pursuit of
selfish ends appear[s] to us', he wrote, 'altogether artificial and
futile'. 19
Malinowski attacked Durkheim and Mauss openly. Criticizing
the views of Steinmetz, he wrote:
His views are fully endorsed by the great French sociologists
Durkheim and Mauss, who add besides one more clause: that
responsibility, revenge, in fact all legal reactions are founded in
the psychology of the group and not of the individua1.20
His own conclusion was as follows: 'The savage is neither an
258 Society and Politics in India
extreme "collectivist" nor an intransigent "individualist" - he is,
like man in general, a mixture of both'. 21 Characteristically, he
did not take much trouble to enquire whether it was the same
mixture in every case.
The stricture on 'clan absorption' entered by Malinowski in
1926 went straight past Mauss when he presented his paper to
the Royal Anthropological Institute in 1938. What is remarkable
about that paper is that it ignores altogether the new ethnography
which in Britain and elsewhere was transforming the study of
primitive society. There is no reference to any of Malinowski's
monographs or to any of the monographs that the first batch
of his pupils - Bateson, Firth, Evans-Pritchard - had already
begun to publish. The ethnographic sources used by Mauss
are of the old type: Frank Hamilton Cushing, Barbara Freire
Marecco, Elsie Clews Parsons, and articles by Boas and others
in the Annual Reports ofthe Bureau ofAmerican Ethnology. They
are in the main the same as the ones used by Durkheim in
The Elementary Fonns, and by Durkheim and Mauss in Primitive
Classification.
For a time at least, Malinowski appears to have been wholly
successful in his effort at 'bringing the men back in'. But did
he overwork his argument about universal individualism? The
general consensus among anthropologists today is that he
clearly did. In a penetrating critique, Edmund Leach argued
that Malinowski's empiricism reached the point of theoretical
sterility very quickly because of his dogmatic insistence that
all human beings were alike in acting as pragmatic individuals.
'Malinowski's biggest guns are always directed against notions
that might be held to imply that, in the last analysis, the
individual is not a personality on his own possessing the capacity
for free choice based in reason'. 22 And again, 'It was dogma
for Malinowski that all human beings are reasonable (sensibly
practical) individuals'. 23 This kind of dogma made it impossible
to examine differences between societies and cultures in a
serious or systematic way.
Without taking anything away from Leach's critique of
Malinowski, we can still say that the ethnographic record flatly
contradicts Durkheim's thesis about the position ofthe individual
in primitive society and Mauss's thesis about the place of the
 Individual and person as subjeas for sodology 259
person in primitive consciousness. Those who have contributed
to this record include anthropologists like Evans-Pritchard and
Godfrey Lienhardt who are sceptical about, if not opposed to,
empiricism and pragmatism. Gregory Bateson, who published his
remarkable study of the latmul of New Guinea in 1936, though
trained by Malinowski in field methods, had serious reservations
about his theoretical approach. But about the latmul themselves,
he tells us that they have 'a highly individualist culture and will
readily respect the law-breaker if he have but sufficient force
of personality'. 24 Personal pride and its public e"J)ression are
recurrent features of the latmul way of life. Further, 'A man
achieves standing in the community by his achievements in war,
by sorcery and esoteric knowledge, by shamanism, by wealth, by
intrigue, and, to some extent, by age'. 25
But the most decisive evidence comes from Evans-Pritchard
whose professed theoretical sympathies were all against Mal-
inowski and all in favour of Mauss. He has left us vivid accounts
of the Nuer which alone, in all the ethnographic literature, can
compare with Malinowski's accounts ofthe Trobriand Islanders.
'The Nuer constitution', Evans-Pritchard tells us, 'is highly indi-
vidualistic and liberatarian'. 26 In talking about Nuer character,
he remarks upon 'their profound inclividualism'.27 Then he says,
'The Nuer has a keen sense of personal dignity and rights'.28
And, again, it is difficult to see how the Nuer can be said to
commit sin or suffer misfortune if they lack a concept of the
person. 29 Finally, we have Godfrey Lienhardt's recent essay on
the Dinka which tells us with exquisite irony that if we cannot
see the person behind the individual Dinka, it may be our failing
rather than their lack.30
Returning for the last time to Malinowski, it is certainly true
that some anthropologists represent the others about whom
they write a little too closely in their own image, and we must
be watchful about this. But we must be watchful also about
those anthropologists who represent others not as copies but
as inversions of themselves. The objective should be to avoid
the extremes of both variants of the comparative method, one
of which overlooks difference while the other dwells only on
difference.
260 Society and Politics in India
In his essay on the person Mauss uses the ancient civilizations
of India and China as a bridge to pass from the primitive to
the modem Western world. But it is a slender and precarious
bridge, hardly equal to the burden of conflicting evidence that it
has to bear. He mentions, en passant, the classical Hindus' ideas
of aham and ahamkara, but not in a manner calculated to divert
attention from the main argument that the 'category of person'
as against the mere 'sense of person' comes into its own only in
modem Western society.
We have seen that, contrary to Durkheim and Mauss, in the
simplest societies the individual was by no means completely
absorbed by the group or completely devoid of a sense of
personal identity and worth. The individual might, in fact, enjoy
considerable freedom of action in such societies, and his sense
of personal dignity and worth might sometimes be quite acute.
Can we say from this that the individual or person must have
had considerable significance in Indian society since that society
represented a more advanced type than any primitive society?
Such an inference would be fully in keeping with Durkheim's
and Mauss's evolutionary perspectives, but it would be mistaken.
The lessons we learn from primitive societies can be useful in
the study of Indian society, but they cannot be conclusive.
Despite its high level of intellectual culture and the consider-
able mental agility of its philosophers, traditional Indian society
provides better evidence of Durkheim's 'alveolar system' than
any one might encounter in the world of primitive man. It was
here if anywhere that boundaries between groups were clear
and sharp, and individuals confined to bounded social spaces
to such a degree that one might plausibly assert that the group
predominated over the individual. Durkheim's description of
the alveolar system seems to correspond remarkably with the
model oftraditional Indian society, whereas alongthe Sepik or the
Amazon we are very far removed from any such type of society.
However, the prominence of the alveolar system - of groups
clearly demarcated from each other - did not prevent the
development of a highly articulate consciousness in regard
to individual, person and self. Durkheim clearly believed that
the segmental type of society went hand in hand with an
absence of individual consciousness or the awareness of the
 Individual and person as subjeas for sodology 261
individual as a morally significant agent. In such a society the
individual played a part or enacted a role, but was not the
object of metaphysical or ethical valuation. Mauss took over the
argument from Durkheim, and only the grip of that argument
can account for the astonishing statement at the beginning of
his essay: 'I shall show you how recent is the philosophical term
"self" '3 1, astonishing because Mauss enjoyed a reputation as a
Sanskritist.
In writing about non-Western societies neither Durkheim
nor Mauss showed much awareness of the variety of ways
in which consciousness and thought might be related t-0 the
social order. The Iatmul might tolerate or even encourage
many acts of individual deviation from social convention, and
yet lack an articulated consciousness of the meaning of indi-
vidual freedom. The Hindus might demand strict conformity to
social convention from the individual, and yet develop complex
philosophical arguments about the nature and significance of the
self or person.
Instead of bypassing the philosophical discussion of the self
in the Hindu literature as Mauss did, one might dwell on the
obsessive concern with the self in it, and still reach the same
conclusion as Mauss did about the individual or person in Indian
society. It goes without saying that the conception of the self in
Advaita Vedanta is vecy different from the conception in modem
Western thought. Nevertheless, it does cause some anxiety to
see the ease with which the conclusion is reached that Indians,
including modem Indians, lack an appreciation ofthe individual
or person.
Hindu conceptions of the self cannot be studied in the same
way as Iatmul or Nuer ideas about the individual. The method
has to be different because the nature and sources of data are
different. In the case of the Nuer or the Iatmul the same
person studies social structure on the one hand, and religion
and ideology on the other, whereas the case is vecy different
with traditional Hindu society. In the Hindu case there are
several traditions of discourse regarding individual, person and
self and some of them reach far back in time. The discourse has
been kept alive by intellectual specialists who have disputed with
each other with the aid of a complicated, not ~o say an esoteric,
262 Society and Politics in India

technical apparatus for two thousand years or more. This is not

a subject which the general anthropologist with a strong dose of
common sense can handle on his own.
It is not uncommon for Indian and Western philosophers
to talk past each other on the subject ·of the individual or
person. Professor Kalidas Bhattacharyya, one of the ablest of
contemporary Indian philosophers, wrote in an essay which is
a model of scholarship, 'Regarding the Indian attitude toward
the individual, Westerners hold certain views which are not
wholly correct'. 32 After listing four such views, he went on to
observe, 'Of these four views, the last one is wholly incorrect,
and the first, second and the third are correct only insofar
as they represent the standpoint of the Advaita Vedanta'.33
Not all philosophers pay heed to such admonitions, and the
anthropologist is either unaware of them or is unable to see
the reason behind them.
Most contemporary Indian philosophers who have written on
the subject have stressed that there was not one traditional
view of the individual or person but several. Bhattacharyya
observed: 'in fact, there is no one Indian view of the status of
the individual'. 34 He distinguished the view of the individual in
Advaita Vedanta from the views in other philosophical systems
such as Nyaya, Samkhya and Mimamsa, and pointed out that
the denial of the plurality of individuals and the assertion of
their oneness, which was a characteristic of the former, was not
a feature of other philosophical systems, including other systems
of Vedanta. He also maintained that the prominence enjoyed
by Advaita Vedanta was a result of recent historical factors.
Professor Raju, writing about the individual in Indian epis-
temology, also stressed the plurality of perspectives, concluding
thus: 'All advanced cultures are in this sense pluralistic, their
members leaning toward one or the other perspective according
to circumstances'.35
These different philosophical systems co-existed not only
with each other, but with popular conceptions, legal institutions,
social regulations and economic arrangements of the most
bewildering variety. In course of time the social and economic
arrangements became increasingly rigid, leaving the individual
less and less space for movement outside his allotted positions in
 Individual and person as subjeas for sociology 263
society, and changes also took place - to some extent in tune with
the ones described - in legal institutions and modes of thought.
But the alveolar system ofsociety was not a primitive condition; it
was a later development and it did not extinguish the concern for
the individual in the realm ofabstract ideas. Rather, that concern
became more and more detached from the ongoing institutional
arrangements ofsociety. The individual or person, as both agent
and value, has once again become significant with the effort to
find a new place for him in these very arrangements under
greatly altered historical conditions in which the distinction
between 'endogenous' and 'exogenous' factors has lost its old
significance.
Historically, the great impetus for rethinking the individual
came with the introduction by the British in the nineteenth
century ofnew legal and economic arrangements, and new ideas
about the rights and responsibilities ofthe individual. There can
be no doubt that these legal and economic arrangements were
substantially new, and therefore the place of the individual in
them had to be thought of anew. Indians not only acquired new
ideas from western sources, but they went back to their own
sources to seek out new meanings from old ideas. Many ofthose
who wrote about the individual or person - and they included
poets and philosophers as well as social and religious reformers
- genuinely believed that they were engaged in a rediscovery of
their own past. The view that traditional India 'knew nothing of
the individual' must either ignore their effort or treat it as a kind
of self-delusion.
Two of the makers of modern India, Tagore and Gandhi,
each stressed in his own way the respect due to the individual
and the responsibility due from him, not only in some ultimate
metaphysical sense but also as a member of society. There is
no doubt that both Tagore and Gandhi, as indeed others of
their generation and the two preceding ones, had been greatly
influenced by Western ideas, but it would be quixotic to maintain
that they had become disconnected from their own traditions of
religious and philosophical thought.
Tagore was one ofthe most influential poets ofhis generation,
remembered to this day by two independent nations to whom
he gave their national anthems. He was a poet and not a social
264 Society and Politics in India

theorist, but his critique of caste and untouchability is among

the most profound and moving examples ofits kind. Tagore was
unusual among Indians of his time in so far as he stood clearly
against nationalism. 36 The nation, he argued, was an artificial
construction which should never be placed on the same plane
of value as the individual or humanity. He wrote extensively
about the human personality which he felt should be set free
from artificial social restraints including those of caste. He was
a cultivated person who read widely, Walt Whitman being one
of his favourites, but he always referred back to the Upanishads
in making his case for the person. 37 The point is not whether his
was the correct interpretation of the µpanishads but that it was
the interpretation of a person whose voice was acknowledged by
millions of Bengalis and other Indians.
The same argument may be madt about Gandhi who was in
other respects a very different person from Tagore. Although
he was an indefatigable writer of newspaper articles, pamphlets,
letters, memoranda, etc., he influenced people less by his writing
than by his life which many regarded as exemplary. Gandhi
believed very strongly in the dignity of labour and in individual
self-reliance, and he required everyone who lived with him to
do some manual work. The anthropologist N.K. Bose, who was
closely associated with him for a time, has left an invaluable
account of the life and work of Gandhi. 38
An intellectual ferment had been· going on in India for a
hundred years when Mauss wrote his essay on the person. Why
did he and other anthropologists of his persuasion who brought
India into their discussion ofthe individual or person pay so little
attention to the history of ideas in India in the nineteenth and
twentieth centuries? Mauss himselfprovides a clue that is highly
revealing. At the end of his essay on the category of the person,
he wrote: 'It was formed only for us, among us'. 39 The same
possessiveness towards what is without doubt one of the major
values of modem Western civilization is to be found, though it
is not as clearly revealed, in Durkheim's essay on individualism
as well as in more recent writings on the subject by European
scholars.
The possessive attitude towards individualism as a value
is not infrequently expressed in an arrogant and aggressive
 Individual and person as subjeas for sodology 265
manner. Nationalist Indians respond to that with a characteristic
mixture of vehemence and bombast. After all, their present
Constitutional and legal order commits them to placing a
certain value on the individual, and they do not like being told
that that value is the exclusive possession of a totally different
intellectual tradition. Hence, the more strident among them set
out to demonstrate that they not only have a claim on that value
but that they have an older and a better claim on it than anyone
else. One's faith in the comparative method must be very strong
indeed to survive such onslaughts from both sides.

The search for an authentic identity for the individual in India's

traditional culture should not be taken to indicate a failure
to recognize the subordination of the individual to the group
in contemporary Indian society. That recognition was clearly
articulated by Jawaharlal Nehru. Writing about Indian social
structure, he underlined the primacy of the group: village,
caste and joint family. He observed, 'In all these three it is
the group that counts; the individual has a secondary place'.40
And again, 'An individual was only considered as a member of
a group; he could do anything he liked so long as he did not
interfere with the functioning of the group'. 41 Nehru's attitudes
towards the individual and the group were mixed: he was wholly
against caste, but he believed that something of value might be
preserved in the village and the joint family. He strove to create
more room for the individual in society, not just more scope for
self-realization in a metaphysical sense but more opportunity for
movement across the entire social space. To be sure, he was
repelled by untempered individualism as an economic doctrine
of the utilitarian kind; but so, indeed, was Durkheim.
It is obvious that family, kin and community signified, and
continue to signify, something very different for Indians in all
walks of life in comparison with what they signify in modern
Western society. Indians in general, and to some extent irrespec-
tive of region and religion, retain strong feelings of obligation
towards family, kin and community, although we should not
ignore altogether the changes being brought about by modern
education, professional employment and·urban living. Also, we
should riot over-stress the moral obligation to the group, for
266 Society and Politics in India

