http://www.oit.org/public/english/bureau/inst/download/andre.pdf.

Caste, inequality and affirmative action

by André Béteille
International Institute for Labour Studies Geneva

Preface
This paper is based on a lecture by Professor André Béteille of the University of Delhi,
given at the ILO in Geneva on 21 February 2002. The talk was followed by an
extended session for questions and answers. The Institute would like to thank
Professor Béteille, for making available the transcript for publication.
André Béteille served as Professor of Sociology in the University of Delhi where
he taught since 1959. His research interests include stratification and social class,
equality and social justice and race, caste and ethnicity. In addition to papers in
scholarly journals, he has published Caste, Class and Power (1965); Castes: Old and
New (1969); The Idea of Natural Inequality and other Essays (1983); Essays in
Comparative Sociology (1987); Society and Politics in India (1991); The Backward
Classes in Contemporary India (1992). His most recent publications are: Chronicles of
Our Time (Penguin Books, 2000), Antinomies of Society (Oxford University Press,
2000), Sociology (Oxford University Press, 2002) and Equality and Universality (Oxford
University Press, 2002).

Caste, inequality and affirmative action

André Béteille
I am really very happy to be here with you this afternoon but I also feel a bit
daunted by the magnitude of the task that is assigned to me. I am going to talk about
caste, inequality and affirmative action, all in the course of 45 minutes. These are really
three separate topics, although they are related to each other. I take it that your main
interest is in affirmative action and I will pay special attention to that subject. Some of
you may be familiar with the programme of affirmative action in the United States and
in other countries. There are many distinctive features of the programme in India, and
in order to appreciate these features and understand how the programme has been
faring in the last several decades, it is necessary to know something about the
structure of society and the problems which the programme seeks to address.
Each one of these topics – caste, inequality and affirmative action – is large and
complex. I will begin with caste, and I will try to indicate the basic features of the
system. It is true that caste has given rise to a massive programme of affirmative
action in India, but there are also programmes of affirmative action which have been
devised in other contexts where caste does not exist. So caste has to be understood to
some extent on its own terms.
Certainly, the caste system is changing. I think that is about the only point on
which the specialists are agreed. What is the direction in which it is changing? There
are sharp disagreements on that. Some would say, and this is my opinion, that if you
take a long-term view of caste, the indication is that it is becoming weaker and not
stronger, taking the system as a whole. But there are others who would argue that
state action has in fact increased the strength of caste. So even on that point there are
disagreements among the authorities and I do not want to discuss these
disagreements today – whether caste is becoming stronger or weaker, whether
affirmative action is making caste stronger or making it weaker. This is an argument
which is debated over and over again in India. Some say that affirmative action is
essential, it is necessary in order to eliminate, or at least to reduce, the inequalities that
have been inherited from the past. Others say that affirmative action in fact reinforces
the sense of separate identity between members of different castes.
Now caste, of course, is closely related to inequality but inequality is not all
there is to caste. There are many other aspects of caste which need to be considered
at least briefly. And inequality is not simply a matter of caste. There are other aspects
of inequality in Indian society which are extremely important. In the 1950’s and 1960’s,
after the country became independent, there were sharp disagreements between
Marxists and liberal theorists as to whether, in fact, caste was the real basis of