those who are fully at ease in the idiom of caste and community
may use. these instrumentally to attain individual ends, as in the
case of the managers of political machines whether in India or
the United States.
The joint family is a favourite subject for discussion among
Indians. It is, according to inclination and circumstance, con-
demned by some for destroying individuality and extolled by
others for keeping in check the excesses of individualism. It is
true that most Indians, including most Hindus, do not live in
joint families in the sociological sense ofthe term, but family and
kin count for much more among them than they do in the West.
There are two aspects of this: firstly, a wider range of relatives
is recognized, and, secondly, the obligations due to them are
more numerous and more varied. It is often pointed out that
one does not choose one's relatives, and in India this is largely
true also of one's spouse. But to what extent is one allowed,
within the system, to treat the relative as a person, according to
his personal qualities and irrespective of the formal structure of
rights and obligations? This is an important question on which,
unfortunately, very little systematic information is available.
Societies differ not only in the extent to which they offer
a choice of roles outside the domain of kinship, but also in
the extent to which they· respect privacy within that domain.
Privacy within the family has been valued in England and certain
other European countries for a long time, and the high value
placed on it there, but not necessarily elsewhere in the West,
probably antedates the major changes in the division of labour
and the occupational structure associated with the Industrial
Revolution. 42 On the other hand, what Indians value is not
privacy within the family but gregariousness among relatives.
It is very likely that there is a link between the value placed
on privacy and the respect for the person, but that link has to
be demonstrated by careful comparative research and cannot be
asserted as an established truth.
The village is a very different kind of group - if 'group' be
the right word - from family and caste which are both based on
kinship in some sense of the term, or at least on the sentiment
of kinship. While it would be misleading to argue that the Indian
village lacks 'sociological reality', loyalty to it does not interfere
 Individual and person as subjects for sociology 267
very seriously with the pursuit of individual opportunity when
it makes itself available. Indians have moved increasingly from
villages to towns and cities in search of education and employ-
ment in sectors of the economy where individual initiative and
individual success count for much more than in the traditional
village. It is true that most Indians continue to live in villages, but
that may be as much a question of opportunity as of value.
The disintegration of the Indian village from the onslaught of
British imperialism became a favourite theme with nationalists
who constructed a rosy picture of the traditional village as a
haven of unity, harmony and co-operation. Nostalgia for the
harmony of village life was often very strong among precisely
those Indians who had left the village and achieved success in
professional life in towns and cities. 43 Some of them sought
to give flesh and blood to their nostalgia with plans for the
reconstitution of the Indian polity on the basis of village auton-
omy.44 They argued in the Constituent Assembly, invoking the
authority of Gandhi, that the village and not the individual
should be the irreducible unit of the new Constitution of India;
but they failed.
Not everyone subscribed to the nostalgia for the Indian village
or the myth of village unity and harmony. Dr B.R. Ambedkar,
hailed as the principal architect of the Constitution, relentlessly
exposed the hiatus between the myth and the reality ofthe Indian
village. In a celebrated speech in the Constituent Assembly he
attacked the villages as cesspools of bigotry, superstition, petty
strife and merciless oppression. He warned against attempts
at reviving the spirit of the traditional village and pleaded,
successfully, for making the individual and not the village (or
any other group) the principal bearer of rights and capacities.45
While many might speak up for the Indian family system and
some even for the village community, the defence of caste has
proved to be a far more difficult endeavour, at least since the
end of the last century. Even Mahatma Gandhi, whose foibles
his countrymen treated with unusual indulgence, had in the end
to give up his defence of caste.46 I regard this collapse of the
public defence ofcaste to be a turning point in India's ideological
history. To be sure, Indians continue to live by caste in a variety
of different spheres. But in public life, they will at best justify the
268 Society and Politics in India
use of caste by saying that it is a means to an end, not an end
in itself..
Caste had a very different place in the Indian consciousness
in earlier times. The social subordination of the individual to
caste is best seen in the Dhannashastra which may be regarded as
the charter of the traditional Hindu social order. But even here
one has to proceed with caution. As Professor Kane has pointed
out, the traditional authors were familiar with both sadharana
or samanya dhanna (i.e. dhanna common to all mankind) and
varnashrama dhanna (i.e. dhanna specific to caste and stage of
life). Kane observes, 'In the midst of countless rules of outward
conduct there is always insistence on the necessity to satisfy
the inner man (antara-purusa) or conscience'.47 However, he
makes it plain that the writers of the Dhannashastra concerned
themselves principally with varnashrama dhanna and not with
samanya dhanna.
In everyday affairs the individual was expected to go along
with his caste, and a combination of economic and religious
constraints kept each caste in its allotted place. No single
individual or group of individuals could easily upset the massive
and elaborate architecture that became gradually established.
Religious movements emerged from time to time whose leaders
sought to free the individual from the constraints of everyday
life, but their long-term impact was small. N.K. Bose argued
that these movements could have very little social impact so
long as the economic structure of society remained unchanged.
Attributing the failure of the 'tremendous upheaval created
in society by Chaitanyadeva' to the fact that 'the economic
organization of the village remained in its ancient form', he
concluded: 'I think that it is this that ensured the triumph of
Raghunandan in regard to both social practices and religious
beliefs'. 48
British rule not only exposed Indians to new moral and
intellectual currents, it also imposed new economic and political
arrangements on them. Western ideas, beliefs and values posed
a challenge of unprecedented magnitude. Indian intellectuals
were quick to respond to the challenge. They wrote novels,
pamphlets, tracts and, of course, poetry, and they started
newspapers in both English and the vernaculars to express
 Individual and person as subjeas for sociology 269
their ideas. The response, as one would expect, was highly
various. Some were totally opposed to the 'new individualism';
a few embraced or professed to embrace it wholeheartedly; the
rest were more cautious in their response.
An important part of the historical development of the
individual is his development as a legal person. British rule
played a major catalytic role in initiating the development of
a legal personality for the individual unencumbered by the
burden of caste. This happened despite the great caution with
which the British proceeded, after a first phase of enthusiasm,
in regard to the personal laws of their Indian subjects. As
Srinivas has pointed out, the rulers of independent India were
able to dispense with the circumspection that the British felt
was necessary and desirable in their case.49
A watershed in the legal emancipation of the individual from
his caste was the passage in 1850 of the Removal of Caste
Disabilities Act. Particularly among Hindus of the higher vamas
many of the legal rights of the individual were acquired by
virtue of his birth in a particular caste, and expulsion from
caste amounted in such cases to civil death. The Act of 1850
gave some protection to the individual in such matters, but it
did not by any mean.s do away with all the disabilities due to
caste. These disabilities were progressively whittled down in
the next hundred years, and the Constitution of 1950 sought
to replace a legal order based on the privileges and disabilities
of groups by one based on the rights and responsibilities of
individuals.
This is not the occasion for discussing in detail the place
assigned to the individual in the Constitution of India which is
a large subject on which I have commented elsewhere.so What I
would like to stress here is that the making of the Constitution,
which took more than two years, was itself a historical process of
great importance. The Constituent Assembly brought together
Indians from different sections of society representing different
points of view, and they put forward claims on behalf of units
of every conceivable kind. Primacy was in the end assigned to
the rights of the individual, save where provisions had to be
made for protecting the interests of minorities and Backward
Classes.
 270 Soci.ety and Politics in India
Some place has thus been found for groups as well as individ-
uals in the Constitution of India. The provisions for the former
are of two different kinds. T~ere are those which seek to protect
the religious, linguistic and cultural interests of minorities; their
justification is sought in the value assigned to pluralism by
both traditional Indian culture and modem democratic theory.
Provisions of the second kind relate to positive discrimination
in the interest of the Backward Classes or weaker sections
of society; these bring out more openly the conflict between
individual and collective interests because they restrict to some
extent the rights of individuals to equality of opportunity. The
continued and even the increasing force ofthese latter provisions
has led some to argue that, far from eliminating the traditional
identities of caste and community, the new Constitution has in
some ways reinforced them.

I have tried to make in the foregoing two points regarding. the

historical development ofthe individual in India from the middle
of the last century till the present. Firstly, this development did
not begin in the nineteenth century on a tabula rasa: the encoun-
ter with the West introduced new ideas but also raised echoes of
old ones, and they all contributed to the process ofdevelopment.
Secondly, such development as there has been, has been neither
1 uniform nor continuous; it has been greater in some domains

than in others, and in some phases the group has not only held
its own against the individual but even enhanced its strength.
Having stressed the uneven development of the social per-
sonality of the individual in contemporary India, I would like
to examine very briefly in this concluding section some of the
factors that make the development uneven and discontinuous.
Although there are different ways of looking at the individual
or the person - and I have stressed this all along - what the
sociologist cannot fail to note are the points of tension between
individual and collective identities. It was a part of Durkheim's
argument that socially the individual comes into his own with
the attenuation of collective identities of the kind based on race,
caste and community, and that this attenuation could be seen
as a uniform and continuous process at least in contemporary
Western societies.st
 Individual and person as subjeas for sociology 271
Collective identities have proved to be far more durable in
the modem world than was envisaged by the sociological theory
of the nineteenth century. This is true almost everywhere -
Europe, America and the Soviet Union - but I shall take only
the case of India. Since I have written more fully elsewhere
about the persistence of collective identities, 52 I shall confine
myself to a few brief observations here.
Tribe, caste, sect and denomination did not vanish from the
public domain with the coming of independence; they appear,
in some sense, to have become even more strongly entrenched
there. The Muslim identity has survived the partition of India
without significant loss of strength, and the Sikh identity is now
probably at its strongest. The makers of the Indian Constitution
were a little like Durkheim in their belief that a new economic
and political order would create more opportunities for the
individual and greater confidence in him; and they also believed
that India was firmly set on the course ofrapid technological and
economic change.
The experience of most contemporary societies suggests that
short-term fluctuations in the strength of collective identities
tend to obscure such long-term trends in their growth or
decline as might operate below the surface. Since there is
no single convincing theory of these long-term trends, the
short-term fluctuations are often seized upon as evidence of
the direction of change. A common view of the resurgence
of collective identities is that it represents a return to the
initial condition of Indian society after a brief and superficial
encounter with the Western values of individualism, liberty and
equality. This view appeals naturally to those who stress the
differences between civilizations, regarding each civilization as
an irreducible whole.
There is, however, another view of the same phenomenon
which maintains that the group - tribe, caste, sect or denomi-
nation - acquires a new lease of life because of the insecurity
experienced by the individual in an unstable and intractable
social environment. The proponents of this view point out that
the sense of collective identity is often more acute in towns and
cities, particularly industrial cities, than in the more traditional
rural areas. According to N.K. Bose:
272 Society and Politics in India

In Calcutta the economy is an economy of scarcity. Because there

are not enough jobs to go around everyone clings as closely as
possible to the occupation with which his ethnic group is identified
and relies for economic support on those who speak his language,
on his coreligionists, on members of his own caste and on fellow
immigrants from the village or district from which he has come.
By a backwash reliance on earlier modes of group identification
reinforces and perpetuates differences between ethnic groups.SJ
Here the phenomenon is seen more as an aspect of transition
than as a return to an original condition.
Dr Ambedkar, who spoke so forcefully in the cause of the
individual in the Constituent Assembly, persuaded the same
Assembly to make special provisions to safeguard the interests
of the Scheduled Castes and Scheduled Tribes. He asked for
special provisions on the ground that the individual members
of these groups remained severely disadvantaged because of
the disabilities to which they had been collectively subjected
for centuries. He saw that there might be a conflict between
the special claims of particular groups and the general claims
of all individuals, and he urged restraint in regard to the
former, saying that they should not be allowed to 'eat up'
the latter. 54 The provisions were intended to enable these
highly-disadvantaged sections of society to catch up with the
rest, particularly in matters of education and employment, and
not, as were those in favour of religious or linguistic 'minorities',
to safeguard their distinctive identities for all time. 55
The public recognition of the Backward Classes and of the
minorities, and the allocation of quotas in their interest go back
to the period of British rule, and were not innovations designed
by the Constitution of 1950. In fact what that Constitution
sought to do was to restrict rather than expand the claims
of such groups in the public domain. To take one example,
in much of South India quotas in education and employment
had been instituted in the 1920s for religious minorities and the
upper castes as well as for the lowest castes and communities.
The Constitution did away with quotas or reservations for
religious minorities and upper castes, retaining them only for the
Scheduled Castes and Scheduled Tribes, and, somewhat more
 Individual and person as subjeas for sociology 273
ambiguously, for Other Backward Classes or weaker sections
of society. 56 The main purpose behind this restriction was to
provide more scope to the rights created for individuals - equality
before the law and equal protection of the laws, equality of
opportunity- in the part on Fundamental Rights.
Why should the claims of any kind of group be admitted in
a Constitution whose basic objective is to secure the rights of
individuals? And, what kinds of groups are these? In general,
exceptions are made in favour of groups only when they can
be justified by a policy of greater equality overall.57 Thus,
there can be quotas for inferior castes but not, as there also
were in the past, for superior castes. This is the policy of
reverse discrimination for which the Constitution of India has
a variety of provisions. They represent interesting departures,
by no means unknown outside India, from the classical liberal
idea of equality which was defined solely in terms of the claims
ofindividuals.ss
Advocates of positive discrimination do not deny the value
in principle of equality of opportunity between individuals, but
they insist on the distinction between real and formal equality
ofopportunity. 59 They argue that to have real equality of oppor-
tunity we require measures for the redress of the disadvantages
inherited by certain groups on which severe disabilities had been
continuously imposed in the past. That argument generally
concedes that such measures, being measures of redress, have
a specific purpose. In other words, the benefits accruing from
them should be restricted to limited sections of the population
and the measures themselves should be time-bound and not
perpetual.
What has been the record of positive discrimination in the
forty-odd years since independence? How have the rights of the
individual guaranteed by the Constitution withstood the special
claims of groups acknowledged by the same Constitution? It
can hardly be denied that developments in the last forty years
have belied the expectations of the Constitution makers in some
important ways. The groups securing the benefits of positive
discrimination have expanded instead of contracting, and in
some states they now make up well over half of the total
population. 60 Moreover, their claims have become so secure
274 Soci.ety and Politics in India
that it does not appear now that the measures can be reversed
in the foreseeable future.
Public policy in India is governed increasingly by considera-
tions of social justice in which parity between groups receives
more attention than equality between individuals. Critics of the
policy view it as an accommodation to the struggle for power
between organized groups, but it also appeals to a conception
of social justice which probably lies very deep in the Indian
consciousness. However, public policy is not the only force in
India today, and, in any case, it operates in a greatly altered
environment where competition between individuals as well as
groups is widespread if not universal.
It is true that middle-class Indians value group-based quotas
in education and employment both from considerations of secu-
rity and out of a sense of social justice. But they also value the
sweets of success in competition between individuals, and it is a
prejudice, contradicted by the Indian experience, that those who
appreciate quotas for groups on grounds of security and social
justice cannot, at the same time, value competitive success as a
measure of individual ability and achievement.
In India the Government plays a large part in providing
education and employment in what may be called the organized
sector, and parity between groups is always an important con-
sideration for it. But even here one finds significant variations.
Some state Governments, as in Karnataka, are particularly
responsive to the demands of castes and communities; others,
as in West Bengal, are more responsive to the demands of organ-
ized labour which represents a different kind of collectivity.
Moreover, education and employment are also available in the
private sector where other considerations prevail. Government
has never shown a very strong inclination to suppress the private
sector even where it is regulated by considerations which depart
from its own conception of social justice. Indeed, there are
indications that it has in the last few years encouraged the
development, outside the area of its own immediate control, of
a sector in which the main accent is on competition, efficiency
and individual initiative. Capitalism has its own compulsions,
and even Indian society cannot remain wholly immune to them.
The protagonist of competitive capitalism in its pure form
 Individual and person as subjeas for sociology 275
is a very different kind of individual or person from the one
about whose emancipation Nehru or Tagore wrote. The birth
of the 'new individualism' is due as much to a transformation
in the moral awareness of people as it is to a transformation in
the economic structure of society with the attendant changes
in material and moral density about which Durkheim wrote.
Historically, the two transformations have been very closely
intertwined. Durkheim's vision of a society in which the utili-
tarian individualism of Spencer and the economists will be
separated out from the moral individualism of Kant was a
utopia which will perhaps always remain out of the reach of
human societies as they are actually constituted. It is not that
there is no individual in contemporary Indian society, but we
should not expect to find him everywhere or in only a particular
idealized form.
 Notes and sources
Introduction
1. A. Beteille, 'Some observations on the comparative method', Economic and
Political Week(y xxv, 40, 6 October 1990, pp. 2255-63.
2. E. Durkheim, The Rules of Sociological Method (Macmillan: London,
1982) p. 157.
3. C. Levi-Strauss, Le regard eloigne, (Pion: Paris, 1983); translated as The
View from Afar (Basic Books: New York, 1985).
4. See also M.N. Srinivas, Social Change in Modern India (University of
California Press: Berkeley, 1966).
5 For a brief personal account of this department, see A. Beteille, 'A career
in a declining profession', Minerva XXVIII, 1, Spring 1990, pp. 1-20.
6. M.N. Srinivas, 'Itineraries of an Indian social anthropologist', International
Social ScienceJournal XXV, 1-2, 1973, pp. 129-48.
7. N.K. Bose wrote on a very broad canvas. See, for instance, his The
Stmcture ofHindu Society (translated from the Bengali with an Introduction
by A. Beteille). (Orient Longman: Delhi, 1975).
8. M.N. Srinivas (ed.), India's Villages (Government of West Bengal:
Calcutta, 1955); M. Marriott (ed.), Village India (University of Chicago
Press: Chicago, 1955).
9. A. Beteille, Caste, Class, and Power (University of California Press:
Berkeley, 1965).
10. See my Introduction in N.K. Bose, The Stmcture of Hindu Society
(Delhi, 1975).
11. For a brief but interesting contrast between the two 'structuralisms',
see M. Sahlins, Culture and Practical Reason (University of Chicago Press:
Chicago, 1976), pp. 1-54.
12. L. Dumont, A South Indian Subcaste (Oxford University Press: Delhi,
1986) (Originally published in French by Mouton in 1957).
13. L. Dumont, Hierarchy and Marriage Alliance in South Indian Kinship
(Royal Anthropological Institute: London, 1957).
14. L. Dumont, A South Indian Subcaste (Delhi, 1986), p. 4.
15. The opening number of Contributions to Indian Sociology (1, 1957) had
a joint statement by the two editors that had something of the character of
a manifesto. The first three numbers were principally the joint work of the
two editors who, through a series of reviews and commentaries, spelled ·out
their own position on the study of Indian society and culture. Although it soon
 Notes and sources 277
opened its pages to other contributors, it has remained, by and large, faithful
to its original inspiration.
16. L. Dumont, Homo hierarchicus (Paladin: London, 1972), p. 74.
17. M. Marriott (ed.), India through Hindu Categories (Sage: New Delhi,
1990). '
18. A. Beteille, 'Sociology and ethnosociology', International Social Science
Journa/XXVI, 4, 1974, pp. 703-4.
19. A. Beteille, 'Some observations on the comparative method', Economic
and Political Weekly XXV, 40, 1990.
20. J. Fabian, Time and the Other (Columbia University Press: New
York, 1983).
21. A. Beteille, 'The problem', Seminar 157 (11,e Social Sciences), September
1972, pp. 10-14.
22. J. Baechler, La Solution indienne (Presses Universitaire de France:
Paris, 1988).
23. N.K. Bose, 'The Hindu method of tribal absorption', Science and Culture
VII, 1941, pp. 188-94.
24. A. Beteille, 'Homo heirarchicus, homo equalis', Modern Asian Studies 13,
4, 1979, pp. 529-48; see also my 'Some observations on the comparative
method', Eronomic and Political Weekfy XXV, 40, 1990.

Chapter 1 : Race, caste and gender

1. If I were to specify a turning point, I would choose the symposium on

Caste and Race organized by the CIBA Foundation and held in London on
19, 20 and 21 April 1966. The conference was chaired by Gunnar Myrdal,
and papers were presented by G.D. Berreman, Louis Dumont, Edmund
Leach and Surajit Sinha, among others. These papers, along with a record
of the discussions, were published in a book, Caste and Race (de Reuck and
Knight, 1967). I had been invited to the conference, but in April/May 1966
I was lecturing at the Centre of Indian Studies in Paris at the invitation of
Professor Dumont; I decided to stay behind in Paris, although Professor
Dumont himself went to the conference.
2. This is Leach's phrase, applied by Dumont to the work of Berreman
and others in his contribution to the CIBA volume, (De Reuck and Knight,
1967, p. 28).
3. Statistics of atrocities against the Scheduled Castes and Schedules Tribes
are published annually in the Report of the Commission for Scheduled Castes
and Scheduled Tribes (Controller of Publications: New Delhi). Atrocities are
grouped under Murder, Violence, Rape, Arson and Others, and figures are
arranged state-wise. Atrocities have also been listed in Kamble (1981). For
278 Society and Politics in India
an account by a sociologist of the exploitation of Scheduled Caste women,
see Trivedi, 1976.
4 See, for instance, the March 1987 issue of The Lawyer's Colleaive. The 4th
National Conference on Women's Studies held at Andhra University, 28-31
December 1988 discussed several papers on the subject; these are, however,
not yet available in published form. ·
5 Davi!l, Gardner and Gardner could trace ten of the sixty-five black holdings
worth $900 or more to gifts by white fathers to their coloured offspring or
commonlaw wife. However, they also note that 'The evidence definitely
indicates that in the great majority of cases where real estate has been given
to coloured individuals by whites the relation from which the gift resulted
was based not upon kinship but upon sexual partnership' (1941, p. 298).
The point simply is that there is a code governing the relationship between
father and natural child, not that it is the same code as the one governing
the relationship between father and lawful child.
6. Traditionally, only sons could be adopted, and only by men, the choice of
the son to be adopted being governed by the idea of putrachhaya (putra son,
chhaya shadow), i.e. that he must bear the likeness of a real son. All this
has changed under the Hindu Adoptions and Maintenance Act, 1955 which,
among other things, ignores caste.
7. The literature on race and intelligence is voluminous and controversial.
Much of the recent controversy has centred on the question whether the
belief in the black's inherently inferior intelligence has a scientific basis.
Some say that it has and others that it does not have, but few would contest
that the belief itself is widespread. See Kamin, 1974.
8. I refer in particular to Marriott's various essays on village Kishan Garhi,
published in the 1950s and 1960s. Dumont's monograph on the Pramalai
Kallar, first published in French in 1957, and now available in English
(Dumont, 1986), is by any account one of the best monographs on any
Indian community. It is, however, notable that in his general work on India,
Homo hierarchicus, he has hardly referred to his own fieldwork, relying on the
fieldwork of others with which he could not have been equally familiar and
which he must often have judged to be inferior to his own.