Page 7
André Béteille
2
inequality in Indian society. Some would say that it was, while others would say that the
real basis of inequality was the ownership, control and use of land. For the latter, it
was the agrarian social structure which provided the real basis of inequality in Indian
society. In other words, there are aspects of caste which can be discussed
independently of the issue of inequality, and there are aspects of inequality which can
be discussed independently of caste.
I will try to bring together caste and inequality and to present the programme of
affirmative action in the context of our understanding of the inequalities of caste that
continue to exist. Many students of Indian society have argued that caste was the
country’s fundamental social institution. If there was anything which gave Indian society
its distinctive features, it was the caste system and caste was the basis of Hindu
society over a very long period. It was very much tied up with Hindu attitudes towards
ritual purity and pollution, to Hindu conceptions of hierarchy as the basic order of
society, and in that sense they would say that caste is the fundamental institution of
Hinduism, that it represents the social morphology, so to say, of Hindu society.
Hinduism is a religion whose social aspect is expressed in the division of the population
into castes and subcastes of many different kinds.
Caste has existed in India, particularly among the Hindus, over a very long
period of time. It has long been a subject of discussion, and not simply among
contemporary or modern social theorists. Caste was extensively discussed in the
classical literature of Hinduism which tried to explain and even to justify the system. So
there has been an elaborate discussion over a long period of time of the logic of caste,
so to say. If you look at medieval European history, you will find elaborate discussions
of the logic of the feudal system and the “estates”. But India has a much greater
continuity over time in the discussion of caste. So it was in that sense the definitive
institution of Hindu society.
Caste was not confined to the Hindus and this is something which must be kept
in mind. Divisions of a kind which are very similar to the divisions of caste among the
Hindus have existed also among the Muslims in India and on the subcontinent over a
very long period of time. When the British administrators began to construct a social
map of India, they used their 10-year censuses as occasions for writing about the
division of the population into its basic groups and communities. These census reports
which start from the end of the nineteenth century provide very interesting information
on the divisions and subdivisions of Indian society. The census of 1921, which, of
course, was of undivided India, had a special report on caste among the Muslim
population. The detailed discussion of what the census commissioner described as
caste among the Muslims generated a strong reaction from the Muslim intelligentsia
who argued that caste was antithetical to the principle of Islam. How could one talk
about caste among Muslims in India? But in the course of time, most students of Indian
society and of the subcontinent have come to accept that caste-like divisions exist not
only among the Muslims but also among Christians and other religious communities.
They also exist in incipient form among the tribal communities in India.
The division of the population into a large number of groups which were ranked
in some kind of a hierarchy was found throughout India and it was not confined to the
Hindus. But of course, it was only among the Hindus that you found not simply the
practice but also the theory of caste. Over a period of 2000 years, the Hindus had
developed a model of society in which the hierarchical division of the population was
regarded as constitutive of their society and indeed of human society as such. It is
Page 8
André Béteille
3
distinctive of caste among the Hindus that you have not only the practice of caste, not
only the division of society into innumerable groups, but you also have a theory which
seeks to describe, explain and justify these divisions.
When one uses the term “caste” in English, one is actually translating two
distinct terms in the classical as well as the modern languages of India. The first term is
varna and the second is jati. Varna and jati have both been described as caste. They
are not unrelated to each other but they are not the same, and it is very important to
understand the distinction between the two in order to understand the social logic of
caste. The scheme of varnas has been elaborated in India among the Hindus for at
least 2000 years. It has no counterpart among the other religious communities, and
that is extremely important. The theory of varna lays down the logic of the system just
as many medieval texts laid down the logic of the system of estates in medieval
Europe. According to these classical texts and also according to later accounts, there
were four and only four varnas – Brahman, Kshatriya, Vaisya, and Sudra. These are
the four varnas and they are ranked in the order shown. The distinctive feature of the
system of varnas is that it was believed to be invariant, to be present throughout the
country, throughout the length of the subcontinent. It was the same order, Brahman,
Kshatriya, Vaisya, and Sudra which was believed to prevail throughout the country.
Further, it was not only invariant, it was also believed to be permanent. The same order
of the social world divided into Brahman, Kshatriya, Vaisya, Sudra remained
unchanged over the centuries.
When we look at the history of modern India, we realize that the traditional
order of varnas may not change, but it becomes obsolete. Fewer areas of life tend to
be governed by varna than was the case in the past. And one can see this
obsolescence of the social order of varnas in the course of the last 50 years.
Discussions of caste in the regional languages in India make less and less use of the
category of varnas. If one talks to modern Indians, if one looks at the newspapers, the
discussion of caste is no longer in the language of varnas as was the case until the
middle of the nineteenth century. Today the discussion of caste is in the language of
jati which is the other aspect of caste.
Now, what are jatis? I said that there are four and only four varnas – Brahman,
Kshatriya, Vaisya, Sudra – but the jatis are innumerable. I myself studied a village in
South India in the early 1960’s. In that one single village, there were as many as 35 or
36 distinct jatis. If you take the region as a whole, you might find the population divided
and subdivided and sub-subdivided into a couple of hundred different jatis. So, they are
numerous and they are ranked, but the order of ranking is not as tidy or as rigid as the
order of ranking of the varnas. Not only that, the jatis are not the same in all parts of the
country. They vary from one part of India to another. Each region has its own distinctive
complement of jatis, and, whereas the varnas are believed to have been unchanging
and immutable, we know that new jatis were being created and old jatis were dying out
all the time. So, it is a much more flexible and dynamic system than the system of
varnas.
Today, when one talks about caste in India, one is really talking about jatis
rather than varnas. What does this mean in sociological terms? It simply means that
when one talked about varnas, one talked about the hierarchy that not only existed in
fact, but also had a certain legitimacy in the normative order of society. Whereas, when
one talks about jatis, one is talking about divisions of a kind which do not enjoy the
legitimacy that the system had in the past. If I may use an analogy from other societies,

Page 9
André Béteille
4
I would say that the system of varnas corresponded to the medieval system of estates.
“The rich man in his castle, the poor man at his gate, God made them high and lowly
and ordered their estate.” That is a Church of England hymn, as a matter of fact. It is a
system whose legitimacy is accepted and taken for granted. In contrast, the jatis
operate rather like the ethnic groups in the United States do. The differences must
also be noted: the varna order had far greater stability and continuity over time than the
system of estates in medieval Europe did, and the system of jatis in contemporary India
is far more elaborate than the system of ethnic groups in the United States.
There is a ranking of jatis but there is a difference between groups which are
ranked in terms of de facto criteria and the social hierarchy whose basic principles are
accepted generally, if not universally, in society. If one looks at the traditional order of
Hindu society and of Indian society in general, I would say that it was a hierarchical
social order in the sense that inequalities not only existed, but that they were accepted
as right, proper and desirable. All this was considered to be part of the natural scheme
of things, not just inequalities between the varnas but also between the jatis, because
the varna system provided a kind of framework, a benchmark for ranking the different
jatis. It was considered right, proper and desirable that they should be ranked in this
way. This applied not just to the ranking of varnas and jatis, but also to the ranking of
men and women. That was the traditional order, it was the most complete and the most
elaborate hierarchical social order known to human history. And the legitimacy of
ranking was taken for granted by and large. It was questioned from time to time, but
over a long historical period the legitimacy of this ranking of jatis and varnas was taken
for granted in the classical, legal and religious literature of Hinduism. There is a whole
range of classical legal texts which lay out the basic principles of the hierarchical order,
but the epitome is to be found in the Manusmriti or the Manavadharmashastra. In the
entire literature, the logic of caste is explained and the Manusmriti gives the blueprint of
a quintessentially hierarchical society.
Today, the Constitution of India provides a blueprint for an egalitarian society.
The Constitution is not based on the premise of hierarchy, but on the premise of
equality. But of course, you do not just erase or cancel out age-old inequalities simply
by adopting new principles in a Constitution. So what else has to be done in order to
reduce the inequalities of the past? This is what affirmative action or positive
discrimination addresses itself to. The object is to reduce the level of inequality in a
society which has had a hierarchical order over a very long period of time. Affirmative
action or positive discrimination was not the only major programme adopted when the
country became independent and created a Constitution based on the premise of
equality. For example, there was a massive programme of agrarian reform which was
also designed to give greater thrust to the pursuit of equality. I would say that these
were the two main programmes – agrarian reform and affirmative action – designed to
improve the conditions of what came to be known as the backward classes in Indian
society.
I have often been asked outside India about the insensitivity of Indians in using
a term like this – the backward classes. It shows a certain kind of insensitivity and I
have often been asked if I feel embarrassed to use this phrase when talking about a
section of my own society. Well, yes and no. The term ‘backward’ has become a
fighting word in India today so it is not necessarily something that one wants to sweep
under the carpet. It has become a fighting word although it was not always a fighting
word and this question of terminology, in fact, bedevilled the administrators of India
even before independence. One major component of the backward classes consists of