SOURCES

AIR, All India Reporter (Government of India: New Delhi, 1976).

Berreman, G.D., 'Caste in India and the United States', American Journal of
Sociology 66, 1, 1960, pp. 120-7.
Berreman, G.D., 'Caste in cross-cultural perspective' in Japan's ltroisible
Race, G. De Vos and H. Wagatsuma (eds) (University of California Press:
Berkeley, 1966).
 Notes and sources 279
Berreman, G.D., 'Stratification, pluralism and interaction" in Caste and Race,
A. Reuck andJ. Knight (eds) ij. & A. Churchill: London, 1967).
Berreman, G.D., 'The concept of caste' in International Encyclopaedia of the
Social Sciences, D.L. Sils (ed.), vol. 2, 1968.
Cayton, H. and St C. Drake, Black Metropolis (Harcourt Brace: San
Diego, 1945).
Davis, A., B.B. Gardner and M.R. Gardner, Deep South (University of
Chicago Press: Chicago, 1941).
Davis, K., 'Intermarriage in caste society', American Anthropologist 43, 1941,.
pp.376-95. .
de Reuck, A. and J. Knight (eds), Caste and Race, Comparative Approaches Q.
& A. Churchill: London, 1967).
De Vos, G. and H. Wagatsuma (eds),Japan's Invisible Race: Caste in Culture
and Personality (University of California Press: Berkeley, 1966).
Dollard, J., Caste and class in a southern to111n (1937) (Doubleday: New
York, 1957).
Dumont, L.,'Caste, Racism and "stratification'", Contributions to Indian
Sociology 5, 1961, pp. 20-43.
Dumont, L. La civilisation indienne et nous (Armand Colin: Paris, 1964).
Dumont, L., Homo hierarchicus (Gallimard; Paris, 1966).
Dumont, L., 'Caste: a phenomenon of social stratification or an aspect of
Indian culture?' in Caste and Race, A. de Reuck and J. Knight (eds) Q. &
A. Churchill: London, 1967).
Dumont, L., A South Indian Sub-caste (1957) (Oxford University Press:
Delhi, 1986).
Fortes, M., Kinship and the Social Order (Routledge: London, 1969).
Freeman, J.M., Untouchable (Stanford University Press: Stanford, 1979).
Freeman, R.B., 1"e Black Elite. New York: (McGraw Hill: New York, 1976).
Ghurye, G.S., Caste and Race in India (1932) (Popular Prakashan: Bom-
bay, 1969).
Glazer, N. and D.P. Moynihan (eds), Ethnicity (Harvard University Press:
Cambridge Mass., 1975).
Hughes, E.C., 'Anomalies and projections', Daedalus 94, 4, 1965, pp.
1133-47.
lnden, R.B. and R.W. Nicholas, Kinship in Bengali Culture (University of
Chicago Press: Chicago, 1977).
Kamble, N.D., Atrocities on Scheduled Castes (Ashish Publishing House: New
Delhi, 1981).
Kamin, LJ., The Science and Politics of IQ (Lawrence Erlbaum Associates:
Hillsdale, NJ., 1974).
Kane, P.V., History ofDhannasastra (Bhandarkar Oriental Research Institute:
Poona, 1974) vol. 2, part 1.
Kapferer, B. (ed.), Transaction and Meaning (Institute for the Study of Human
Issues: Philadelphia, 1976).
280 Society and Politics in India
Levi-Strauss, C., Structural Anthropology (Basic Books: New York, 1963).
Manu, The Laws of Manu, tran. G. Biihler (Motilal Banarasidas: Delhi,
1964).
Marriott, M. 'Hindu transactions' in Transaction and Meaning, B. Kapferer
(ed.) (Philadelphia, 1976).
Marriott, M. and R.B. Inden, 'Caste systems' in Encyclopaedia Britannica, 15th
edn, 1980 Macropaedia, vol. 3.
Mayer, A.C. 1960. Caste and Kinship in Central India. (Routledge: Lon-
don, 1960).
Montagu, A., Man's Most Dangerous Myth (Oxford University Press: New
York, 1974).
Myrdal, G., An American Dilemma, The Negro Problem in Modern Democracy
(Harper & Row: New York 1944).
Myrdal, G., 'Chairman's Introduction' in Caste and Race, A. de Reuck and J.
Knight (eds) 0- & A. Churchill: London, 1967).
Parsons, T., 'Some theoretical considerations of the nature and trends of
change of ethnicity' in Ethnicity, N. Glazer and D.P. Moynihan (eds)
(Harvard University Press: Cambridge Mass., 1975).
Pinkney, A., The Myth of Black Progress (Cambridge University Press:
Cambridge, 1985).
Risley, H.H., The People ofIndia. (W. Thacker: Calcutta, 1908).
Schneider, D.M.,American Kinship (Prentice-Hall: Englewood Cliffs, 1968).
Shils, EA and H.A. Finch (eds), Max Weber on the Methodology ofthe Social
Sciences (The Free Press: Glencoe, 1949).
Trivedi, H.R., Scheduled Caste Women. (Concept: New Delhi, 1976).
Warner, W.L., 'American caste and class', American Journal ofSociology 42,
1, 1936, pp. 234-7.

Chapter 2 : Race, caste and ethnic identity

1. For an interesting discussion see Caste and Race, Comparative Approaches,
A. de Reuck and}. Knight (eds) (London, 1967).
2. Among the more notable community studies going back to this period are
John Dollard, Caste and Class in a Southern Town (New Haven, 1937) and
Allison Davis, Burleigh B. Gardner and Mary R. Gardner, Deep South, A
SocialAnthropological Study ofCaste and Class (Chicago, 1941).
3. G. Myrdal, An American Dilemma, The Negro Problem in Modern
Democracy (New York, 1944), pp. 667, 688.
4. P.L. van den Berghe, Race and Racism, A Comparative Perspective (New
York, 1967), speaks of whites, Africans, Asians and coloureds as constituting
the four 'castes' or 'colour-castes' of South African society.
5. G. Myrdal,AnAmerican Dilemma (New York, 1944).
6. A. Beteille, Castes: Old and New, Essays in Social Structure and Social
 Notes and sources 281
Stratification (Bombay, 1969), p. 3.
7. I. Karve, Hindu Society: An Interpretation (Poona, 1961), p. 16.
8. For American examples see the case studies by J. Dollard (Caste and Class
in a Southern Town (New Haven, 1937)) and by A. Davis, B. Gardner and
M. Gardner (Deep South, A Social Anthropological Study of Caste and Class
(Chicago, 1941)); for an Indian case study see E.K. Gough, 'Caste in a
Tanjore village' in Aspeas of Caste in South India, Ceylon and North-west
Pakistan, E.R. Leach (ed.) (Cambridge, 1960), p. 49.
9. G.D. Berreman, Hindus ofthe Himalayas (Berkeley, 1963), pp. 243-5.
10. See A. de Reuck and J. Knight (eds), Caste and Race, Comparative
Approaches (London, 1967).
11. L. Dumont, 'Caste, racism and "stratification": reflections of a social
anthroplolgist', Contributions to India Sociology 5, 1961, pp. 20-43.
12. E.R. Leach, 'Introduction: What we should mean by caste' in Aspeas
of Caste in South India, Ceylon and North-west Pakistan, E.R. Leach (ed.)
(Cambridge, 1960).
13. SJ. Tambiah presents this opposition as a 'gross simplification' in a
discussion reported in Caste and Race, A. de Reuck and J. Knight (eds)
(London, 1967),pp.328-9.
14. W. Lloyd Warner, 'Introduction: Deep South -A Social Anthropological
Study of Caste and Class' in Deep South, A Social Anthropological Study of
Caste and Class, A. Davis, B. Gardner and M. Gardner (eds) (Chicago, 1941),
pp. 3-14.
15. G. Myrdal,AnAmerican Dilemma (New York, 1944).
16. W. Lloyd Warner, 'Introduction' in Deep South, A Social Anthropological
Study of Caste and Class, A. Davis, B. Gardner and M. Gardner (eds)
(Chicago, 1941), pp. 3-14.
17. G.Myrdal,AnAmericanDilemma(NewYork, 1944),p.113.
18. Kingsley Davis, 'Intermarriage in caste society', American Anthropologist
43, 1941,pp.386-7.
19. L. Dumont, 'Caste: a phenomenon of social structure or an aspect
of Indian culture?' in Caste and Race, A. de Reuck and J. Knight (eds),
pp. 28-38.
20. L. Dumont, Homo hierarchicus (Paris, 1966).
21. I. Karve, Hindu Society: An Interpretation (Poona, 1961).
22. E.R. Leach (ed.), Aspects of Caste in South India, Ceylon and North-west
Pakistan (Cambridge, 1960).
23. G.D. Berreman, Hindus ofthe Himalayas (Berkeley, 1963), pp. 243-5.
24. A. Beteille, 'The politics of "non-antagonistic" strata', Contributions to
Indian Sociology, New Series 3, 1969, pp. 17-31 (see Chapter 5). One way
in which conflicts between castes were structured in the past was through
the opposition·between the 'right-hand' and the 'left-hand' castes prevalent in
many parts of South India; see J.H Hutton, Caste in India: Its Nature, Funaion,
and Origins (Bombay, 1961).
282 Society and Politics in India
25. G.D. Berreman, 'Caste in cross-cultural perspective', in Japan's Invisible
Race: Caste in Culture and Personality, G. De Vos and H. Wagatsuma (eds)
(Berkeley, 1966), p. 297.
26. G.D. Berreman, 'Caste in cross-cultural perspective', in Japan's Invisible
Race, G. De Vos and H. Wagatsuma (eds) (Berkeley, 1966), p. 297.
27. P.L. van den Berghe, Race and Rm:ism, A Comparative Perspective (New
York, 1967).
28. For typical village studies see A.C. Mayer, Caste and Kinship in Central
India, A Village and its Region (London, 1960); and A. Beteille, Caste,
Class, and Power, Changing Patterns of Stratification in a Tanjore Village
(Berkeley, 1965).
29. A. Beteille, Caste, Class, and Power (Berkeley, 1965).
30. A. Beteille, Caste, Class, and Power (Berkeley, 1965).
31. McKim Marriott, 'Caste ranking and food transactions: a matrix analysis'
in Structure and Change in Indian Society, M. Singer and B.S. Cohn (eds)
(Chicago, 1969), pp. 133-71.
32. R. Dahrendorf, Class and Class Conflict in an Industrial Society (Lon-
don, 1959).
33. M.N. Srinivas, 'Varna and Caste' in M.N. Srinivas, Caste in Modern India
and Other Essays (Bombay, 1962), pp. 63-9.
34. I. Karve, Hindu Society: Anlnterpretatwn (Poona, 1961).
35. G.S. Ghurye, Caste and Race in India (London, 1932).
36. I. Karve, Hindu Society: An Interpretation (Poona, 1961).
37. I. Karve and K.C. Malhotra, 'A biological comparison of eight endoga-
mous groups of the same rank', Cu"entAnthropology 9, 1968, pp. 109-16.
38. A. Beteille, Caste, Class, and Power (Berkeley, 1965).
39. A. Beteille, Caste, Class, and Power (Berkeley, 1965).
40. H.H. Risley, The People ofIndia (Calcutta, 1908), p. 29.
41. H.H. Risley, The People ofIndia (Calcutta, 1908), p. 29.
42. P.C. Mahalanobis, 'A revision ofRisley's anthropometric data', Samkhya
I, 1933, pp. 76-105; G.S. Ghurye, Caste and Race in India (London, 1932).
43. G.S. Ghurye, Caste and Race in india (London, 1932), p. 111.
44. G.S. Ghurye, Caste and Race in India (London, 1932), p. 107.
45. D.N. Majumdar and C.R. Rao, Race Elements in Bengal: A Q]Jantitative
Study (Calcutta, 1960).
46. D.N. Majumdar and C.R. Rao, Race Elements in Bengal (Calcutta,
1960), p. 102.
47. D.N. Majumdar and C.R. Rao, Race Elements in Bengal (Calcutta,
1960), p. 103.
48. I. Karve and K.C. Malhotra, 'A biological comparison of eight endoga-
mous groups of the same rank', Cu"entAnthropology 9, 1968, pp. 109-16.
49. I. Karve and K.C. Malhotra, 'A biological comparison of eight endoga-
mous groups of the same rank', Cu"entAnthropology 9, 1968, p. 115.
50. L.D. Sanghvi and V.R. Khanolkar, 'Data relating to seven genetical
 Notes and sources 283
characters in six endogamous groups in Bombay', Annals of Eugenics 15,
1950-1, pp. 52-76.
51. L.D. Sanghvi and V.R. Khanolkar, 'Data relating to seven genetical
characters in six endogamous groups in Bombay', Annals of Eugenics 15,
1950-1, p. 62.
52. A. Beteille, 'Race and descent as social categories in India', Daedalus 96,
1967, pp. 444-63.
53. In a discussion reported in Caste and Race, A. de Reuck and J. Knight
(eds) (London, 1967), pp. 110-11.
54. A. Beteille, 'Race and descent as social categories in India', Daedalus 96,
1967, pp. 444-63.
55. H.S. Morris, 'Ethnic Groups' in International Encyclopaedia ofthe Social
Sciences, David L. Sills (ed.), vol. 5, 1968, p. 167.
56. N. Glazer and D.P. Moynihan, Beyond the Melting Pot: The Negroes, Peurto
Ricans,Jtll)s, Italians, and Irish ofNt11J York City (Cambridge, Mass., 1963).
57. J.S. Fumivall, Colonial Policy and Practice: A Comparative Study ofBurma
and Netherlands India (New York, 1956).
58. F. Barth, 'Introduction' in Ethnic Groups and Boundaries, The Social
Organization ofCulture Difference, Fredrik Barth (ed.) (London, 1969, p. 14).
59. H. Eidheim, 'When ethnic identity is a social stigma' in Ethnic Groups
and Boundaries, F. Barth (ed.) (London, 1969), pp. 39-57.
60. F. Barth, 'Introduction' in Ethnic Groups and Boundaries (London,
1969), p. 10.
61. I am indebted for this information to Jonathan P. Parry who has made
an intensive study of the hill Rajputs in Kangra district.
62. L.I. Rudolph and S.H. Rudolph, The Modernity of Tradition: Political
Development in India (Chicago, 1967); A. Beteille, 'Caste and politics in
Tamilnadu' in Castes: Old and Nt11J (Bombay, 1969) (see Chapter 4).
63. N. Glazer and D.P. Moynihan, Beyond the Melting Pot (Cambridge,
Mass., 1963).
64. R. Kothari (ed.), Caste in Indian Politics (New Delhi, 1970).
65. L.I. Rudolph and S.H. Rudolph, 'The political role of India's caste
associations', Pacific Affairs 33, 1960, pp. 5-22.
66. S.S. Harrison, 'Caste and the Andhra communists', American Political
Science Review 50, 1956.
67. O.L. Lynch, The Politics ofUntouchability (New York, 1969).
68. A. Beteille, 'The future ofthe Backward Classes: the competing demands
of status and power', Perspectives, Supplement to the Indian Journal of Public
Administration 11, 1965, pp. 1-39 (see Chapter 7).
69. C. von Fiirer-Haimendorf, 'The position of the tribal population in
modem India' in India and Ceylon: Unity and Diversity, P. Mason (ed.)
(London, 1967), pp. 182-222.
70. D.N. Majumdar and C.R. Rao, Race Elements in Bengal: A fJ!Jantitative
Study (Calcutta, 1960).
284 Sodety and Politics in India

71. C. von Fiirer-Haimendorf, 'The position of the tribal population in

modem India' in India and Ceylon: Unity and Diversity, P. Mason (ed.)
(London, 1967), p. 188.
72. See, for instance, the issue of Seminar 24, August 1961, devoted to
Communalism.
73. A. Spear, 'The position of the Muslims, before and after partition'
in India and Ceylon: Unity and Diversity, P. Mason (ed.) (London, 1967),
pp. 33-4.
74. D.N. Majumdar and C.R. Rao, Race Elements in Bengal (Calcutta,
1960), p. 102.
75. S.S. Harrison, India: The Most Dangerous Decade (Bombay, 1960).
76. See, for instance, the issue of Seminar 23, July 1961, devo.ted to North
and South.
77. R.L. Hardgrave Jr., The Dravidian Movement (Bombay, 1965).
78. A. Beteille, 'Race and descent as social categories in India', Daedalus 96,
1967, pp. 444-63.

Chapter 3 : The concept oftribe with special reference

to India
1. R. Redfield, Peasant Society and Culture: An Anthropological Approach to
Civilization (Chicago University Press: Chicago, 1956); see also his Papers
2 vols, M.P. Redfield (ed.) (Chicago, 1962-3).
2. Kroeber's interest in civilizations is expressed in many of his writings,
published over a long period. A good sample is to be found in A.L.
Kroeber, Configurations of Culture Growth (University of California Press:
Berkeley, 1944).
3. See R. Redfield, Peasant Society and Culture (Chicago, 1956). The
phrase is due to A.L. Kroeber, Anthropology (Harcourt Brace: New York,
1948), p. 284.
4. N.K. Bose, The Struaure ofHindu Society, Translated with Introduction by
A. Beteille (Orient Longman: New York, 1975). K.S. Singh, Tribal Society in
India: An Anthropo-historical Perspeaive (Manohar: New Delhi, 1985).
5. L.H. Morgan, Ancient Society (1877) (Belknap Press: Cambridge, Mass.,
1964).
6. E. Durkheim, The Division of Labour in Society (1893) (The Free Press:
New York, 1933) was, as Sahlins has noted, the starting point of the
anthropological discussion of segmentary systems. See also E. Durkheim, The
Rules ofSociological Method (1895) (New York, 1938), especially Chapter 4.
7. N.K. Bose, Tribal Life in India (National Book Trust: Delhi, 1971);
S. Fuchs, Aboriginal Tribes of India (Macmillan: London, 1973); C.
von Fiirer-Haimendorf, Tribes of India (University of California Press:
Berkeley, 1982).
 Notes and sources 285
8. The phrase was popularized by D.N. Majumdar, A Tribe in Transition: A
Study in Culture Pattern (Longman: Calcutta, 1937).
9. E.R. Service, Primitive Social Organization (Random House: New York,
1962).
10. M.D. Sahlins, 'The segmentary lineage: an organization of predatory
expansion', American Anthropologist LXIII, 1961, pp. 322-45; Tribesmen
(Prentice-Hall: Englewood Cliffs, 1968). .
11. M.D. Sahlins, 'The segmentary lineage', American Anthropologist LXIII,
1961, p. 325.
12. E.E. Evans-Pritchard, The Nuer (Clarendon: Oxford, 1940).
13. M. Fortes and E.E. Evans-Pritchard (eds), African Political Systems
(Oxford University Press: London, 1940).
14. Sahlins makes no reference to Durkheim's work in his first paper
although the omission is made good in his book published seven years later.
See 'The segmentary lineage',AmericanAnthropologist LXIII, 1961, p. 325.
15. See E. Gellner, Saints ofthe Atlas (Weidenfeld and Nicholson: London,
1969); also his Muslim Society (Cambridge University Press: Cambridge,
1981). The pioneering anthropological work is E.E. Evans-Pritchard, The
Sanusi ofCyrenaica (Clarendon Press: Oxford, 1949).
16. F. Barth, Political Leadership among Swat Pathans (Athlone: London,
1959).
17. M.D. Sahlins, 'The segmentary lineage', American Anthropologist LXIII,
1961, p. 322.
18. M.D. Sahlins, Tribesmen (Englewood Cliffs, 1968), p. 20.
19. M. Godelier, Perspectives in Mar.mt Anthropology (Cambridge University
Press: London, 1977), translated from French: Horizons, trajets mamstes en
anthropologie (Maspero: Paris, 1973).
20. M. Godelier, Perspectives in Mar.mt Anthropology (London, 1977), p. 70.
21. M. Godelier, Perspectives in Mar.mt Anthropology (London, 1977), p.
87. This argument, as indc.ed Godelier's whole approach, leans heavily
on Engels.
22. M. Fortes and E.E. Evans-Pritchard (eds), African Political Systems
(London, 1940), p. 8.
23. See J.P. Digard, 'On the Bakhtiari' in The Conflict of Tribe and State
in Iran and Afghanistan, R. Tapper (ed.) (Croom-Helm: London, 1983), pp.
331-6. Digard asks, 'Moreover, how can it be maintained that a segmentary
structure is inherently contradictory with a class structure, when, as a matter
of/act, these two forms of organization co-exist and "function" simultaneously
in several societies, including the Bakhtiari?' (p. 332).
24. M. Fortes and E.E. Evans-Pritchard (eds), African Political Systems
(London, 1940), p. 8.
25. N.K. Bose, Tribal Life in India (Delhi, 1971).
26. B.K. Roy Burman, 'Transformation of tribes and analogous social
formations', Economic and Political Weekbi XVIII, 27, 2 July 1983, pp.
286 Society and Politics in India
1172-4; see also his 'The post-primitives of Chota Nagpur' in UNESCO,
Trends in Ethnic Group Relations in Asia and Oceania (UNESCO: Paris, 1979),
pp.102-41.
27. M.H. Fried, The Notion of Tribe (Cummings Publishing Co.: Menlo
Park, 1975).
28. M. Godelier, Perspectives in Marxist Anthropology (London, 1977).
29. M.H. Fried, The Notion ofTribe (Menlo Park, 1975), Chapter 2.
30. Writing about the Kamar and the Gond who have close cultural
affiliations, Fiirer-Haimendorf observed in 1951: 'The Kamars consider
the Gonds as their social superiors, but nevertheless they do not countenance
unrestricted social intercourse with their Gond neighbours. Sexual relations
with a Gond is sufficient ground for excommunication and a woman who eats
food cooked by Gonds is at once expelled from the tribal community. Aman,
on the other hand, may eat Gond food, but Kamars and Gonds do not freely
intermingle at feasts and ceremonies' (C. von Fiirer-Haimendorf, Foreward
in S.C. Dube, The Kamar (Universal Publishers: Lucknow, 1951), p. v).
31. M.H. Fried, The Notion ofTribe (Menlo Park, 1975), p. 30.
32. T. Trautman, Dravidian Kinship (Cambridge University Press: Cam-
bridge, 1981).
33. On this, however, there is some difference of opinion. See, for instance,
the three-part article by L. Dumont, 'Marriage in India: the present state of
the question'. Contributions to Indian Sociology 5, 1961, 7, 1964 and 9, 1966.
34. E.E. Evans-Pritchard, The Nuer (Oxford, 1940), Chapter 4.
35. M.H. Fried, 'On the concepts of "tribe" and "tribal society"' in Essays
on the Problem of Tribe, J. Helm (ed.) (American Ethnological Society:
Washington, 1968), p. 15.
36. E. Colson, The Makah Indians (Manchester University Press: Manches-
ter, 1953), p. 62.
37. E. Colson, 'Contemporary tribes and the development of nationalism' in
Essays on the Problem ofTribe, J. Helm (ed.) (Washington, 1968), p. 202.
38. J.W. Anderson, 'Khan and Khel: dialects of Pakhtun tribalism' in The
Conflict of Tribe and State in Iran and Afghanistan, R. Tapper (ed.) (London,
1983), p. 121.
39 The most illuminating contemporary discussion is in E. Gellner, Muslim
Society (Cambridge, 1981); see also his Saints ofthe Atlas (London, 1969).
40. D. Brooks, 'The enemy within: limitations on leadership in the Bakhtiari'
in The Conflict of Tribe and State in Iran and Afghanistan, R. Tapper (ed.)
(London, 1983), p. 338.
41. This remark is controversial, but here I follow D.D. Kosambi, An
Introduction to the Study ofIndian History (Popular Prakashan: Bombay, 1956),
especially Chapter 1.
42. N. Ray, Nationalism in India (Aligarh Muslim University: Aligarh, 1973);
N.K. Bose, Tribal Life in India (Delhi, 1971).
43. N. Ray, Nationalism in India (Aligarh, 1973), p. 123.
 Notes and sources 287

44. For an alternative formulation of the distinction see S. Bouez, Reciprocite

et hierarchie (Societe d'ethnographie: Nanterre, 1985), pp. 14--15.
45. D.D. Kosambi, An Introduction to the Study of Indian History (Bombay,
1956), p. 8.
46. D.D. Kosambi, An Introduction to the Study of Indian History (Bombay,
1956), pp. 26-30.
47. B.K. Roy Burman, 'The post-primitives ofChota Nagpur' in UNESCO,
Trends in Ethnic Group Relations in Asia and Oceania (Paris, 1979), p. 26.
48. Among the better known are H.H. Risley, The Tribes and Castes ofBengal,
2 vols (Bengal Secretariat Press: Calcutta, 1892); E. Thurston, Castes and
Tribes of Southern India, 7 vols (Government Press: Madras, 1909); W.
Crooke, The Tribes and Castes ofthe North-western Province and Oudh, 4 vols
(Calcutta, 1896); R.E. Enthoven, The Tribes and Castes of Bombay, 3 vols
(Government Central Press: Bombay, 1920-2).
49. D.D. Kosambi, An lntroduaion to the Study oftndian History (Bombay,
1956), p. 25; italics where capitalization in original.
50. I. Habib, 'Caste in Indian History', being the first of two Kosambi
Memorial Lectures delivered in Bombay in February 1985, ms.
51. I. Habib, 'Caste in Indian History', p. 12.
52. N.K. Bose, 'The Hindu method of tribal absorption', Science and Culture
VII (1941), pp. 188-94.
53. N.K. Bose, The Structure of Hindu Society, trans A. Beteille (New
York, 1975. ·
54. For a recent description of the melange of tribes and castes in Assam see
A. Candie, The Assamese (Curzon Press: Philadelphia, 1984). On the Khasi,
see P.R.T. Gurdon, The Khasis (Macmillan: London, 1914), and for a more
recent work, H. Barch, History and Culture ofthe Khasi People (Published by
the author: Calcutta, 1967).
55. F. Barth, Political Leadership among Swat Pathans (London, 1959). See
also F. Barth,'The system cf social stratification in Swat, North Pakistan'
in Aspeas of Caste in South India, Ceylon and North-west Paltistan, E.R. Leach
(ed.) (Cambridge, 1960), pp. 113-46.
56. E. Colson, The Maleah Indians (Manchester, 1953), p. 11.
57. E. Colson, The Maleah Indians (Manchester, 1953), p. 1.
58. B.K. Roy Burman, 'Transformation of tribes and analogous social
formations', Economic and Political Week?, XVIII, 27, 2July 1983.
59. B.K. Roy Burman, 'The post-primitives ofChota Nagpur' in UNESCO,
Trends in Ethnic Group Relations in Asia and Oceania (Paris, 1979), p. 112.
60. G.S. Ghurye, The Scheduled Tribes (Popular Book Depot: Bombay,
1959).
61. R. Tapper, 'Introduction' in The Conjlia of Tribe and State in Iran and
Afghanistan, R. Tapper (ed.) (London, 1983), pp. 46-7.
62. K.S. Singh, Tribal Society in India (New Delhi, 1985), especially
Chapter 6.
288 Sodety and Politics in India
63. G.S. Ghurye, The Scheduled Tribes (Bombay, 1959), articulates the
nationalist point ofview forcefully.
64. For a succinct account of the Constitutional position, see M. Galanter,
Competing Equalities (Oxford University Press: New York, 1984), pp. 147-53.
65. This case is described in K.S. Singh, Tribal Society in India (New Delhi,
1985), p. 80.
66. A. Beteille, Individualism and the Persistence ofCollective Identities (Univer-
sity of Essex: Colchester, 1984).

Chapter 4 : Caste and politics in Tamilnadu

I. R.W. Nicholas, 'Factions: a comparative analysis', Political Systems and the
Distribution ofPower, ASA Monographs 2 (London and New York, 1965).
2. P.R. Brass, Factional Politics in an Indian State: The Congress Party in Uttar
Pradesh (Berkeley and Los Angeles, 1965).
3. See A. Beteille, Caste, Class and Power: Changing Patterns ofStratification
in a Tanjore Village (Berkeley and Los Angeles, 1965).
4. E.A. Shils, 'Primordial, Personal, Sacred and Civil Ties', British Journal
ofSociology 8, 1957, pp. 130-45. ·
5. Cf. M.N. Srinivas, Caste in Modern India and ·other Essays (Bombay,
1962), p. 5.
6. See A. Beteille, Castes: Old and New, Essays in Social Structure and Social
Stratification (Bombay, 1969), Chapter 5.
7. S.F. Nadel, 'Dual Descent in the Nuba Hills' in African Systems of
Kinship and Marriage, A.R. Radcliffe-Brown and D. Forde (eds) (London,
1950), p. 337.
8. J.H. Hutton, Caste in lnd1'a, 3rd edn (Bombay, 1961), pp. 205-6.
9. See A. Beteille, Caste, Class and Power (Berkeley and Los Angeles, 1965),
especially Chapter 3.
10. M.N. Srinivas, Caste in Modern Indian and Other Essays (Bombay,
1962), p. 5.
11. See A. Beteille, Castes: Old and New (Bombay, 1969), Chapter 5.
12. M.N. Srinivas, Caste in Modern India and Other Essays (Bombay,
1962), p. 5.
13. For the concept of merging series, see J. Goody, 'The Fission of
Domestic Groups among the LoDagaba' in The Developmental Cycle in
Domestic Groups, J. Goody (ed.) (Cambridge, 1958), p. 60.
14. A. Beteille, Caste, Class and Power (Berkeley and Los Angeles, 1965),
Chapter 3.
15. A lineage is itself often internally segmented.
16. See A. Beteille, Castes: Old and New (Bombay, 1969), Chapter 1.
17. The Brahmins in Tamilnadu constitute between 3 and 4 per cent of the
total population.
 Notes and Sources 289
18. The Poondi Wandiyars, the Kapisthalam Muppanars and the Ukkadai
Thevars.
19. Quoted in C.V. Subba Rao, Life and Times of Sir KV. Reddi Naidu
(Rajahmundry, 1957), pp. 17-19. I am grateful to Miss G. Uma for having
drawn my attention to this book and for having provided me with other
interesting material on the non-Brahmin movement.
20. E.K. Gough, 'The social structure of a Tanjore village' in Village India,
McKim Marriott (ed.) (Chicago, 1955); 'Caste in a Tanjore village' in Aspects
of Caste in South India, Ceylon and North-111est Pakistan, E.R. Leach (ed.)
(Cambridge, 1960).
21. D. Sivertsen, When Caste Barriers Fall (New York, 1963).
22. A. Beteille, Caste, Class and Power (Berkeley and Los Angeles, 1965).
23. E.K. Gough, 'The social structure of a Tanjore village' in Village India,
McKim Marriott (ed.) (Chicago, 1955), p. 38.
24. This was written when the Congress was in power; it applies even more
to the present situation.
25. S. Harrison shows the tie-up between the Communists and the
DK in his book, India: The Most Dangerous Decades (Bombay, 1960),
pp.182-90.
26. D. Sivertsen, When Caste Barriers Fall (New York, 1%3), p. 126.
27. R. Dahl, Who Governs? (New Haven, 1961).
28. Quoted in C.V. Subba Rao, Life and Times of Sir KV. Reddi Naidu
(Rajahmundry, 1957), pp. 17-23.
29. C.V. Subba Rao, Life and Times (Rajahmundry, 1957), p. 30.
30. C.V. Subba Rao, Life and Times (Rajahmundry, 1957), pp. 30-9.
31. C.V. Subba Rao, Life and Times (Rajahmundry, 1957), p. 53.
32. R.L. Hardgrave Jr., The Dravidian Muvement (Bombay, 1965), p. 19.
33. R.L. Hardgrave Jr., The DrllVidian Muvement (Bombay, 1965), p. 21.
34. Quoted in Subba Rao, Life and Times (Rajahmundlkry, 1957),
p.19.
35. K.B. Krishna, The Problem ofMinorities (London, 1939), pp. 154-5.
36. R.L. Hardgrave Jr., The Dravidian Muvement (Bombay, 1965), p. 16.
37. M.N. Srinivas, 'The social system of a Mysore village' in Village India,
McKim Marriott (ed.) (Chicago, 1955).
38. M.N. Srinivas, 'The social system of a Mysore village' in Village India,
McKim Marriott (ed.) (Chicago, 1955), p. 18.
39. M.N. Srinivas, 'The social system of a Mysore village'.
40. In this sense, Kumbapettai, Thyagasamudram and Sripuram are excep-
tional rather than general.
41. L.I. Rudolph and S.H. Rudolph, 'The political role of India's caste
associations', Pacific Affairs XXXIII, 1, 1960, pp. 5-22.
42 The Mail, 13 February, 1962.
43. C.V. Subba Rao, Life and Times (Rajahmundry, 1957), p. 107.
44. A. Beteille, Caste, Class and P(JJl)er (Berkeley and Los Angeles, 1965).
290 Sodety and Politics in India
45. The Mail, 9 June 1961.
46. This case was reported in detail in the September and October 1957
issues of The Hindu; see, in particular, the issue of 15 September 1957, 16
September 1957 and 22 September 1957.
47. M.N. Srinivas, Caste in Modern India and Other Essays (Bombay,
1962), p. 75.
48. E.K. Gough, 'The social structure of a Tanjore village' in Village India,
McKim Marriott (ed.) (Chicago, 1955).
49. M. Weber, The Religion ofIndia (New York, 1958).
50. I am grateful to Mr E.A. Ramaswamy for the important information
on Coimbatore where he is making an intensive study of textile work-
ers' unions.
51. E.R. Leach (ed.), Aspects of Caste in South India, Ceylon and North-'lfflt
Pakistan (Cambridge, 1960), p. 6.
52. E.R. Leach (ed.),Aspects ofCaste (Cambridge, 1960), p. 7.
53. E.R. Leach (ed.),Aspects ofCaste (Cambridge, 1960), p. 7.
54. See A. Beteille, Castes: Old and New (Bombay, 1969).
55. P.R. Brass, Factional Politics in an Indian State (Berkeley and Los
Angeles, 1965).
56. L.I. Rudolph and S.H. Rudolph, 'The political role of India's caste
associations', Pacific Affein XXXIII, 1, 1960, pp. 5-22. ·
57. R.A. Dahl, Who Guvems? (New Haven, 1961), p. 34.
58. P.R. Brass, Factional Politics in an Indian State (Berkeley and Los
Angeles, 1965).
59. S.M. Lipset, Political Man (London, 1963), p. 31.

Chapter S: The politics of'non-antagonistic' strata

1. In talking about Bailey in the following pages, I shall confine myself
to the views expressed by him in this paper which is probably his most
comprehensive general statement on caste. These seem to me to be at
variance with some of his earlier views which I think would support the
basic argument of this chapter.
2. There is an interesting discussion in R.W. Nicholas 'Factions: a
comparative analysis', Political Systems and the Distribution of Po'fl)er, ASA
Monograph 2 (London, 1965).
3. For an interesting discussion see S. Ossowski, Class Structure in the Social
Consciousness (London, 1963), pp. 110-18.
4. At this level the caste system is above all a system of ideas and of values, a
formal system, comprehensive ·and rational, a system in the intellectual sense
of the term.
5. Our first task consists in grasping this intellectual system, this ideology.
 Notes and Sources 291
SOURCES
Aron, R., La Lune de classes (Paris, 1964).
Bailey, F.G., 'For a sociology of India', Contributions to Indian Sociology
III, 1959.
Beteille, A., 'A note on the referents of caste', European Journal of
Sociology V, 1964.
Dahrendorf, R., Class and Class Conflict in Industrial Society (London, 1959).
Dumont, L., Homo hierar&hicus (Paris, 1966).
Frykenberg, R.E., Guntur District, 1788-1848 (Oxford, 1965).
Frykenberg, R.E., (ed.), Land Control and Social Sttucture in Indian History
(Madison, WISc., 1969).
Lange, 0. (ed.), 'Political Economy of Socialism' in Problems of Political
Economy ofSocialism, 0. Lange (ed.) (Delhi, 1962).
Leach, E.R., Political Systems ofHighland Burma (London, 1954).
Leach, E.R., 'What we should mean by caste' in Aspects ofCaste in South India,
Ceylon and North-T11eSt Pakistan, E.R. Leach (ed.) (Cambridge, 1960).
Lockwood, D., 'Some remarks on "the social system"', British Journal of
Sociology VII, 1956.
MacPherson, C.B., The Real World ofDemocracy (Oxford, 1966).
Mannheim, K., Ideology and Utopia (London, 1960).
Marcuse, H., Suviet Mar.mm (New York, 1961).
Mayer, A.C., 'Change in a Malwa Village', Economic Weekly, 24 Septem-
ber 1955.
Metcalfe, T.R., 'From Raja to landlord: the Oudh Talukdars, 1850-1870' in
Land Control and Social Sttucture in Indian History, R.E. Frykenberg (ed.)
(Madison, WISC., 1969).
Nemchinov, V.W., 'Changes in the class structure of the Soviet Union',
Transactions ofthe Third World Congress ofSociology VIII (London, 1957).
Nicholas, R.W., 'Factions: a comparative analysis', Political Systems and the
Distribution ofPMPer, ASA Monograph 2 (London, 1965).
Ossowski, S., Class Sttucture in the Social Consciousness (London, 1963).
Ossowski, S., 'Old notions and new problems: interpretations of social
structure in modem society', Transactions of the Third World Congress of
Sociology III (London, 11956).
Srinivas, M.N., Social Change in Modern India (Berkeley, 1966).
Wesolowski, W., 'Les notions de strates et de classe dans la societe socialiste',
Sociologiedu Tr11Vail1X, 1967.