Page 10
André Béteille
5
those who are called Dalits today, or the Untouchables or the Harijans as Mahatma
Gandhi called them. J. H. Hutton, a distinguished anthropologist who was also the
commissioner for the 1931 census, wrote that this section of the Indian population had
come to be called the ‘depressed classes’, which was an odious phrase to use in to
reference to a whole section of society. He was very unhappy with the term and
favoured ‘exterior castes’ in its place. Finally, the Government of India in 1935, that is
before independence, settled for ‘scheduled castes’. Thus, the policy of positive
discrimination in India, which I am going to talk about, has antecedents in colonial rule.
It was not created ex nihilo with the Constitution of India. It has antecedents in colonial
rule and right from that time we see the two faces of positive discrimination, a desire to
bring about greater equality, greater social justice as well as a desire to use it as an
instrument for devising politics. It has also been used as an instrument of divisive
politics since independence.
Successive governments in independent India have used positive discrimination
as an instrument for furthering specific political ends. So, what is tantalizing about
positive discrimination in India is that it is not just a question of furthering social justice,
it is also a question of maintaining a certain balance of power between different groups
in society. This did not start with independence and the adoption of a new Constitution.
In fact it started at least 30 years before the new Constitution was adopted. So, what is
this programme of positive discrimination which has such a long history in India? I will
now very briefly indicate the basic features of positive discrimination before I invite
questions. I am often asked: “Is positive discrimination a good thing or is it a bad
thing?” I think it is impossible to give a blanket answer to that question because the
programme is so large and so elaborate and has so many different aspects that almost
inevitably one sees that some parts of it are good, other parts are not so good and it is
very difficult to disentangle what is good in positive discrimination from what is not. It is
not easy to determine how much it has contributed to reducing inequalities and how
much it has contributed to the heightening of rivalries, tensions and hostilities between
groups. If one wants to examine it more closely, I think there are two distinctions that
have to be kept in mind. The first relates to the beneficiaries of positive discrimination –
that is to say, those groups for whose benefit the programme of positive discrimination
has been constructed and elaborated over the last 80 years. So one must get some
idea of the interested beneficiaries of positive discrimination. They are not a single
homogeneous block; in fact, they are very diverse and one must keep this diversity in
mind in order to evaluate the success and the failure of positive discrimination. One
might take the view that positive discrimination is justified in the case of some of these
groups, but not in the case of others. Secondly, one has to distinguish between the
different kinds of programme that come under the broad policy of positive
discrimination; here, one might say that positive discrimination is justified in certain
areas of public life but not in others. At least, one must have some appreciation of the
wide range of the beneficiaries of positive discrimination as well as a sense of the
variety of policies which come under the umbrella of positive discrimination.
Let me first say something about the intended beneficiaries of positive
discrimination, who are broadly known as the backward classes. Who are the
backward classes? They are not a single homogeneous block, and this must always be
kept in mind. And what must also be kept in mind is that the backward classes are not
really classes in the ordinary sociological sense of the term. They are really groups of
communities and not classes. These groups of communities fall into three broad
divisions: firstly, there are the scheduled tribes, secondly, there are the scheduled
castes and thirdly, there are the other backward classes (OBCs).

Page 11
André Béteille
6
Let me very briefly give you an idea of these three divisions among the
backward classes. The smallest group consists of the scheduled tribes. In a very rough
way, they can be compared to the native American population of the United States but
the analogy is deceptive beyond a certain point. They are the tribal population of Indian
society and they are the smallest division of the backward classes. However, although
they only comprise a little over 8.5 per cent of the population, that means over 85
million people. That is not a very small population by the standards of European
countries or by the standards of countries anywhere in the world. The scheduled tribes
are very diverse, and this is directly relevant to the discussion, because the critics of
positive discrimination point out that in India, it benefits not the most disadvantaged
members of the backward classes but the least disadvantaged among them.
The scheduled tribes are divided into more than 400 distinct tribes varying
enormously in size, occupation, economic condition, habitat and many other things.
Their general characteristic so far has been their physical isolation in the hill and forest
areas either on the frontiers of the country or in the interior. But their isolation is no
longer as conspicuous as it was till the middle of the nineteenth century. So today it is
often very difficult to decide whether a community is a tribe or a caste. The British were
very keen to distinguish the tribes from the castes, but they frequently confused them.
If you look at the terminology which they used in the census reports, you will find that
this is a very genuine problem, and perhaps I can address this point when I answer
your questions.
The scheduled tribes constitute the smallest section of the backward classes.
The next consists of what are known as the scheduled castes. Whereas the scheduled
tribes have been isolated in hill and forest areas, the scheduled castes have been
segregated. They have been very much a part of the economy of land and grain on
which Indian society was based. But they were segregated and were not allowed
access to many civic amenities. They were denied access to education and the other
advantages through which one could achieve success in economic and social life. They
comprise over 16 per cent of the population, so around 165 million persons. But they
are also highly divided. First of all, they are regionally divided. India is a country of
many regions, each with its own distinctive language, and the scheduled castes are
divided by language just like the rest of the population, although they come at the
bottom of the hierarchy. This means that there is a distinction between the scheduled
castes of Tamil Nadu who speak Tamil like the non-scheduled caste population of
Tamil Nadu and the scheduled castes of Bengal who speak Bengali. Even if we
confine our attention to the scheduled castes of one particular region, there also we will
find a considerable amount of differentiation as well as ranking. The ranking of castes
did not stop at the boundaries of untouchability, it went right down. So the scheduled
castes or the so-called untouchables were themselves also ranked. It is this that makes
the problem of the scheduled castes quite different from the problem of the blacks in
the United States. It is this that makes the problem of administering a policy of positive
discrimination so much more difficult and so much more complex in the Indian case
because the benefits of the programme tend to be gathered not by the weakest among
the scheduled castes but by the strongest among them. There are analogies to this of
course in the affirmative action programme in the United States. But in India the
problem is far more complex.
Finally, there are what are called the “other backward classes”. The Constitution
of India was drawn up at a time when the country had been partitioned on the basis of
religion so the mood was against further divisiveness. Therefore, in the Constitution of