Chapter 6 : Networks in Indian social structure

I. E.E. Evans-Pritchard, The Nuer (Oxford, 1940).
2. E.E. Evans-Pritchard, The Nuer (Oxford, 1940).
292 Society and Politics in India
3. This procedure, it may be noted, delimits only the 'formal' structure. The
understanding of the 'informal' or operative structure requires additional
tools. The delimitation of the formal structure is, however, an essential
preliminary to the delineation of the informal structure.
4. McKim Marriott (ed.), Village India (Chicago, 1955).
5. E.E. Evans-Pritchard, 'Kinship and the local community among the Nuer'
in African Systems of Kinship and Marriage, A.R. Radcliffe-Brown and D.
Forde (eds) (Oxford, 1950), p. 365.
6. E.E. Evans-Pritchard, The Nuer (Oxford, 1940), p. 264 (our italics).
7. This process was grasped, with rare perception, by Emile Durkheim in
The Division ofLabour in Society (1893) (fhe Free Press: New York, 1933).
Durkheim showed how the segmental structure with its sharply defined
outlines gave place to the organized structure with its interpenetration
of groups with the change from mechanical to organic solidarity (pp.
177-90, 256-75).
8. This is not to deny that networks existed in traditional society also. Their
role in contemporary India has, however, become more important. For a brief
discussion on networks in traditional India see p.
9. The village was (and still is) clearly divided into three territorial segments:
the agraharam (where .Brahmins live), the non-Brahmin streets and the cheri
(inhabited by untouchables). ·
10. See J.A. Barnes, 'Classes and committees in a Norwegian Island parish',
Human Relations VII, I, 1954, pp. 39-58.
11. J.A. Barnes, 'Classes and Committees', Human Relations VII, I, 1954,
pp. 39-58.
12. I. Karve, 'What is caste? (1) Caste as extended kin', The Economic Weekly,
Annual Number, January 1958.
13. This is not contradicted by the fact that the network of formal kinship
relations (based upon genealogical ties rather than effective exchange of
obligations) expands with an expansion of the limits of endogamy.
14. M.J. Levy Jr., The Structure ofSociety (New Jersey, 1952), pp. 88ff.
15. F.G. Bailey, 'Politics and society in contemporary Orissa' in Politics and
Society in India, C.H. Philips (ed.) (Londonl, 1963).
16. I. de Sola Pool, 'The mass media and politics in the modernization
process' in Communications and Political Development, L.W. Pye (ed.) (New
Jersey, 1963).
17. See, for instance, M. Singer, 'The Social Organization of Indian
Civilization', Diogenes, Wmter 1964.

Chapter 7 : The future of the Backward Classes

I. Comprising the Scheduled Tribes, the Scheduled Castes and the Other
Backward Classes; the Denotified Tribes, who constitute a small minority,
 Notes and Sources 293
are not considered separately here. For a general account, see Lelah Dushkin,
'The Backward Classes', The Economic and Political Weekly 29 October, 4 and
18 November, pp. 1665-9, 1695-1706, 1729-38.
2. Karl Mannheim, Ideology and Utopia (New York, 1936).
3. Gunnar Myrdal, Value and Social Theory (London, 1958).
4. Marc Galanter, 'Equality and "Protective Discrimination" in India', Rutgers
Law Review XVI, 1, 1961, pp. 42-74.
5. A. Beteille, 'Question of Definition' in Tribal India, Seminar 14, 1960,
pp. 15-18.
6. B.S. Guha, 'Indian Aborigines and Who They Are', in The Adivas.is (Delhi,
1955), p. 31.
7. Government of India, Report of the Backward Classes Commission (New
Delhi, 1956), p. 48.
8. The author is grateful to Mr L.P. Singh, Home Secretary, Government
oflndia, for supplying this information.
9. M.N. Srinivas, 'Sanskritization and Westernization' in Society in India,
Aiyappan and Bal Ratnam (eds) (Madras, 1956).
10. M.N. Srinivas, Religion and Society Among the Coorgs of South India
(Oxford, 1952).
11. M.N. Srinivas et al., Caste: A Trend Report and Bibliography (Oxford,
1959), p. 140.
12. M.N. Srinivas, 'The Dominant Caste in Rampura', American Anthropolo-
gist 61, 1, 1959,pp. l-16.
13. M.N. Srinivas, 'Sanskritization and Westernization' in Society in India,
Aiyappan and Bal Ratnam (eds) (Madras, 1956).
14. J.H. Hutton, Caste in India (Bombay, 1961), p. 205.
15. J.H. Hutton, Caste in India (Bombay, 1961), p. 205.
16. J.H. Hutton, Caste in India (Bombay, 1961), p. 206.
17. F.G. Bailey, Politics and Social Change: Orissa in 1959 (Berkeley,
1963), p. 51.
18. Government of India, Report ofthe Commissioner for Scheduled Castes and
Scheduled Tribes for the Year 1961-62 (New Delhi, 1963), p. 6.
19. The author is grateful to his colleague, Mr G.S. Aurora, for the
information on the Meenas.
20. M.N. Srinivas, 'Sanskritization and Westernization', in Society in India,
Aiyappan and Bala Ratnam (eds) (Madras, 1956).
21. Harold R. Isaacs, 'A Reporter at Large, The Ex-Untouchables' I, The
New Yorker(December, 1964), pp.138-40.
22. Courtesy Mr L.P. Singh.
23. Harold R. Isaacs, 'A Reporter at Large, The Ex-Untouchables' I, The
New Yorker (December, 1964), pp. 138-40.
24. Harold R. Isaacs, 'A Reporter at Large, The Ex-Untouchables' II, p.75.
25. Martin Orans, The Santai, A Tribe in Search ofa Great Tradition (Detroit,
1965), p. 101.
294 Soci.ety and Politics in India
26. Myron Weiner, 'The Politics of South Asia' in The Politics of the
DevelopingAreas, Almond and Coleman (eds) (Princeton, 1960), p. 192.
27. F.G. Bailey, Politics and Social Change: Orissa in 1959 (Berkeley, 1963).
28. The Weekry Mail, 18 February 1962.
29. L.P. Vidyarthi, 'The historical March of the Jharkhand Party: A.
Study of Adivasi Leadership in Tribal Bihar', Sociological Bulletin 1, 2,
1964, p. 5.
30. N.K. Bose, 'Change in Tribal Cultures Before and After Independence',
Man in India 44, 1, 1964, p. 5.
31. N.K. Bose, 'Change in Tribal Cultures Before and After Independence',
Man in India 44, 1, 1964, p. 7.
32. F.G. Bailey, Politics and Social Change: Orissa in 1959 (Berkeley,
1963), p. 41.

Chapter 8 : Equality as a right and as a policy

1. R. Dworkin, 'Reverse Discrimination', in R. Dworkin, Taking Rights
Serious!), (Duckworth: London, 1977), pp. 223-39.
2. For a discussion of equality as conceived in tenns of the 'anti-
discrimination' principle, see O.M. Fiss, 'Groups and the Equal Protection
Clause' in Equality and Preferential Treatment, M. Cohen, T. Nagel and T.
Scanlon (eds) (Princeton University Press: Princeton, 1977), pp. 84-154.
3. R.H. Tawney, Equality (1931) (Unwin: London, 1964).
4. The most consistent exposition of this point of view is in F.A. Hayek, The
Constitution of Liberty (Routledge: London, 1960), but see also R. Nozick,
Anarchy, State and Utopia (Basil Blackwell: Oxford, 1974), and M. Friedman,
Capitalism and Freedom (University of Chicago Press: Chicago, 1962).
5. B. Sivaramayya, 'Equality and inequality: the legal framework' in Equality
and Inequality: Theory and Practice, A. Beteille (ed.) (Oxford University Press:
Oxford, 1983), pp. 28-70.
6. The classic formulation of this linkage is in A. de Tocqueville, Democracy
in America, 2 vols (Knopf: New York, 1956), but see also A. Beteille,
'Individualism and Equality', Cu"ent Anthropology 27, 2, April 1986, pp.
121-34 (see Chapter 9). ·
7. A very good historical account is to be found in J.R. Pole, The Pursuit of
Equality in American History (University of California Press: Berkeley, 1978).
8. R. Dworkin, 'Reverse Discrimination' in R. Dworkin, Taking Rights
Serious!), (Duckworth: London, 1977), p. 232.
9. R. Dworkin, 'Reverse Discrimination', p. 227.
10. This was the position consistently adopted in the Constituent Assembly
by B.R. Ambedkar, Chairman of the Drafting Committee and independent
India's first Minister of Law. See A. Beteille, Individualism and the Persistence
ofCollective Identities (University of Essex: Colchester, 1984).
 Notes and Sources 295
11. L. Dumont, From Mandeville to Ma,:r (University of Chicago Press:
Chicago, 1977); L. Dumont, Essais sur l'individualisme (Seuil, 1983);
S. Lukes, Individualism (Basil Blackwell: Oxford, 1973); A. Beteille,
'Individualism and Equality', Current Anthropology 27, 2, April 1986, pp.
121-34. (see chapter 9).
12. L. Dumont, Homo hierarchi&us (Paladin: London, 1972); N.K. Bose, The
Struaure ofHindu Society (Orient Longman: Delhi, 1975). See also the Kale
Memorial Lecture by M.P. Rege, Concepts ofJustice and Equality in the Indian
Tradition (Gokhale Institute: Pune, 1985).
13. B. Sivaramayya, Inequalities and the Law (Eastern Book Co.: Lucknow,
1984), p. 5.
14. I have discussed some of the ambiguities of the official classification in
my Commonwealth Lectures, 'Positive Discrimination and Social Justice',
delivered in the University of Cambridge, April - May, 1985.
15. Tara Patel (ed.), Remuval of Untoudu,bility in Gujarat (Gujarat Univer-
sity: Gujarat, 1973); I.P. Desai, Untouchability in Rural Gujarat (Popular
Prakashan: Bombay, 1976). See also M.J. Moffatt,An Untouchable Community
ofSD1'th India (Princeton University Press: Princeton, 1979).
16. R.G. Karmarkar, The Protection ofCivil Rights Aa, 1955 (N.R. Bhalerao:
Pune, 1978), p. v.
17. J. Woodburn, 'Egalitarian societies',Man 17, 1982, pp. 431-51.
18. See my 'Equality of Opportunity and the Equal Distribution of Benefits',
Arthavijnana 27, 2, 1985, pp. 96-114.
19. The empirical material on the association between caste and occupation
is uneven. See V.S. D'Souza, Inequality and its Perpetuation (Manohar: New
Delhi, 1981); J. Sarkar, Caste, Occupation and Change (B.R. Publishing
Corporation: Delhi, 1984).
20. R.H. Tawney, Equality (Unwin: London, 1964), p. 103.
21. O.M. Fiss, 'Groups and the Equal Protection Clause', in Equality and
Preferential Treatment (Princeton, 1977), pp. 84-154.
22. They speak of 'goals' rather than 'quotas'. For a balanced account, see
R.A. Rossum, kvme Discrimination (Marcel Dekker: New York, 1980).
23. The only quotas specified by the Constitution are those relating to
sears in Parliament and in the state Legislatures, but these are subject to
a time limit.
24. For a detailed discussion see M. Galanter, Competing Equalities (Oxford
University Press: Oxford, 1984).
25. R. Dworkin, 'Why Bakke Has No Case' Neri) York Revier/) of Books, 10
November 1977.
26. M. Weiner et al., India's Preferential Policies (University of Chicago Press:
Chicago, 1981).
27. R. Dworkin, 'Reverse Discrimination' in R. Dworkin, Taking mghts
Seriously (Duckworth: London, 1977), p. 225.
28. B. Sivaramayya, Inequalities and the la'IIJ (Lucknow, 1984), p. 36.
296 Sodety and Politics in India
However, other interpretations of the relationship between Articles 14
and 16 may be found in the State of Kerala vs. N.M. Thomas, AIR,
1976: SC 490.
29. ConstituentAssembly Debates: Official Report, vol. 7, 1950, p. 702.
30. AIR, 1976: SC 527.
31. AIR, 1976: SC 536.
32. B. Sivaramayya, Inequalities and the Law (Lucknow, 1984), p. 89.
33. A. Beteille, Individualism and the Persistence of Collective Identities
(Colchester, 1984). ·
34. Karnataka Backward Classes Commission, Report, Government of
Kamataka (Bangalore), 1975.
35. B. Sivaramayya, Inequalities and the Law (Lucknow, 1984), pp. 88-91.

Chapter 9 : Individualism and equality

1. Dumont (1972, p. 42), for instance, writes: 'As opposed to modem society
traditional· societies . . . know nothing of equality and liberty as values ...
known nothing, in short, of the individual'.
2. This is the title of the French edition; the English edition, published
simultaneously, is called From Mandeville to Man". The subtitle has roughly
the same meaning in both editions.
3. This began to change from the middle of the nineteenth century onwards.
The first major step was the Caste Disabilities Removal Act of 1850.
4. This view was widely expressed in the press when large concessions were
announced in Gujarat and Madhya Pradesh on the eve of the elections in
1985 and, earlier, in Bihar in 1977.
5. Anti-reservation agitation and riots began in Gujarat and Madhya
Pradesh in March 1985 and continued with great severity in Gujarat for
several months.
6. These issues have in the last few years been discussed extensively in the
national press, especially in The Times ofIndia.
7. A recently retired judge of the Supreme Court writes: 'It was in the
nineteenth century that the demand for civil rights came to be articulated
by the Indians. The birth of the Indian National Congress provided the
platform to give expression to the desire for civil rights which then
meant equality with the Englishmen' (Reddy, 1976, pp. 10-11). See also
Ganguli (1975).
8. Cf. Ballhatchet (1979, p. 6): 'Competitive examinations for the Indian Civil
Service began in 1855, and an Indian candidate, Satyendranath Tagore, was
successful in 1863. The Civil Service Commissioners reacted in characteristic
fashion by manipulating the marking scheme so as to impede subsequent
Indian candidates'.
 Notes and Sources 297
SOURCES
Arnold, M., 'Equality' in Mised Essays Gohn Murray: London, 1903),
pp. 48-97.
Ballhatchet, K., Race, Sex, and Class Under the Raj (Vikas: New Delhi, 1979).
Bankimchandra Chatterji, (1382 Bengali calendar) Bankim Rachanabali (in
Bengali) (Sahitya Samsad: Calcutta, n.d.), vol. 2.
Beteille, A., 'Homo hierarchicus, Homo equalis', Modem Asian Studies 13,
1979, pp. 529-48.
Beteille, A., The Idea of Natural Inequality (London School of Economics:
London, 1980).
Beteille, A., The Backward Classes and the New Social Order (Oxford
University Press: Delhi, 1981).
Beteille, A., Individualism and the Persistence of Collective Identities
(University of Essex: Colchester, 1984).
Beteille, A., Equality of Opportunity and the Equal Distribution of Benefits
(Kale Memorial Lecture, Gokhale Institute of Politics and Economics,
Poona, 1986).
Beteille, A., Positive Discrimination and Social Justice (Commonwealth Lec-
tures, Cambridge University), ms., n.d.
Burckhardt, J., The Civilization of the Renaissance in Italy (1860) (Random
House: New York, 1954).
Coleman, J.S., 'Equality of opportunity and equality of results', Harvard
Educational Review 43, 1, 1973, pp. 129-37. Constituent Assembly Debates:
Official Report, vol. 7 (New Delhi, 1950).
Duby, G., The Three Orders (University of Chicago Press: Chicago, 1980).
Dumont, L., Homo hierarchicus (Paladin: London, 1972).
Dumont, L., From Mandeville to Mar.r (University of Chicago Press: Chicago,
1977).
Dumont, L., 'On value', Proceedings of the British Academy 66, 1980,
pp. 207-41.
Dumont, L., Essais sur l'individualisme (Seuil: Paris, 1983).
Durkheim, E., 'Individualism and the intellectuals' (1898) trans S. Lukes and
J. Lukes, Political Studies 17, 1969, pp. 114-30.
Dworkin, R., 'DeFunis vs Sweatt' in Equality and Preferential Treatment, M.
Cohen. T. Nagel and T. Scanlon (eds) (Princeton University Press:
Princeton, 1977).
Elias, N., The Court Society (Basil Blackwell: Oxford, 1983).
Fiss, O.M., 'Groups and the equal protection clause' in Equality and
Preferential Treatment, M. Cohen, T. Nagel and T. Scanlon (eds) (Princeton
University Press: Princeton, 1977), pp. 84-154.
Friedman, M.! Capitalism and Freedom (University of Chicago Press: Chicago,
1962).
Galanter, M., Competing Equalities (Oxford University Press: London, 1983).
298 Society and Politics in India
Galton, F., Hereditary Genius (1869) (Watts: London, 1950).
Ganguli, B.N., Concept of Equality (Indian Institute of Advanced Studies:
Simla, 1975).
Glazer, N.,Affirmative Discrimination (Random House: New York, 1975).
Goldman, A.H., Justice and Reverse Discrimination (Princeton University Press:
Princeton, 1979).
Government of India, Report of the Bacleward Classes Commission (Manda/
Commission) 4 (New Delhi, 1981).
Hayek, F.A., The Constitution ofLiberty (Routledge: London, 1960).
Hayek, F.A., 'Individualism: true and false' (1946) in Individualism and
Economic Order (University of Chicago Press: Chicago, 1980), pp. 1-32.
Holmstrom, M., Industry and Inequality (Cambridge University Press: Cam-
bridge, 1984).
Huizinga, J., The Waning ofthe Middle Ages (Edward Arnold: London, 1924).
Huxley, T.H., 'On the natural inequality of man', The Nineteenth Century no.
155, 1890,pp. 1-23.
Joseph, K. andJ. Sumption, Equality Oohn Murray: London, 1979).
Letwin, W. (ed.),Against Equality (Macmillan: London, 1983a).
Letwin, W., 'The case against equality' in Against Equality, W. Letwin (ed.)
(Macmillan: London, 1983b).
Levi-Strauss, C., Structural Anthropology (Basic Books: New York, 1963).
Lucas,J.R., 'Against equality', Philosophy, October 1965, pp. 296-307.
Lukes, S., Individualism (Basil Blackwell: Oxford, 1973).
Macfarlane, A., The Origins ofEnglish Individualism (Basil Blackwell: Oxford
1978).
Maine, H.S., Lectures on the Early History ofInstitutions (1875) Oohn Murray:
London, 1914).
Maine, H.S.,Ancient Law (1861) (Oxford University Press: Oxford, 1950).
MalinoV:'ski, B., Crime and Custom in Savage Society (Routledge: London,
1926).
Nozick, R.,Anarchy, State and Utopia (Basil Blackwell: Oxford, 1974).
Ossowski, S., Class Structure in the Social Consciousness (Routledge: London,
1963).
Pollock, F. and F.W. Maitland, The History ofEnglish Law Before the Time of
Edward I (1895) (Cambridge University Press, 1968), vol. 1.
Rawls,J.,A Theory ofJustice (Oxford University Press: London, 1973).
Reddy, P. Jaganmohan, Social Justice and the Constitution (Andhra University
Press: Waltair, 1976).
Rossum, R.A., Reverse Discrimination (Marcel Dekker: New York, 1980).
Simmel, G., 'Individual and society in eighteenth- and nineteenth-century
views of life' (1917) in The Sociology ofGeorg Simmel, K. Wolff (ed.) (Free
Press: Glencoe, 1950), pp. 58-84.
Simmel, G., On Individualism and Social Forms, D.N. Levine (ed.) (University
of Chicago Press: Chicago, 1971).
 Notes and Sources 299
Sivaramayya, B., 'Equality and inequality: the legal framework' in Equality and
Inequality, A. Beteille (ed.) (Oxford University Press: Delhi, 1983).
Srinivas, M.N., India 2000 A.D.: The Next Fifteen Years (Austin: University of
Texas Press, 1986).
Tocqueville, A. de, Democracy in America, 2 vols (1835-40) (Knopf: New
York, 1956).
Tripathi, P.K., Some insights into Fundamental Rights (University of Bombay:
Bombay, 1972).
Williams, B., 'The idea of inequality' in Philosophy, Politics, and Society,
2nd series, P. Laslett and W.G. Runciman (eds) (Basil Blackwell:
Oxford, 1962).

Chapter 10 : Individual and person as subjects for

sociology
1. There is a long line of prominent American sociologists who have written
about the self, from G.H. Mead, Mind, Selfand Society (University ofChicago
Press: Chicago, 1934) to E. Goffinan, The Presentation of Self in Everyday
Life (Doubleday: 1959).
2. For a recent critical account see A. Bharati, 'The self in hindu thought
and action' in A.J. Marsella, G. DeVos and F.L.K. Hsu (eds) Culture and Self,
(favistock Publications: London, 1985), pp. 185-230.
3. E. Durkheim, 17,e Division of Labour in Society (1893) (Macmillan:
London, 1984.
4. E. Durkheim, 17,e Elementary Forms ofReligious Life (1912) (Allen & Unwin:
London, 1915), p. 270.
5. M. Mauss, 'A category of the human mind: the notion of person, the
notion of "self"' in his Sociology and Psychology (1938) (Routledge: London,
1979), pp. 57-94.
6. N. Luhmann, 'The individuality of the individual: historical meanings
and contemporary problems' in T.C. Heller et al. (eds), Reconstructing
Individualism, (Stanford University Press: Stanford, 1986), p. 313.
7. Two of his most influential pupils, Evans-Pritchard, Professor of
Social Anthropology at Oxford (1946-70) and Fortes, Professor of Social
Anthropology at Cambridge (1950-73) both became increasingly critical
of Malinowski, if not hostile to him, although Malinowski's intellectual
influence was very strong on them, particularly on Evans-Pritchard. For
derogatory and hostile remarks made by the former in print, see E.E.
Evans-Pritchard, Social Anthropology (Cohen and West: London, 1951), pp.
93-6, and A History ofAnthropological 17,ought (Faber and Faber: London,
1981), pp. 197-200; what was put down in print was mild compared to what
was said in coversation. Fortes was more guarded, but far from flattering
in his references to Malinowski in M. Fortes, Kinship and the Social Order,
300 Society and Politics in India
(Routledge: London, 1969), pp. 4-5.
8. See the Introduction by Levi-Strauss in M. Mauss, Sociologie et
Anthropologie, (Presses Universitaires de France: Paris, 1960), pp. ix-Iii.
See also the introduction in L. Dumont, Essais sur l'individualisme (Seuil:
Paris, 1983), pp. 11-30, especially pp. 11-18.
9. Malinowski is not so much as mentioned in a recent Oxford symposium on
the person: M. Carrithers, S. Collins and S. Lukes (eds), The Category ofthe
Person (Cambridge University Press: Cambridge, 1985), whereas the editors
took pains to make a fresh English translation of Mauss's essay on the person
to serve as the opening chapter in the volume.
10. Meyer Fortes has left a brief but vivid account of Mauss on that occasion.
See M. Fortes, 'The Concept of Person' in his Religion, Morality and the
Person (Cambridge University Press: Cambridge, 1987).
11. See M. Mauss 'A category of the human mind: the notion of person, the
notion of "self"' in his Sociology and Psychology (London, 1979), pp. 57-94
and M. Carrithers, S. Collins and S. Lukes (eds), The Category ofthe Person
(Cambridge, 1985).
12. E. Durkheim, The Division ofLabour in Society (London, 1984).
13. E. Durkheim, 'Individualism and the Intellectuals' (1898), trans S. Lukes
and J. Lukes, Political Studies 17, 1969, pp. 114-30.
14. M. Mauss, 'A category of the human mind' in M. Mauss, Sociology and
Psychology (London, 1979).
15. The point was made forcefully in his celebrated Marett Lecture, 'Social
anthropology: past and present', republished in E.E. Evans-Pritchard, Essays
in Social Anthropology (London, 1951). See also 'The position of women in
primitive societies' in his The Position ofWomen in Primitive Societies and Other
Essays in Socia/Anthropology (Faber and Faber: London, 1965), pp. 37-58; and
his Theories ofPrimitive Religion (Clarendon Press: Oxford, 1965).
16. Durkheim, The Division ofLabour in Society (London, 1984), p. 84.
17. M. Mauss, The Gift (1925), (Routledge: London, 1954), p. 3.
18. Compare C. Geertz, Works and Lives (Polity Press: Cambridge, 1988)
on Evans-Pritchard: 'On the Akobe as the Isis, men and women are brave
and cowardly, kind and cruel, reasonable and foolish, loyal and perfidious,
intelligent and stupid, vivid and boring, believing and indifferent, the better
the one than the other' (p. 71).
19. B. Malinowski, Crime and Custom in Savage Society (Kegan Paul, Trench
& Trubner: London, 1925), p. 127.
20. B. Malinowski, Crime and Custom in Savage Society (London, 1925),
p. 57.
21. B. Malinowski, Crime and Custom in Savage Society (London, 1925),
p.57.
22. E.R. Leach, 'The epistemological background to Malinowski's empiricism'
in Mind and Culture, R. Firth (ed.) (Routledge: London, 1957), p. 127.
23. E.R. Leach, 'The epistemological background to Malinowski's empiricism'
 Notes and Sources 301
in Mind and Culture, R. Firth (ed.) (London, 1957), p. 127.
24. G. Bateson, Naven (1936) (Wddwood house: Aldershot, 1980), p. 31.
25. G. Bateson, Naven (Aldershot, 1980), p. 124.
26. E.E. Evans-Pritchard,'The Nuer of the Southern Sudan' in M. Fortes
and E.E. Evans-Pritchard (eds), African Political Systems (Oxford University
Press: Oxford, 1940), p. 296.
27. E.E. Evans-Pritchard, The Nuer (Clarendon Press: Oxford, 1940), p. 90.
28. E.E. Evans-Pritchard, The Nuer (Oxford, 1940), p. 171.
29. E.E. Evans-Pritchard, Nuer Religion (Clarendon Press: Oxford, 1956).
30. G. Lienhardt, 'Self: public, private' in The Category of the Person, M.
Carrithers, S. Collins and S. Lukes (eds) (Cambridge, 1985), pp. 141-55.
31. M. Mauss, 'A category of the human mind' in M. Mauss, Sociology and
Psychology (London, 1979), p. 62.
32. K. Bhattacharyya, 'The status of the individual in Indian metaphysics' in
The Status ofthe Individual in East and West, C.A. Moore (ed.) (University of
Hawaii Press: Hawaii, 1968), p. 47.
33. K. Bhattacharyya, 'The status of the individual', p. 47.
34. K. Bhattacharyya, 'The status of the individual', p. 48.
35. P.T. Raju, 'Indian epistemology and the world and the individual' in
The Status of the Individual in East and West, C.A. Moore (ed.) (Hawaii,
1968), p. 134.
36. Rabindranath Tagore,Nationalism (Macmillan: London, 1917).
37. Radindranath Tagore, Personality (Macmillan: London, 1917).
38. N.K. Bose, My Days with Gandhi (Nishana: Calcutta, 1953); see also his
Seleaions from Gandhi (Navajivan Publishing House: Ahmmedabad, 1957).
39. M. Mauss, 'A category of the human mind' in M. Mauss, Sociology and
Psychology (London, 1979), p; 90.
40. Jawaharlal Nehru, The Discovery ofIndia (1946) (Asia Publishing House:
Bombay, 1961), pp. 247-8.
41. J. Nehru, The Discuvery ofIndia (Delhi, 1961), p. 252. Nehru also pointed
to the significance of the sannyasi or renouncer as an alternative to the
individual-in-society. On this see also N.K. Bose, 'Return of the Individual',
Man in India 34, 3, 1954.
42. A. Macfarlane, Marriage and LmJe in England 13()()-1840 (Basil Blackwell:
Oxford, 1986); also his The Origins ofEnglish Individualism, (Basil Blackwell:
Oxford, 1978).
43. Niharranjan Ray, Nationalism in India, (Aligarh Muslim University:
Aligarh, 1973), pp. 142ff.
44. Jayaprakash Narayan continued his advocacy of a return to village
self-government into the post-independence period. He presented his case
in a widely circulated but unpublished tract, A Plea for the Reconstitution of
Indian Polity (c. 1960).
45. Constituent Assembly Debates: Official Report, vol. 7 (New Delhi, 1950),
p.39.
302 Society and Politics in India