Page 12
André Béteille
7
1950, special provisions were only made for the scheduled castes and the scheduled
tribes. But it did not stop there because one of the directive principles of state policy
said that some measures may also be adopted for the benefit of other backward
classes without specifying what these measures should be and who the other
backward classes were. So commissions were set up and as a result we have today a
third section of the backward classes consisting once again of numerous castes and
communities, which are even more diverse in their economic and social standing than
the scheduled castes or the scheduled tribes and which are known as the other
backward classes. They comprise anything between 25 to 50 per cent of the
population.
One can give exact figures for the scheduled tribes and the scheduled castes,
but not for the other backward classes. What the British did was to enumerate all
castes and communities throughout the country. At the time of independence it was felt
that the benefit of this was outweighed by the divisive implications of enumerating
caste in the census. So the enumeration of castes was discontinued in the censuses
from 1951 onwards with the exception of the scheduled castes and the scheduled
tribes, who have been enumerated separately. So these are the three broad divisions,
very different in their social standing. Some of the other backward classes are and
have been politically dominant in their areas although their social and ritual standing in
the traditional hierarchy was not very high. Whereas the scheduled caste and
scheduled tribes, one would agree on the whole, have been severely disadvantaged
and deprived, this is not true, or not true to the same extent, of the other backward
classes.
These groups are the intended beneficiaries of positive discrimination. What are
the benefits? There are three main kinds of benefit and I will just enumerate them and
then stop. The first is what may be described as political reservation. This means that a
certain number of seats in Parliament and in the state legislatures are reserved for
members of the scheduled castes and the scheduled tribes, roughly in proportion to
their strength in the population. Political reservation in Parliament and in the state
legislatures is only for the benefit of the scheduled castes and the scheduled tribes, not
for the other backward classes. Political reservations are written into the Constitution of
India and the provisions reveal the ambivalence of the makers of the Constitution as
well as of policy makers in contemporary India. The constitutional provisions for
political reservations for the scheduled castes and scheduled tribes are mandatory.
That is, the Constitution says that ‘seats’ shall be reserved in Parliament for such and
such. However, when the provisions were made mandatory in 1950, it was decided that
this would apply only for ten years, so they would last for a single decade. But since
then the Constitution has had to be amended every ten years to keep on extending
political reservations for the scheduled castes and the scheduled tribes. I think the
general idea is that these should not be written into the Constitution permanently, that
at some future time the need for these special provisions will no longer exist. But that
future seems to be moving further and further away. So that is political reservation.
The second kind of reservation which is even more contentious than the first is
known as job reservation. Job reservations apply mainly to government appointments
at union and state level and also to organizations which are substantially funded by the
government. The provisions for job reservation apply not only to the scheduled castes
and the scheduled tribes but also to the other backward classes. Over the years there
has been an extension of job reservations for the benefit of the other backward
classes. This has now become the most contentious issue, whether the wholesale

Page 13
André Béteille
8
extension of job reservations for the other backward classes accords with the spirit of
the Constitution or not. For job reservations, unlike political reservations, the provisions
are not mandatory; they are enabling provisions. The Constitution says that the state
may take such measures as are necessary for the special benefit of the other
backward classes.
Finally, there are what may be called reservations in education. These, again,
are matters of contention, for reservation exists not only in general arts and science
courses but also in medical and engineering schools. Those who are familiar with the
subject will perhaps know that in the United States two of the most contentious legal
cases have been over reservation; first, in law school in the case of Marco De Funis,
and then in medical school in the case of Alan Bakke. In India, reservation in law
schools somehow has not been a subject of such bitter contention. But reservation in
medical and engineering schools is very controversial. Once again, when assessing
the merits of this, one must ask whether provisions that are justified for the scheduled
castes and scheduled tribes on a limited scale should be extended on such a large
scale. What is the problem with extending these reservations on such a very large
scale? The logic of reservations or of positive discrimination is that special
opportunities should be created for some over and above the general provisions for
equality of opportunity for all. Now, if you create special opportunities for some, then it
does eat into the provision of equal opportunities for all.
The question is one of balance and, once again, this question came up at the
time of the making of the Constitution of India. And the man who piloted it through the
constituent assembly was himself from an untouchable caste: Dr. Ambedkar, a
formidable lawyer. In putting these proposals forward to the constituent assembly,
Ambedkar argued that there are several conflicting aims which have to be reconciled.
First, there is the aim of providing equal opportunity for all, irrespective of caste, creed
and community. That is very important. The second aim is to create special
opportunities for those sections of society which have been severely deprived and
disadvantaged. But he went on to add that these special opportunities should not be so
extensive as to ‘eat up’ the general provision of equality of opportunity for all. Those
were the words he used. And now many people feel that the extension of reservations
to the other backward classes on such a large scale has carried the policy of
reservations so far that in many ways it threatens the more general principle of equality
of opportunity irrespective of caste, creed and community which is also inscribed into
the Constitution of India. So, very broadly speaking, this is the framework within which
the problem of positive discrimination or affirmative action in India has to be assessed.
Page 14
9

Debate
Question - I would like to hear a little bit more about the middle part of your topic,
which is inequality.
André Béteille
- Inequality is a very important topic in itself and again, there are two
aspects. One relates to inequalities in the agrarian social structure, that is, the problem
of inequalities in the ownership, control and use of land. These inequalities continue
but not in the extreme form in which they existed at the time of independence. At that
time a number of measures were adopted and some of them were in fact successful.
For instance, the abolition of the zamindari or the system of estates was generally
successful although there were numerous evasions of the law relating to the
resumption of estates by the Government. Today, you no longer find landlords owning
five, ten or 15 villages as was fairly common until the 1950’s. Then the regulation of
tenancy. I think that has also had some positive effects, although it had negative
effects as well. Its unintended consequences should not be lost to sight. For instance,
you have a relationship between landowner and tenant, and the Government comes in
and changes the terms of the tenancy in favour of the tenant. What does the landlord
do? The landlord evicts the tenant and transforms him into a landless labourer, tells
him that if he wants to work on the land he has to renounce his claim as a tenant and
work as a wage labourer. That has happened, but on the whole I think that over the last
50 years, the conditions of tenancy have become less unequal and less oppressive
than they were in the past. The biggest problem, which was not fully anticipated, was
the increase of population. I remember a conversation with K. N. Raj, one of our
leading economists, whose heart was very much in the land reforms and who was also
instrumental in drafting parts of the first five-year plan. I asked him then what he
thought went wrong. He said: “You know if I were to answer that question in one single
sentence I would say that we did not anticipate this tremendous increase in
population.” Many of the gains of agrarian reform were swallowed up by the massive
increase in population, so that the conditions of landless agricultural labourers continue
to be very weak and very precarious. Some changes have come about but those
changes, I would say, have led to a lowering of the top rather than a raising of the
bottom. There are also provisions for the regulation of wages which have had some
effect on the conditions of agricultural labourers.
There is another aspect of inequality which relates to affirmative action, and
that is the creation of a new middle class in India. I think that is extremely important. It
is not as if people rank each other only in terms of caste. Occupation, education and
income have become increasingly important in Indian society. Now it is true that the
different castes are not equally represented through the entire hierarchy of
occupations. The upper castes are more common in the superior non-manual
occupations. There is an over-representation of the upper castes in such occupations.
And there is an overwhelming over-representation of the lower castes in the inferior,
menial and manual occupations. But some shaking up has certainly been taking place
and I would say that affirmative action has played a part in this. However, it is

Page 15
Debate
10
extremely difficult to say that this would not have happened at all without affirmative
action. This is very difficult to judge. Some of it I think would have happened even
without affirmative action, but I feel that affirmative action has made a positive
contribution. It has helped create a middle class among the untouchables and the
tribals, and certainly among the other backward classes. So these are the two aspects
of inequality. The increasing differentiation and ranking of people not in terms of caste
but in terms of income, occupation and education of the kind that you find in societies
which have had a large middle class over a much longer period of time. Although a
middle class began to emerge in the presidency capitals of India like Calcutta, Bombay
and Madras by the end of the nineteenth century, it has grown enormously in size in
recent decades, and with this growth, a new kind of inequality has emerged.
Q-
In India the utilization rate of job reservation by Dalits is very low, hardly 5
to 6 per cent of the total. I would like to know whether you have definite figures on that.
My second question is in connection with your answer just now. Would you say that the
rise of the Bharatiya Janata Party (BJP) in India today is connected to the size of the
middle classes and their fear of being threatened by these positive measures for the
other backward classes?
AB -
Your first question is an extremely important one, but it is very difficult to
give a clear-cut answer because it is true that at the highest levels the Dalits are under-
represented, but this is changing very rapidly. You see, one of the problems is counting
the number of Dalits in the educational sector. Do you include for instance, faculty
appointments in universities? If you do that, then you will find that they are very poorly
represented. There has been a long controversy over universities, which are
autonomous institutions, and whether they are obliged to have quotas in faculty
positions. Some universities have had them, others have not had them, and now they
are under pressure to have reservations in faculty positions, and one doesn’t know
what the outcome will be. But if you take one particular sector of the occupational
system, and a very crucial one, that is the Indian Administrative Service (IAS), which is
the élite corps, these provisions have been there since the beginning of the
Constitution and if you go back to the 1950’s you will find one or two Dalits being
recruited to the IAS. As far as I know in the last ten years quotas have been filled for
recruitment to the top level of the civil service, which of course does not mean that
throughout the entire range of the IAS you will find Dalit officers in proportion to their
strength in the population. You will find them more proportionately represented in the
younger age cohort than in the senior age cohort, but there are already secretaries to
the Government of India who are Dalits.
Your second question. Well, the middle class always looks after itself in every
society. I don’t think that opposition to reservation for the other backward classes was
associated specifically or singularly with the BJP rather than with the Congress party. I
don’t think so. I think these are other and deeper causes for the rise of the BJP. I am
the kind of sociologist who is uncomfortable with the view that the rise and fall of
political parties necessarily reflects basic trends in society. It may do so, over a long
period of time, but fluctuations in the electoral fortunes of political parties are linked to
transformations in the basic structure of society through mediations which are very
complex and little understood. So I am a bit puzzled by the newspaper view that the
middle classes are expanding and the BJP is looking after their interests. It is
interesting that side by side with the rise of the BJP, the middle class among the Dalits
has become more articulate and much more vocal and Dalit leaders are now saying
that it is essential to create a Dalit middle class.