46. M.K. Gandhi, Vamashramadharma, (Navajivan Publishing House: Bom-

bay, 1962).
47. P.V. Kane, History of Dharmasastra 2nd edn. (Bhandarkar Oriental
Research Institute: Poona, 1974), vol. II, part I, p. 7.
48. N.K. Bose, The Structure of Hindu Society (1949) (Orient Longman:
Delhi, 1975), p. 137.
49. M.K. Srinivas, Social Change in Modern India (University of California
Press: Berkeley, 1966).
50. A. Beteille, The Backward Classes and the New Social Order, (Oxford
University Press: Oxford 1981).
51. E. Durkheim, The Division ofLabour in Society (London, 1984).
52. A. Beteille, Individualism and the Persistence of Collective Identities (fhird
Fuller Bequest Lecture) (University of Essex: Colchester, 1984).
53. N.K. Bose, 'Calcutta: A premature metropolis', Scientific American 213,
3, 1965, p. 102.
54. Constituent Assembly Debates: Official Report, vol. 7 (New Delhi, 1950),
701.
55. In the case of the Scheduled Tribes, as against the Scheduled Castes
(and also the Other Backward Classes), the preservation of language, religion
and culture is an additional consideration.
56. M. Galanter, Competing Equalities (Oxford University Press: Oxford,
1984). I have discussed these in my unpublished Commonwealth Lectures,
Positive Discrimination and SocialJustice (University of Cambridge, 1985).
57. A. Beteille, 'Equality as a right and as a policy', LSE Q!mrter/y l, l, 1987,
pp. 75-98 (see Chapter 8).
58. For arguments relating to the United States, see O.M. Fiss, 'Groups and
the equal protection clause' in Equality and Preferential Treatment M. Cohen,
T. Nagel and T. Scanlon (eds) (Princeton University Press: Princeton, 1977),
pp. 84-154. See also R. Dworkin, 'Reverse Discrimination' in his Taking
Rights Serious/y (Duckworth: London, 1977), pp. 223-39.
59. A. Beteille, 'Equality of opportunity and the equal distnbution of
benefits',Artha Vijnana 27, 2, 1985, pp. 96-114.
60. Firstly, the Scheduled Castes as a proportion of the total population
increased between 1951 and 1981 from 14.4 to 15.75 per cent, and the
Scheduled Tribes from 5.4 to 7.76 per cent, mainly as a result of the
inclusion of new groups. Secondly, new lists of the Other Backward Classes
have been made and enlarged in both North and South India.
 Index
Abor tn'be, 73 Annual Reports of the Bureau of
Adi-Dravida untouchables, 85, 167 American Ethnology, 258
Adivasi, 154, 161-162, 164, Anthropology, 3; American, 8,
171-172, 174-176, 178,180, 37, 60; British, 8; as a study of
182, 185, 187-190; problem in culture, 3; European, 9; in India,
India, 52-3 4, 58; of India, 7, 33-4, 41, ·137;
Adoption in Hindu law, 28 political, 12; social, 1-3; and
Adult franchise, 103, 112 sociology, 3; and tribal studies, 4,
Advaita Vedanta, 261-2 47,57,58
Affirmative action, 201, 207, 237; in Anthropologists: American, 124;
the US, 192, 205, 208 British, 138, 140; European, 124;
Affirmativediscrimination,229-230 and the study of civilizations, 58
Afghanistan, 66; Pathans in, 51, Anuloma marriage, 20-1
tribes in, 76 Ao Naga tribe, 78
Africa, 129,257; sub-Saharan, 57, Arab world / Arabs, 58, 60
60; tribes in, 61, 65 Arnold, Matthew, 218,297
Agrippa, 135 Aron,Raymond,134,291
Ahmadiya caste, 88, 109 Aristotle, 135
Ahir caste, 127, 166 Arya Samaj, 165
Ahom tribe, 68, 73 Aryan domination, 55-6, 90
All India Congress Committee, 101 Aryans,55
Amazon, the, 260 Asia, 57, 66
Amazonia, 63 Asia, 57, 66
Ambattan caste, 1OS Assam, 186
Ambedkar, Dr.BR, 110, 172, 187, Australia, 57-58
210,228-230,267,272
Backward classes, 13, 52, 150-91,
American: anthropology, 8, 37,
205,211,219,223,269-70,
60, oflndia, 9; culture, ideas of
272; composition of (India), 52;
natural substance in, 24-5, 27-8,
percentage of Indian population,
30-1; equality, 41; kinship, 18,
52,150,153
23-5, 32; values, 42
Backward Classes Commission,
American Constitution, 192-3
154,156
American Indian tribes, 74; Indian
Backward Oasses Movement, 173
Reorganisation Act of 1934, 74
Bakke case, 208, 209
Andaman Islands; tribes of, 4
Baiga tribe, 64, 78
Anderson,] W, 286
Bailey, F G, 125-27, 131-33,
Anglo-Saxon countries, 255
304 Society and Politics in India
137-38, 147, 169, 181, Bombay, 55, 143
188,291-4 Bombay State, castes in, 48
Ballahatchet, K, 296-7 Book view, 17; and indology, 8,
Bangalore, 226 33-4; as opposed to field view, 6
Bangladesh, 47 Bose, N K, 5, 6, 185-86, 264,268,
Bankimchandra Chatterji, 228, 297 271-2,276,284-7,294,301-2;
Banswara district, 171 and fieldwork, 5; and tribal
Bareh, H, 287 studies, 12, 62, 73-5
Barnes,} A, 144,292 Bouez, S, 287
Bl\rth, Fredrik, 50, 60, 283, Brahmin,26,43,48
285,287 Brahmins, 172, 206; in Bombay
Bateson, Gregory, 258-59, 301 State, 48; in Maharashtra,
Behaviourism, 17 47; and non-Brahmins, 47;
Bemba tribe, 62 subdivisions of, 44-5, 47-8;
Bengal, 47; conversion to Islam in, Tamil,44-5,81,85-106
54 Brahmanical tradition, 6
Bengalis,264 Brass, Paul R, 80, 118, 288, 290
Bernier, 72 Britain, 5, 194, 217, 232
Berremen, GD, 16, 18, 38, 41-2, British, 193, 235; anthropology, 3;
277-8,281-2 civil servants, 71; rule in India,
Beteille, Andre, 126, 219-21, 58, 77,91, 102, 110-11, 128,
224,230,232,235,276-7, 162,167,189,205,224,228,
280-4,288-92,294-7,302;and 267, 269, 272; society, 193;
Dumont, 8, 12, 14, 238-249; and sociology, 3
fieldwork, 4, 12, 169, 175, 181, Brooks, D, 286
184; and social stratification, 13; Buddhism, 172
sociology of, 1O; and Srinivas, Burckhardt,Jacob, 215,219,
13; and structuralists, 5; and 245,297
University of Delhi, 4 Burman, BK Roy, 285-7
Bhagat Movement, 171 Calcutta, 55, 143
Bharati A , 299 Calcutta High court, 22
Bhattacharyya, Prof Kalidas, Calcutta University, 4, 5
262,301 Candie, A, 287
Bhil tribe, 73, 171 Capitalist Society, 133, 136-7
Bhumijtribe, 76,170 Caribbean, blacks in, 51; Indians in,
Bihar, 75, 78,177,186 51; whites in, 51
Bisipara village, 169, 188 caste, 42, 44, 129, 237, 271; Adi-
Blacks; in America, 15, 20-1, vasi, 53, 154, 161-2, 164, 171-2,
24-6,30-1,37-40,42-3,48, 174-6, 178,180,182,185,
52; as castes, 18; Myrdal on, 15, 187-90; associations as pressure
32,37,39 groups, 51, 56, 83, 106-7, 118,
Boas, 57 · 163; atrocities against, 23, 277;
Bodily substance: notion of, in in Bengal, 46; in Bihar, 46; as a
caste, 24; notion of in US, 24
 Index 305
biological model, 45; blood and Chandrasenia Kayastha Prabhu,
natural substance in, 24, 28-30, 48; Chetti, 102, 10S; Chettiyars,
33; Brahmin, 43-5, 47, 81, 102; Chozhadesa Vadama, 4S,
85-106, 113, 116-7, 120,127, 86; Desastha, 128; Dhakar,
141,143,145,169,172, 175-6, 127;Ezhavas, 1S6;Gaunda,
184; and class, 16, 83, 116, 125; 10S, 107; Gujjars, S4;Jats, S4,
as collective identity, 11, 52, 84; 88, 127; Kalla, 81, 86, 88, 102,
competition in, 128; conflict in, 106, 107-9, 114, 167;Kamma,
43, 123, 125, 132; as a corporate 128; Kayastha, 47; Koknasth
group, 126; as cultural model, Brahman, 48; Kollan, 10S;
39-41; as cumulative inequality, Kurmi-Mahato, 78; Lingayat, 106,
38; characteristics of, 39; 1S6, 174, 190; Lushai, 174;
conversion to Christianity, 171; Mahar, 174; Maratha, 48, 106,
Dravidian, 91; dominant, 80-1, 127; Marava, 88, 109; Meena,
88-9, 105-8, 111,113, 120-1, 83, 17l;Mudaliars,91, 102, 10S;
126-8, 130-2, 157,201;and Nadar, 106-8; Naidu, 102; nairs,
endogamy,38-40,44-5,51, 102; Namasudras, S4; Nyogi,
171; and ethnicity, 51, 93; 16, 128; Okkaligas, 86, 106,
and the factory system, 117; 164, 166, 174, 190; Padayachi,
food transactions between, 29, 102, 105, 107, 118; Palla, 83,
30; and gender, 11, 18-20, 112, 169; Panchamas, 87; Pans,
38, 197; Harijans, 43, 85-7, 169, 181; Pasi, 46; Patella,
89-90, 101,106, 109-15, 170; Pramallai Kalla, 86; Raj
161-2, 164, 166-85, 187-90, Gond, 170; Rajput, S4, 127,
206; as harmonic inequality, 171; Raju, 102, 128; Reddi, 102;
42; hereditary membership of, Shri Vaishnava, 44-S, 8S, 96,
39, 198; hierarchy in, 41; and 97,127;Shudras,87;Smartha
Hindu Dharmashastras, 20-1; Brahmins,44,8S,96-7, 11S,
homogeneity within, 38; as a 127; Svapachas, 198; Tachchan,
form of inequality, 18-20, 171, 10S; Tattan, 10S; Thengalai,
194, 197; inferior, 84-5, 201, 97;Thevar, 10S, 107-8, 11S,
203-4; intermarriage, 20-1, 51, 183; Vadagalai, 97; Vadama,
84; and jati, 44; and kinship, 4S, 77, 8S; Vadadesha Vadama,
27, 29, in Bengali culture, 28; 4S, 86; Vannan, 10S; Vanniya,
Kshatriya, 20, 180; low castes, 83, 108; Vattiman, 4S, 8S;
48; loyalties, 116-7, 119, 130; Vellala, 83, 86, 102, 10S~;
in Madras Sfllte, 46; names Velama, 128; and mobility,
of, Adi-Dravida untouchables, 160; and nasal index, 46;
85, 167;Ahmadiya,88, 109; non-Brahmin,4S,47-8,8S-7,
Ahir, 127, 166; Ambattan, 10S; 89-96, 106, 109-17, 127,143,
Antyaja, 198; Astasahashram, 14S, 17S, 184; and physical
4S, 8S; Baidya, 47; Bhumij, difference, 37, 4S-9; and
170; Brihacharanam, 4S, 8S; politics, Sl-2, 79-121, 124,
Chamar, 46; Chandala, 20, 198; 126, 130-1; purity of women
306 Society and Politics in India
in,23,38;quotas,205--6,208-9, 83,194
211-3, 224; and race 11, 15-16, Class system, 38, 129
23,26;29,33,37,38,45--6, Coimbatore, 99, 117
48-9, 51; as religious value, Coimbatore district, 105, 107
41; as segmentary system, 45, Coleman, J S , 238, 297
84--6, 88, 109; sexual unions Collective identity, 12, 213, 221,
between, 20-2; stereotypes, 225,229,230,271; and caste,
31; as structural system, 40, 11; in India, 7, 11
45;subdivisions,42-5,84,88; Collective representation, 3, 7
subordinate, 126; superior, 84-5; Colson,Elizabeth,65, 74,286-7
as a system ofvalues and ideas, Colour: colour caste, 37-8, 42;
8; and tribe, 60, 63, 67-8, 71-3, Commissioner For Scheduled
75, 78; distinction of, 58; in the Castes and Scheduled Tribes,
United Provinces of Agra and 153,169
Oudh, 46; in the US, 15-16, Commonweal Party, 107
38-9, similarities and differences Commonwealth, 57
with India, 37-9, 49; unity, 85-7; Communal politics, 79, 110, 186
upper, 48; in urban areas, 116; Communal problem in India, 53
varna, 44, 157; women in, 23, 38 Communal quotas, 208
Caste Disabilities Removal Act of Communist Party, 95, 111, 113
1850, 198,222,269 Comparative method, 2, 3, 10, 17,
Catholics, 217 33,35,37,215,237,265;and
Cayton, Hand St C Drake, 15, 279 the study of the individual, 253
Chaitanyadeva, 268 Congress Party, 93--6, 98-101, 1()4,
Chandala caste, 20, 198 107-9, 115,120,181, 186-8
Chetti caste, 102, 105 Conservatives, 136, 217
Chattiyars caste, 102 Constitutent Assembly, 210, 228,
Chicago: anthropolgy as ethnoso- 230-1,267,269, 172
ciology, 9 Constitution: Indian, 77, 192-4,
Chicago University, 9 198-9,201,205,208,221-3,
Chiefdom; and tribe, 60-1, 69, 76 228,267, 269-72; ammendments
China,58,67, 76,253-5,260 to, 194,212-3,224,First
Chinese: civilization, 73; in amendment, 194, 205; part III,
Malaysia, 50-1 · Fundamental Rights, 192-4,
Chingleput district, 105 222-3, 273; Part IV, Directive
Chotanagpur tn"bals, 4, 75, 78, Principles of State Policy, 192,
171,185 201, 206, 209, 230; Part XVI;
Christian: missions, 185 Special Provisions Relating
Christians, 47; in India, 53, to Certain Classes, 223, 230;
222; in Kerala, 53; and the Provisions for equality in, 199,
Non-Brahmin movement in 203-5, 224; Provisions for
Tamilnadu, 86, 109 reservation in, 205-6, 272-3
Christianity, 172, 232 Contributions to Indian Sociology,
Class: conflict, 135--6; in India, 8,218,276
 Index 307
Conversions: from Hinduism to Dravida Munetra Kazhagam, 98-9,
Islam in India, 54 107-9, 120
Coorgs, 162 Dravidians, 55, 91
Culture: study of, 3 Dube, S C, 286
Current Anthropology, 238, 248 Duby, G,297
Cushing, Frank Hamilton, 258 Dubla tribe, 76, 78
Cyrenaicans, 60 Dumont, Louis, 7, 9, 12, 14, 17,
72, 122-3, 126, 132-3, 137-8,
Dafla tribe, 73
216,218,234,237-49,276-9,
Dahl, Robert A, 97, 119, 289, 290
281,286,291,295-7,300;and
Dahrendorf, Ralk, 137-8, 282, 291
Europe, 220-1, 239; Homo
Davis, A and B B Gardner and
Hierarchicus, 7, 9; and India
MR Gardner, 15, 26, 279-81
Davis, Kingsley: on caste in the US 7,8,220-l,239;Pramallai
Kallar, 7; concept ofstructure, 8;
and in India, 15-16, 40, 279, 281
and caste, 8, 33, 39; on race in
DeFunis case, 192, 209
Debra Dun; Berremen's work
America, 33, 39
Durkheim Emile, 58, 60, 135-6,
on,16-17
138,231,244,251-2,255-8,.
Delhi, 143, 148
Delhi University, 4, 5 260-l,264-5,270-1,275-6,
284-5,292,297,299,300,302;
Depressed classes, 101, 109-111,
Constitutional recognition of, 110 collective representation, 3, 216;
Depressed Classes Movement, 187 and comparative method, 2;
notion of social morphology, 3, 7
Desastha Brahmins, 128
De Reuck and} Knight, 277, Duverger,M, 230
279-81,283 Dworkin, Ronald, 192, 195-6,
Desai, I P, 295 208-9, 230,237, 294-5,
De Vos, G and H Wagatsuma,
297,302
279,282 East Africa, 51
Dhakar caste, 127 East Europe, 126, 133, 135
Dharma, notion ofin Hindu East European Scholars, 133
thought, 41 Eclecticism, 15
Dharmashastras, 268; and caste 20, Eidheim, H, 50, 283
198; and intercaste unions, 21 Egalitarianism, 42
Dickens, Charles, 30 .Elias, N, 297
Digard, J P, 285 Empiricism, 15, 17
Dinka tribe, 259 Endogamy, 38, 63; in tribes, 63;
Directive Principles ofState Policy, and castes, 51
192, 201,"230 Engek,Friedrich,285
Dollard,John, 15, 18-19, 21, England,30,99,218,232,241,
31, 38, 279-91; study of 244,266
Southemtown, 19, 20, 39 English language; as a status symbol
DravidaKazhagam, 93-6, 104,111, in India, 173
113, 181-2 English law, 252
308 Society and Politics in India
Enthoven, R E, 287 Fieldwork, 5, 33; extensive, 5;
Equality, _192-214; changing intensive,m 5; Srinivas and, 5
character of, 13; as an ideal, Fieldview, 17; and Indology, 6, 7,
13, 215; in India, 237; and 33; and Srinivas, 6
individualism 14, 216-49; of Finch, H A and EA Shils, 35
opportunity, 201-4, 226-9, 235; Firth, R, 258
as a right, 192-214; as a policy, Fiss, 0 M, 205, 294-5, 297, 302
192-214; as a social fact, 13; in Fortes, Mayer, 3, 27,279, 299-300;
the US, 41,205, 237; as a value, and EE Evans-Pritchard, 285;
13, 41,215 and social structure, 2
Estate society, 129
Forward Block, 107, 115
Ethnic groups(s), 49-51, 53; and
France,36,217,244
caste, 51-2; and culture, 50:
Freeman,] M, 22, 279
and endogamy, 51; Harijans as,
Freeman, RB, 32, 279
52; and hypergamy, 51; in India,
French School of Sociology,
55; and didfferences in language,
255,257
50, 53-5; and physical markers, Fried, Morton H, 62-5, 70-2,
50; and plural societies, 50; and
286
politics, 50-3, 119; and quotas, Friedman, Milton, 235, 237,
205, in the US 49-52, 124; 244,297
and race, 51; and differences in Frykenberg, Robert Eric, 127-8,
religion, 50, 53; and tribes, 53; in 291
the US - similarities with Indian _Fuchs, S, 284
castes, 51 Fundamental Rights, 222-3
Ethnic identity, 49, 50, 55, 237; Furer-Haimendorf, C von, 52-3,
and collective identity, 11, 49; 283-4,286
definition of, 49
Furnivall, J S, 283
Ethnosociology, 9; of Indian
culture, 29 Galanter, M, 224-5, 288, 293, 295,
Europe,66-7, 137,231,239,246, 297,302
248,255,266,271; as estate Galton, F, 232, 298
society, 36, 124 Gandhi, Mohandas Karamchand,
European culture 57 110,263-4,267,302
Evans-Pritchard, E E, 2, 3, 4, 6, Gandhism, 111
65, 140-2,256,258-9,285--6, Ganguli, B N, 227, 298
291-2, 299-301; on Nuer, 3, Garo tribe, 63
60, 140-2 Gaya, 69, 163
Evolutionary approach, 58, 76; in Geertz, C, 300
relation to historical approach, Gellner, E, 285--6
58, 76; in study of tribes, 61, 76 Gender; and caste, 18, 38; lack
Ezhavas of Kerala, 156 of comparative studies of, 18;
Fabian,J, 277 inequalities of, 31, 38, 197; and
Factions, 131 race, 11, 18-20, 38
Fascism, 245 General Elections in India, 111,
 Index 309
180; First, 104, 109; 1962 concept of pollution, 156; culture,
Elections, 108 10, 40, hierarchy in, 41; law,
Germany, 217,241, 244-6 28; ideas of natural substance
Ghurye, GS, 16, 44, 46-7, 279, in, 24; Muslim relations, 53;
282,287-8 religious ideology, as different
Glazer, Nathan, 237,298; and DP from Muslims, 54; society, 71,
Moynihan,279,283 and tribe, 68-71; texts, 8, 122-3,
Gluckman, M, 130 126; undivided family, 252;
Godelier, M, 61-2, 285-6 theory of marriage, 20
Goffman, E, 299 Hindu civifu:ation and the State, 69
Goldman, AH, 237, 298 Hindu Marriage Act of 1955, 21
Gond tribe, 68 Hinduism, 75; and animism, 76;
Goody,J, 288 place of forest in, 75; pluralities
Gough, EK, 95, 97, 116, 281, ofculture in, 161; and polygymy,
289,290 21; religious conversions from,
Gough, Kathleen, 140 172; Sanskritic, 171
Government of India Act 1935, 77 Hierarchy: in Hindu values, 222
Government of India Bill, 99 Historical appraoch, 248,253;
government Service, in India, 92 versus the evolutionary approach,
Buba, B S, 293 76; in study of tribes, 58; to
Gujarat, 47, 64 Western civifu:ation, 34
Gujjars caste, 54 Holmes, Justice, 233
Guntur district, 127 Holmstrom, Mark, 226, 298
Gurdon, P R T, 287 Hughes, Everett C, 25, 279; on
Habib, Irfan, 72, 287 American kinship, 25
Hardgrave, R LJr, 284,289 Huizinga,], 298
Harijans, 43, 52, 85-7, 89, 101, Hutton,] H, 167-8, 281, 288, 293
106, 109-15, 161-2, 164, Huxley, T H, 232-3, 298
166-85, 187-90,206;conversion Hypergamy,21,46,5l;among
to Buddhism, 172; conversion to Rajputs, 51
Christianity, 172; in India, 43, Iatmul of New Guinea, 259, 261
85-7, 89-90; integration through Ideal types, 35
politics, 187; special position in Immanuel murder case, 115,183
India, 90; in Tamilnadu, 85-7, lnden R B, and R W Nicholas, 18,
89,90, 101,106, 109-115 28, 279; and Marriott, 18, 24; on
Harrison S S, 283-4, 289 substance and code in Bengali