Page 16
Debate
11
Q-
You referred to affirmative action in relation to the backward classes: the
scheduled castes, the scheduled tribes and the other backward classes. You also
referred to hierarchies in sex among those classes and to the experience of affirmative
action in the United States which has been most strongly related to measures
introduced on the basis of sex. Can you please tell us about Indian experience in
relation to affirmative action based on sex?
AB -
In the last four or five years the Indian Parliament has had before it a bill on
the reservation for women of 33 per cent of the total number of Parliamentary seats. Of
the many commissions that have been set up in independent India and that have
produced reports, I would single out the commission that was set up, I think sometime
in the late 1960’s, on the status of women in Indian society and which produced a most
judicious and thoughtful report. The report argued against quotas for women, either in
employment or in Parliament. But that has now been reversed, with consequences
which one must try to understand. There has been opposition to the reservation of
seats for women in Parliament and, without taking sides on the merits or demerits of
this, I think that one must examine the arguments that have been put forward. The
arguments against reserving Parliamentary seats for women have been put forward
mainly by leaders of the other backward classes. Their argument has been that over
the years they have struggled to establish a place for themselves in politics. Through
their efforts at mobilization and organization, they have captured a large number of
seats in Parliament. Now one-third of those are going to be taken away and reserved
for women. Who are the women who will get these seats? Their argument is that these
seats will be cornered by women belonging to the upper castes and the middle classes;
they agree to reservation for women, but they want quotas within the quotas and there
is a problem there. The problem is that once you have quotas within quotas in
Parliament, you will have to have quotas, not only for backward caste women, but also
for minority women. If you have quotas for minority women, then you cannot deny
quotas to Muslim women, and that will reintroduce into the Constitution the very thing
against which the nationalist movement and the Constitution of India stood, that is,
reservations on the basis of religion.
Thus, there are various problems with reservations for women. However, if you
look – and I have often been struck by this – at the position of women in Indian society
over the last 100 years, you will find, at least on the surface, that there have been great
advances. For instance, the ICS, the Indian Civil Service under the British, did not
recruit women. It was closed to women. When the country became independent, the
IAS was thrown open to women. At first there were very few. They were recruited in
ones and twos, then the figure gradually increased to ten, 12, even 15 per cent of the
total intake, and it has stuck there. My assessment is that it will not rise much higher
than that, so you have the civil service, the higher civil service, you have the
professions, law, medicine, entertainment, the media.
I have two daughters. One is in a television company and other is in ICICI which
is one of the major financial institutions in the country. Very recently, the widely-read
magazine Business Today had a special issue on successful companies, and it singled
out ICICI, saying that it has been very successful because it is an easy place for
women to work in. Nowadays you see women, all across the board, in the various
professions and in the universities. I would say that there has been a sea change in the
position of women, but this is confined largely to the higher levels of society. You will
not see it to nearly the same extent as you go down the hierarchy. It has been quite
easy to fill positions in the best academic institutions with women and they are very

Page 17
Debate
12
well-represented there. It has been much more difficult to fill those places with people
from the scheduled castes, the scheduled tribes and even the other backward classes,
and I think the reason is once again that the middle class looks after itself. Middle class
parents in India have now decided that it is important to give not only their sons but
also their daughters a very good quality education. There may not be absolute equality
between the son and daughter, but I think that the change in the position of women in
the middle and upper middle classes in Indian society is very striking. This is the
reason why the leaders of the backward classes are opposed to quotas for women in
Parliament. They feel that quotas for women in Parliament will be followed by
demands for quotas for women in jobs and once again the benefits will go to middle-
class women who are mainly from the upper castes.
Affirmative action in India does not mean the same thing as in the United
States. To be very frank, affirmative action in India simply means numerical quotas.
That is what it has come down to - numerical quotas. Quotas for women have been
resisted, but other kinds of legislation, other kinds of benefits for women’s education
have been put forward, though not with complete success. These have benefited
mainly the middle classes and the upper middle classes and now there are two areas
which have quotas for women. One has already been adopted and that is in the village
councils, or Panchayats, which have seats for women. I think it is still too early to say
how this is working. Enthusiasts say that it is working wonderfully and critics say the
women are only there to put a stamp on decisions taken by their husbands. It is a very
large country and a very complex situation. The question of quotas for women in
Parliament is a complex and tangled issue. I am opposed to such quotas because I
think the debate reflects a basic lack of sincerity in the political parties. No party will
oppose quotas for women in Parliament. At most they will say, let us have quotas
within quotas; at the same time, no political party will go out of its way to field women
candidates in the elections. You do not need quotas in Parliament if there is goodwill. If
the political parties are sincere about wanting more women in Parliament, then they
should do what Mr. Tony Blair did in 1997. I must repeat that in India, and the sub-
continent as a whole, family and kinship are extremely important in politics. This is true
of Pakistan, it is true of Sri Lanka, and it is true of Bangladesh. Look at the records of
our political leaders. Family and kinship are extremely important. The women who
benefit are very likely to be women whose husbands, fathers or brothers already have
important positions in Parliament, and that tends to dampen my enthusiasm for quotas
for women.
Q-
I would like to consider the future impact of caste discrimination or
affirmative action in the light of two emerging trends. One is the process of
globalization and privatization taking place in India and the second thing is that I
believe that the impact of caste remains as strong as ever before. It is just that it has
changed its ways of expressing itself. If you look at Indian politics today, there is no
ideology but they are all led by caste identities and caste agendas. The caste mindset
continues. Education, mobility and exposure do not seem to have made an impact on
the basic caste mindset of the Indian community even among the middle class and
those who live abroad. So I am very sceptical about how Indian social groups,
especially those who have been marginalized for several centuries, will find themselves
in a situation where there is no mechanism of positive discrimination.
AB -
There are two major questions wrapped up in your observations. First, I
don’t agree with you that caste remains as strong as it was in the past. I don’t think so.
Let me say why I don’t think so. There are very penetrating, detailed and exhaustive