Hayek, FA, 216, 231, 233, 235-37, culture, 28
240,244,294,297 India, 1, 18-19, 21, 58, 192, 195,
Helm,J, 286 253-5, 260; Backward classes
Hill Rajputs, 51 in, 52; Christians in, 53; as
Hindu Adoptions and Maintenance caste society, 123-4; communal
Act 1955, 278 problem in, 53; equality as a
Hindu: categories of thought, 9; policy in, 192-5, 201, 204-214,
310 Sodety and Politics in India
equality as a right in, 54-55; Juang tribe, 73, 77
Muslims in, 53; multi-religious Judaism, 232
nation, 52; national integration in, Justice (English newspaper), 99
53, 55, 208; physical differences Justice Party, 92-3, 99-104, 110
in, 43, 55; absence of racial
Kalelkar, Kaka, 154, 156
tensions in, 53; Sikhs in, 53;
Kalla caste, 81, 86, 88, 102, 106-9,
social conflicts in, 54; Sociology 114, 167
of, 35; structural view of, 34;
Kamble, ND, 277, 279
traditional, 197, 202
Kamin, L J , 278-9
Indian: Civil Service, 77, 228;
Kamraj, K, 96, 108
intellectuals, 2, 227
Kamma caste, 128
Indian Reorganization Act of
Kane, P V, 21, 268, 279, 302
1934, 74
Kant, Immanuel, 231, 244, 275
Indians in Malaysia, 50, 51
Kapferer, B, 279
Individual, 252-75;
Karma, in Hindu thought, 41
Individualism and equality, 215-49 Harmarkar, R G, 295
Indology, 8; as book view, 8, 33; 34; Karnataka, 274
and Indian sociology, 33 Karve (Mrs) I, 38, 44, 47, 144,
Indonesia, 58 281-2, 292; and KC Malhotra,
Industrial Revolution, 266 48,282
Intellectual climate in India, 1 Kashi, 163
Intellectual tradition in India, 67; Kerala, 90, 211; Christians in, 53
Intelligentsia, 134; in Soviet Kerala High Court, 211
society, 134 Khaldun,lbn,66,67
Iran, 67, tribes in, 76 Khanolkar, V R and L D Sanghvi,
Irish, 50 48,282
Isaacs, Harold R, 174, 176-7, 293 Khasi tribe, 63, 73
Islam reconversions to, 172; place Kinship: American, 18, 23--6;
of desert in, 75 biogenic substance in, 24-27;
Islamic: civilization, 67, 73; State, Dravidian as opposed to North
67; world, 58, 60, 66-7, 76, 202 Indian, 64; and race, 25--6; and
Italy, 219 Schneider, 18, 23-5
Italians, 50 Kollan caste, 105
Iyer,Justice Krishna, 212 Kond tribe, 64
Jamshedpur, 177 Konyak tribe, 73
Japan, 58 Kosambi, DD, 69-72, 286-7
Jarawa tribe, 73 Kothari, R, 283
Jati, 44; difference with varna, 44 Krishna, K B, 102
Jats caste, 54, 88, 127 Kroeber, AL, 57,284
Jharkhand Party, 186 Kumbapettai, 103
Joint family, 266 Kurmi-Mahato caste, 78
Joseph, K andJ Sumption, Kwakiutl, 255
235,298 Lange, Oskar, 134-5, 291
 Index 311
Languages: in Eighth Schedule of Maharashtra, 90, 106, 110, 172-3
the Indian Constitution, 63, 64; Mahars of Maharashtra, 174
in India, S4; lndo-Aryan, S5, 64; Maine, Henry S, 215, 220, 298
Dravidian, 55, 64; and Ethnicity, Maidand,220
SO, S3-S; and race, 55; regional Majumdar, D N, 47, S2, 54, 285;
distribution of in India, S4-S5 and C R Rao, 282-4
Lapps,SO Makah Indians, 6S, 74
Latin America, S7 Malaysia, 54,209; Chinese in,
Leach, ER, 39, 41, 117, 125-7, S0-51; ethnic groups in, 50-Sl;
132-3, 137-8,2S8,277,281, Indians in, SO-Sl; Malays in, 51
290-1,300 Malayalam, 101
Letwin, W, 23S, 298 Malayalis, 102
Levi-Strauss, Claude, 7, 8, 17,249, Malinowski, B, 3, 6, 57, 60, 254-5,
276,280,298,300 257-9,298-300
Levy,MarionJJr, 146,292 Mandeville, Bernard de, 234,
Liberals, 136 237,244
Lienhardt, Godfrey, 259,301 Manheim, Karl, 136, ISO, 291, 293
Lingayats of Mysore, 1S6, 158, 174 Mannargudi taluka, 9S, 113
Linguistic groups: tensions in Manu, the laws of, 280
India,ss Maratha period, 128
Linguistics: and structure, 7 Marava caste, 88, 109
Upset, SM, 120,290 Marcuse, Herbert, 134, 291
Lockwood,David, 138,291 Marecco, Barbara Friere, 258
Lopli tribe, 62 Marriage: anuloma, 20, 21, 39;
London School of Economics, 11 endogamy,39,5l;Hindu,
Lowie,R,3,57 20;hypergamy,21,38,46;
Lucas, J R, 238, 298 monogamy, 21; polygyny, 21;
Luhmann, Niklass, 253, 299 pratiloma, 20, 22
Lukes, S, 216, 29S, 298 Marriott, McKim, 4, 9, 29, 30, 33,
Lushais of Assam, 174 140,277,280,282,292;and
Lynch, D L, 283 Inden, 18, 24, 280
Macfarlane,A,220,298,301 Marsella, AJ, G Devos, and FL K
MacPherson, C B, 129, 220, 291 Hsu,299
Madras,143, 173-4,224 Marx, Karl, 138, 234
Madras Municipal Corporation, 89 Marxian sociologists, 126, 133, 13S
Madras Presidency, 90, 93, 101-2, Marxist theory, 134, 137
110-11 Mason, Philip, 284
Madura district, 106 Mathura, 163
Madurai,69 Mauss, Marcel, 14,251, 2S3-4,
Maghreb,60 256-61,264,299-301
Mahabharata, 74 Mayer, Adrian C, 28, 29, 127, 280,
Mahalanobis, P C, 282 282,291
Maharajapuram village, 114 Mayurbhanj, 77
Mead, G H, 253, 299
312 Sodety and Politics in India
Meena caste, 83, 171 Nagana Gowda Commission,
Megasthenes, 72 156,158
Melanesia, 57 Naicker, EV Rarnasswarny, 104
Metcalf, Thomas R, 124,291 Naidu, KV Reddi, 99, 109
Middle Ages, 220-21 Nair, 102
Middle East, 67 Narnasudra caste, 54
Ministry of Education, 156 Narayan,Jayprakash, 301
Miscegenation, 31, 33 Nasal index, 46
Mobility: horizontal, 142; in India Negroes, 15, 39, 40, 42, 48
202-3, 229; vertical, 142 Nehru,Jawaharlal,265,275
Moffat, M A, 295 Nemchinov, VW, 134,291
Moghul India, 72 Neo-Buddhist Movement, 172
Monogamy: among Protestants, 21 Networks, 140-49
Montgomery, Alabama, 16 Ngwato tribe, 62
Montagu, A, 33, 280 Nicholas, Ralph W, 80,288, 290-1;
Morgan, L H, 58, 60, 62, 284 and lnden, 18, 28
Morris, H S, 283 Non-Brahrnins, 45, 47-8, 85-106,
Morocco, 72 109-15
Morphological approach, 11 Non-Brahrnin Movement in
Mudaliyar, 102 Tarnilnadu, 86, 89-91, 93-4,
Munda,63, 73, 78 99-104, 109, 117
Muslim: culture, 72; identity, 271; North Arcot district, 105
society, 71 North America, 58
Muslim rule in India, 67, 69 Norwegians, 50
Muslims, 47, 89, 101; and the Nozick, R, 233, 298
Non-Brahrnin movement in Nuertribe,3,62,65,259,261
Tarnilnadu, 86, 89, 90, 109; re- Objectivity: in social research, 151
lations with Hindus, 53--4, 56; in Oceania, 257
India, 53; difference in religious Okkaligas, 86, 106, 164, 166,
ideology with the Hindus, 54 174,190
Myrdal, Gunnar, 42,150,277, Onge tribe, 73
280-1, 293; and the American Orans, Martin, 177, 193
Negro, 15, 32, 37, 39; in the Oraon tribe, 63, 73, 171
us, 16 Orissa, 147, 169, 181
Mysore, 86 Ossowski, S, 135, 233, 290-1, 298
Mysore State, 106, 173-4 Other Backward Classes, 151,
Mysore State vs. M R Balaji and 153, 156-160, 163-4, 167,
others, 156 170, 173, 179, 180, 184, 190,
NEFA, 186 224-5, 273; percentage oflndian
Nadar caste, 106-8 population, 157
Nadar sangarn caste association, Oxford University, 4, 5
106-7 Pacific Islands, 57-8
Nadel, S F, 288 Padayachi caste, 102, 105, 107, 118
 Index 313
Pakhtuns nibe, 66 Praja Socialist Party, 108
Pakistan, 73; Pathans in, 51, 73 Pramallai Kalla, 86
Palla caste, 83, 112, 169 Pramallai Kallar; and Dumont, 277
Pan untouchables, 169, 181 Pre-British period in India,
Panchamas, 87, 110 129,189
Panchayat, 81-2, 98, 184 Pre-literate societies, 3
Panchayati Raj, 80, 93, 97, 103, Protection of Civil Rights Act,
105, 121, 182 199,200
Panchayat Union Councils, 107 Protestant Ethnic, 211
Paris, 12,277 Protestantism and monogamy, 21
Parry, Jonathan P, 283 Punjab - Sikhs in, 53
Parsons, Talcott, 32, 280 Race, 58, 237; Aryan, 46; as a
Passin, 49 biological phenomenon, 39;
Patel, Tara, 295 blacks, 15, 20, 21, 24-6, 30-1,
Patella caste, 170 37-40, 42-3, 48, 52; in the
Pathans, 60; in Afghanistan, 51; in Caribbean, 51; and caste, 11,
West Pakistan, 51 151,23,26,29,33,37-40,
Peasant studies, 57 48, 58; as opposed to class,
Person, 250-75 37; conflicts between, 43;
Philosophical systems: Advaita and collective identity 11; as a
Vedanta, 261-2; Mimamsa, 262; cultural fact, 33; and endogamy,
Nyaya, 262; Samkhya, 262 38; and ethnic groups, 51; and
Pinkney, A; 32, 280
gender, 11, 18-20, 38; in India,
Pocock, D F, 8, 137
16, 54-5; as a form of inequality,
Poland, 72 17-20, 25, 31, 40; inter-marriage
Pole, J R, 294 between,21,25-7,38;and
Poles, 50 kinship, 25-6; and language,
Political anthropology, 12 55; ideas of natural substance
Political system, 80 in, 24-5; non-Aryan, 46;
Politics, 79; and caste, 79-121;
occupational structure, 32; quotas
communal, 79; in India, 9-121; in the US, 205; sexual relations
and violence, 113
between, 18-9,21,23,25-6,
Pollock, F, 220, and F W Maitland,
31, 38; and nibe, 52; in the US,
298
15-17,20-1,24-6,30-l,34,
P9lygyny, in Hinduism, 221
37-9; whites, 20-1, 23, 25-7, 31,
Ponavasal village, 114
37-40,42-3,48
Pool, Ithiel, 148
Radcliffe-Brown, 3-8, 57, 140; and
Poona, 70
social structure, 2, 5, 7, 140
Positive discrimination, 204-7,
Raghunandan,268
219,230,237,273; in India as
Raj Gond caste, 170
different from Affirmative action
Raja,MC, 110
in the US, 205
Rajagopalachari, C, 93, 96,
Positivism, 15
100, 104
Power, 80-2: structures of, 80, 82
314 Society and Politics in India
Rajasthan, 64; caste politics in, 83 162; sanskritic model, 162, 169;
Rajput caste, 54, 127, 171 vaishya model, 162
Raju caste, 102, 128 Santal tribe, 63, 73, 177
Raju, PT, 262, 301 Saora tribe, 73
Ramanathapuram district, 183 Sarkar,J, 295
Rameshwaram, 163 Sattur Parliamentary Constituency,
Ramnad district, 105, 107, 111 108
Ramosi tribe, 70 Satyamurti, S, 104
Rao, CR, 47 Scheduled castes, 150, 153,
Rao, C V Subba, 289 155-160, 166-7, 169,170,174,
Ras Phase Pardhi tribe, 70 176-7, 179-82, 188, 190-1,
Rawls,J,235,237,243,298 198-9,203-5,210-11,223-5,
Ray, Niharranjan, 67-8, 286,301 230, 272, 277-8; desire for
Reddi caste, 102 white collar jobs among, 177;
Reddy, PJaganmohan, 298 education levels among, 17+-5;
Redfield, Robert, 4, 5, 7-9, 57, geographical distribution in India,
149, 284; concept ofculture, 5 155; population of in India, 155;
Rege, MP, 295 representation in Civil Services,
Renaissance, 219; Italy, 231,245 175; segregation of, 155; as
Reservation, 205,225, 272-3; of untouchables, 156
seats in Parliament, 182, 223 Scheduled tribes, 77, 150, 153-4,
Risley, Sir Herbert: and caste, 157-8, 160, 169, 174-5, 177,
45,47-8,280,282,287;on 179-82, 188-91,204-5,210-ll,
physical and racial difference, 16, 223-5, 230, 272, 277; desire for
45-6,48 white collar jobs among, 177;
Rivers, 3, 6, 57 education levels among, 175
Roman Catholic Church, 172 Schneider, D M, 28, 280; on
Rossum, RA, 229, 295, 298 American kinship, 18, 23-5,
Round Table Conference, 19, 27, 32; on substance and code,
30-32, 110 24-5,27,29
Royal Anthropological Institute, School as an area of desegregation,
255,258 175
Rudolph L I and S H Rudolph, Segmentary system, 59-61, 64-5
118,283,289,290 Self: concept of, 250-1
Self Respect Movement in
Safavid Iran, 72
Tamilnadu, 104, 172
Sahlins, Marshall D, 59-61, 66,
Sepik, 260
276,284-5
Service, Ellman R, 59, 285
Sanatan Dharma Samaj, 165
Sexual exploitation: and women, 18
Sanghvi, L D and VR Khanolkar,
Shaka, 130
48,282,283
Shils, E A, 288; and H A Finch,
Sanskritization, 160, 162-8, 170-1,
35,280
178-9,182,184;Brahmanical
Shudras,20,22·
model, 162; kshatriya model,
Sikh identity, 271
 Index 315
Sikhs in Punjab, 53 105-6, 115,127, 140-9, 162,
Sikhisni, 172 164-5,239,269,276,282,
Sills, David L, 283 288-91, 293, 299, 302; on field
Simmel, G, 230-4, 240,298; on view oflndia, 6; and Oxford, 5
equality and individualism, 14-31 Srinivasan, Dewan Bahadur, 110
Singer, Milton, 149; and BS Cohn, Sripuram village, 81, 95, 97-9, 103,
282 106, 113-4, 142-3, 147
Singh, K S, 284, 287-8 St Clement of Alexandria, 135
Sinha, Surajit, 277 Stalin, 134, 233
Sivaramayya, Bi228, 294-6, 299 State society, 60, 66
Sivertsen, Dagfin, 95, 97, 289 State and Non-State societies, 59,
Smartha Brahmins, 44, 85, 96-7, 66
115,127 Stratification: on the basis ofcaste,
Social anthropology, 1-3; as 37; dishannonic, 41; on the basis
fOmparative sociology, 3; and ofcolour, 37; harmonic, 41-2
sociology, 3; oflndia, 8, 123 Steinmetz, 257
Social conflict: in India, 54 Structure; in linguistics, 7; as
Social mobility, 142 amorphological conept, 7, 11; in
Social morphology, 11; in the work of Radcliffe-Brown, 7
Durkheim, 3; in India, 11 Structuralists: and Beteille, 5, 7;
Social Network, 143-9 view of India, 34
Social structure, 2, 3, 5, 7, 8 Subbaroyan,P, 100
Socialists, %17; countries, 133, 137; Supreme court, 156, 211;
Society, 122, 134-5 M R Balaji and others vs. the
Sociology, 12; and comparative State of Mysore, 156
method, 34-5; oflndia, 8, 35, Swarajists, 100
123; as a sociology of ideas, 17; Swat Pathans, 73
ofvalues and ideas, 123 Swatantra Party, 96
Sociologists, American, 15 Symbolic order: study of in
South Africa, 23, 37-9, 41-2; anthropology, 10
races in, 42
Tachchan caste, 105
South Arcot district, 105, 107
Tagore, Rabindranath, 263-4,
South Asia, 63
275,301
South Indians, 55
Tallensi tribe, 62
S9uth India, 144; politics in,
Tambiah, SJ, 281
79-121
Tamil, 56, 64, 99, 101; Brahmins,
Soviet: Marxism, 134; scholars,
81, 85-106; as akin to Jews, 94;
133-4; society, 12; state, 66
relations with non-Brahmins,
Soviet Union, 271
85-93; social structure, 110
Spear A, 54, 284
Tamilnadu, 79-121, 167,172,
Spencer, Herbert, 135, 255, 275
182-4: caste politics in, 79-121;
Spiro,42
politics in, 79-121; separatist
Srilanka,54
movement in, 55
Srinivas, MN, 4-7, 13, 86-7,
316 Society and Politics in India
Tamilnadu Toiler's Party, 107 and collective identity, 64, 70-2,
Tanjore district, 4, 77, 88, 91, 77-8; Constitutional guarantee,
• 93-8, 106-8, 111-4, 127, 142, in India, 59, 77; definition of,
148, 169, 175, 181, 184 59, 76-7; in East Africa, 51;
Tanjore village, 4, 12-13, 175, and endogamy, 63, 72, 74;
188,281 and ethnicity, 53; geographical
Tapper, R, 285-7 distribution in India, 155; and
Tattan caste, 105 Hindu society, 68-73, 75-7;
Tawney, R H, 193, 294-5 identity, 69, 185, 187; in
Telugu language, 70, 99, 101-3 India, 52, 63, 67-9, 73, 76-8,
The Lawyer's Collective, 278 150, 153-5,271;andlndian
Thevar Caste, 105, 107-8, anthropology, 4, 71; Islamic, 71;
115,183 and language, 63-4; multi-tribal
Thevar Party, 107 systems, 51; Muslim, 73; names
Thevar, U Muthuramilinga, 115 of, Indian: Abor, 73; Ahom, 68,
Thiruvaiyar Assembly Constituency, 73; Ao Naga, 78; Baiga, 64, 78;
108-9 Bhil, 73, 171; Bhumij, 76, 170;
Thomas, N M, 211 Chandela, 68; Dafla, 73; Dubla,
Thyagasamudram, 97, 103 76, 78; Garo, 63; Gond, 68;
Tinnevelli district, 106 Jarawa, 73;Juang, 73, 77;Khas~
Tirupati, 163 63, 73; Kond, 64; Konyak, 73;
Tocqueville, Alexis de, 215,217, Mahato, 78; Munda, 63, 73, 171;
219-21,240,294,299 Onge, 73; Oraon, 63, 73,171;
Todas tn'be, 4, 78 Patelia, 170; Raj Gond, 170;
Trade Union Politics, 117 Ramosi, 70; Ras Phase Pardhi,
Traditional India, 197 70; Santai, 63, 73, 177; Saora,
Trautman, T, 64, 286 73; Toda, 4, 78; Vaddar, 70;
Tribal, 190; absorption of, 73-4, Vaidu, 70; and non-tribe, 76-7;
154, 184, Hindu method of, northeastern, 4, 73, 155; occu-
73-5; economy, 74-5, 189; pations of, 74; Paleo-mongoloid
identity, 77-8; studies, 4, 47 type, 52; physical differences
Tribes, 57-9; American Indians, between, in India, 52-3; and
74; Andaman Islanders, 4, 73; politics, 53, 184; quotas for, 207;
in anthropology, 57; atrocities and race, 52; as a self contained
against, 22-3; Backward, in India, unit, 57; as secondary phenom-
77; and Bands, 60-1, 76; and enon, 65-70; as segmentary
caste, 60, 63, 67-8, 71-3, 75, system, 59-61, 64-5; and state,
78; and chiefdom, 60-1, 69, 76; 64, 66, 67, 69, 70; in transition,
Chotanagpur, 4, 171; Christian, 59, 154; Veddoid type, 52; and
71, 172, 185; in relation to western civilization, 58
civilization,58,64,66,68, 70, Trichy,91,93-4, 112
72-3, 75-6, 184;classification Tripathi, PK, 233, 299
of: in anthropology, 4, in India, Trived~ H R, 278, 280
4, 59,,62, 71; and class, 61; Trobriand Islanders, 259
 Index 317
Vellala caste, 83, 86, 102, 105-6
Ujjain, 69
Velama caste, 128
United States, 11, 19, 21, 36, 49,
Vidyarthi, LP, 185, 294
50,57,58, 192, 194-5,202,
Village studies, 4, 5
205,207,209,210,219,244,
250, 253, 266, 271: caste in, Warner, W Lloyd, 15-16, 37,
15-16; class in, 15; colour based 280-1: on caste in the US, 15,
stratification in, 15, 23, 31,194, 39; on class in the US, 15
229; equality in, 192-5; Ethnic Weber,Max, 13,35,290:andideal
groups in, 49-50, similarity with types, 35
Indian castes, 51, role in politics, Weiner, Myron, 181, 294, 295
51-2; individual rights in, 194; Welfare State, 194,234
and the Negro problem, 15, 39, Wesolowski, Wlodzimierz, 134,291
40; race in, 15-16, 21, 30-31, West Bengal, 47,274
34, 38-40; Southern, 15, 21, 25, West Europe, 194
31-2, 37--42, 57; Western civilization, 33, 36, 58,254
Untouchability, in India, 188, Western countries, 37, 197;
198-203 anthropology in 3; sociology in, 3
Untouchability (Offences) Act, 199 Western educated elite in India,
Untouchables, 204, 206-7, 229; 188,191
atrocities against, 22-23; mar- Western education, 173, 177-8;
riage with non-untouchables, 200 Western scholars, 133
Upanishads, 264 Western society, 129, 213,
Uttar Pradesh, 47 217,260,265,270: as class
Vadagalai caste, 97 society, 124
Westernization, 162, 165, 173,
Vadama caste, 45, 77, 85
Vadadesha Vadama caste, 45, 86 179,185
Vaddar tribe, 70 Whites, 37, 39: as castes, 15
Vaidu tribe, 70 Whitman, Walt, 264
Vaishya caste, 20 Williams,B,231,299
Van den Berghe, P L, 280, 282 W1Ssler, 57
Vannan caste, 105 Woodburn, J, 295
Vanniya caste, 83, 108 Women: sexual exploitation of, 18
Vanniyakkula Kshatriya Sangam, Women's studies, 19
caste association, 107 Yalman, 126
Varna, 20, 21, 44, 157: difference
Zilla Parishad, 182
with Jati, 44; unions between,
Zulu tribe, 62, 130
20-21
Zuni Indians, 255
Vattima caste, 45, 85