Page 18
Debate
13
discussions of caste from the end of the nineteenth century onwards in books which
were published mainly by British civil servants, some of whom were very keen
observers of Indian society. If you examine what they said about caste, you will find
that for them the decisive feature of caste centered around the observance of rules of
purity and pollution the rituals of purity and pollution, taking food, taking water,
maintaining distance and so on. There can be no doubt whatsoever that there is a
decline in the strength of the ritual rules of purity and pollution. I have no doubt in my
mind about that. I would challenge you to show me any single detailed, ethnographic
account which shows that these ritual rules have become stronger rather than weaker.
But I agree that the politicization of caste is extremely important. I am not discounting
that. But a kind of optical illusion is created by media interest in the politicization of
caste. The media, whether television or newspapers, overplay politics not just in India,
but everywhere. Newspapers are not interested in discussing the way in which ritual
attitudes towards purity and pollution are breaking down. Newspapers and television
channels are much more interested in recording how loyalties of caste are mobilized in
the political process. I do not deny that this is happening. Certainly it has been
happening over the last 50 years, but again with ups and downs, and with enormous
regional variations which should not be ignored.
I see these changes as part of a long-term trend. The regulation of marriage
according to rules of caste has not disappeared but it has certainly weakened
substantially, and I could discuss this with you in some detail. The noted
anthropologist, Louis Dumont argued that the really decisive rule for the regulation of
marriage in caste society was not endogamy, but hypergamy, that is the rule according
to which a man of superior caste takes wives from inferior castes. Bride givers are
inferior, bride takers are superior. In a sense that brings out the hierarchical basis of
caste. Today the rules of hypergamy have not only broken down, but young men and
women whose ancestors used to practice hypergamy do not even know the terms that
were used for these kinds of marriage. I am not saying that inter-caste marriages have
become very common. They have not become very common. But they have become
less uncommon than in the past. Not only that, among the middle classes, although
most marriages are within the caste, it is much easier to violate the rules which
required a man to marry not only within his own caste but also within the subcaste of
his own caste and the sub-subcaste of his own sub-caste. So, in a long term
perspective, I think these changes are quite important.
The association between caste and occupation has not disappeared. It certainly
exists as I myself pointed out but it has become weaker than it was in the past. These
changes are not to be discounted. I agree that caste consciousness has increased
enormously and affirmative action has played a part in that by increasing and
intensifying caste consciousness. But, affirmative action alone cannot be responsible.
Political leaders have found caste a very easy basis for mobilizing political support.
That is the easiest and the laziest course of action for a politician to take. Many years
ago when I was studying this problem in the 1960’s, I gave a talk to a group of
American engineers. They had come to Delhi for a year’s stay and the American
embassy had organized a series of lectures on Indian society, a crash course of one
week. My lecture was on caste and politics, and at the end of it, one man who knew
nothing of the social sciences came to me and said: “What’s so new about all this? In
New York where I come from, we say that the politics of the city is governed by the
three I’s, the Italians, the Irish and the Jews.” Now this analogy can be misleading but
politics certainly played a very important part in intensifying caste identities in India as
well as ethnic identities in America. But again, it has had its ups and downs and one
Page 19
Debate
14
must not ignore regional variations. Compare West Bengal with Bihar, for instance. A
well-known historian used to say to me: “In Bihar only the mosquito is free from
sentiments of caste: it bites everyone irrespective of caste. Everything else is governed
by caste.” Now if you look at West Bengal, the language of politics is very different
there. So without discounting the strong presence of caste in the political system, one
must take other factors into account as well.
I go back again and again to what Dr. Ambedkar said. In Madras presidency in
the 1920’s there were reservations or quotas not only for the backwards but also for the
forwards, so all appointments were made on the basis of quotas and they had a roster
system. There was a comprehensive system of quotas, but it was manageable
because the number of positions was relatively small, and it was also feasible because
the British were not impeded by a Constitution which under Article 16 guarantees
equality of opportunity and under Article 15 prohibits discrimination on the basis of
caste, creed and so on. We are hampered by such a Constitution. I do not think it
would be right to jettison the provision of equality of opportunity for all, irrespective of
caste. So the question is how to balance the two. In the early years of independence, in
the 1950’s and 1960’s, the Government took over the economy, became the main
employer, and acted as a pace setter all across the board. People have now begun to
feel that the Government has actually hamstrung the economy and the market should
be given more room.
Many now feel that government action is not enough for social transformation
and they have begun to turn to non-government organizations (NGOs). I have mixed
views on this. I think that the role of the Government in finding employment for all
cannot be sustained. I am not saying therefore that there should be no planning, but
that people have come to realize that expecting the Government to solve every kind of
problem that arises in society or in the economy has now become counter-productive. I
have a friend who is a great enthusiast for the private sector, and he tries to convince
me that the market solves problems by expanding opportunities and increasing
mobility. But as a sociologist I believe that when the system opens up, there will be
both upward mobility and downward mobility and, even if the net balance of mobility is
up rather than down, there will be political discontent. In a democracy, the political
costs of downward mobility are far higher than the political benefits of upward mobility,
even when upward mobility is greater than downward. After all, if I move up, I attribute
my success to myself. I don’t say that the Government has done wonderful things for
me, but when people move downwards they blame the Government. One has to be
very careful about this, and I would say that opening up to globalization will have its
costs although one should not count only the costs.
Q-
I have a question on reservation in education. In a country where the
development of education is very uneven, you have written that university education
has been developed at the expense of secondary education and secondary at the
expense of primary education and it is also very uneven state-wise. So I would like to
know whether you can expand on the outcome of development in basic education and
move towards reservation at the university level.
AB -
I entirely agree with you. I have myself used that very expression, that in
India, secondary education has been developed at the expense of primary education,
and college education at the expense of secondary education. But the situation is no
longer as rosy for the universities. Things are changing somewhat, but not all that
much. I think there is enormous scope for investing in education, primary education,

Page 20
Debate
15
and I make a distinction between what I call universality and equality. I put more
emphasis on universality, that is making certain basic skills, aptitudes and capabilities
available to all universally, irrespective of merit and irrespective of need. They should
be made universally available to all. If you wish to pursue this further, I am just in the
process of publishing a book called Equality and Universality in which I discuss these
very things.
It is important to understand how and why this lopsided development took
place. It is much easier and less costly to build 25 moderately good universities than to
build so many hundred thousand moderately bad schools. It is very expensive to have
no more than barely adequate schools in every village. It requires enormous resources
and the policy makers did not have the stamina to pursue this. The middle class did not
depend on the state for primary education. The most ardent promoters of equality and
social justice in India would never dream of sending their own sons and daughters to
government schools. They know how to look after their children and they are prepared
to pay for schooling. What they seek to do is to ensure that college education is
virtually free, because their sons and daughters, having had excellent school
education, will walk into the best colleges and the best legal and medical schools. This
is changing, but it takes time, effort and political will to change it. And in a democracy it
is extremely difficult to adopt policies which seriously hurt the interests of the middle
class. You can do it in a totalitarian system. Chairman Mao could do something that Mr.
Nehru could never possibly have got away with. I am a great admirer of Mr. Nehru, not
an admirer of Chairman Mao, but I think one must understand that there are certain
limits to what the political leadership can do in a democracy. If it does not carry the
middle class with it, its political fortunes decline. I do not think that any section of the
political leadership has the energy to alienate the middle class in India. Yes, it will also
carry the backward castes along with it, but all that means is that it will carry the middle
classes among the backward castes with it. No political leadership, no political party
can afford to alienate or antagonize the middle classes. And the egalitarians, the
professed promoters of equality and social justice, some of whom are my best friends,
they all look after the interests of their progeny and their relatives extremely well, and
after all why not?
Q-
I had the pleasure recently to work for some weeks in India and one of the
hottest issues in the media besides cricket and Pakistan was the revision of the history
curriculum in school. Now, apparently there are a few strong leaders in society who
would like to introduce some kind of Hindu history curriculum instead of the existing
secular one. How do you assess the situation? Is this a move towards turning the clock
back? How large do you think is the risk that these forces may succeed, may gain
strength?
AB -
I am very worried by this trend, let me say this straight away. I am very
worried by it, and I have written in the media against it, as many sensible people have
done, but the actual story is rather more complex than the professed champions of
secularism make it out to be. As far as the history curriculum is concerned, the
professed champions of secularism pursued somewhat ruthlessly a policy of officially
sponsored Marxism in the writing of history text books and they alienated the Hindu
fundamentalists and many others as well. I am not saying that this justifies doing what
the proponents of Hindutva want to do now. Two wrongs do not make a right, and this
wrong I think is far more severe than the earlier one. At least the Marxist historians
included many people of intelligence and ability. The historians who are trying to
promote Hindutva do not have the same intelligence or ability. But India is a very large

Page 21
Debate
16
country, and the school curriculum is a state subject, not a union subject. West Bengal
will never go in for this kind of thing. This trend is so absurd and fatuous that I do not
see how it can go much further. I think that the very people who want to turn the clock
back – it’s not just history, they will attack sociology and political science very soon –
are interested in a different kind of education for their own sons and daughters. That is
a very important point that should be kept in mind. This demand will cause a lot of
confusion, a lot of waste, and therefore I am alarmed. When BJP came to power for the
first time, I wrote an article in a newspaper in which I said that all this noise about the
BJP being radically different from the other parties did not carry much conviction. My
feeling was that its foreign policy would basically be the same as the foreign policy of
the Congress party, and its economic policy too would not be very different from the
economic policy of its predecessors. I said that what it would attack was the
educational system, and particularly school education. I do worry very much about that.
I think it is very worrying.
Q-
I have two questions. Firstly on the issue of extending reservations,
particularly in the employment sector. I’m aware that a number of Dalit groups have
been lobbying for some time for the extension of reservations to the private sector also
and I believe that at least one state Government is presently considering the issue
seriously. I believe it is looking at the feasibility of employment reservations in the
private sector, and I would like to hear your views on the feasibility and indeed the
desirability of such an extension. My second question relates to the very lively national
debate that took place at the time of the world conference against racism and racial
discrimination and xenophobic related violence. This was quite a complex discussion
that continued over a long period of time before and during the conference, and I
believe that you spoke against including caste in the agenda of the world conference. I
wonder if you would be prepared to elaborate on your reasoning.
AB -
On the first question, the extension of reservation to the private sector, I
don’t really know how far this will work out. We have very strong constitutional
provisions for protecting property rights, the rights of the individual, and so on, and I
don’t know how far this can be carried through. I think that it will meet very strong
opposition in the private sector. Over the years, I have come to the conclusion that
there must be an active policy of affirmative action as against straightforward numerical
quotas applied across the board without discrimination, for every kind of institution and
every kind of position, and I hope that the attack on quotas and the extension of quotas
will not throw the baby of affirmative action, even aggressive affirmative action, out with
the bathwater. I hope so, but I don’t really know. There is now growing hostility to
affirmative action as such. I have never been opposed to affirmative action. Let me tell
you how I have applied it myself, in a very small way, as a university professor, as a
member of academic selection committees. In selecting students, my policy has been,
other things being equal, to select a Dalit in preference to a non-Dalit. Other things
being a little less than equal, still select the Dalit. But one must ensure that the person
selected can deliver the goods. Official practice has been to equate affirmative action
with affirmative reservation in numerical quotas and to fill the quotas without
considering whether other things are equal, a little less than equal, or completely
unequal. That has been an unhealthy practice, and it has damaged the prospect of
affirmative action of the right kind.
On the question of race and caste, I strongly opposed the inclusion of
untouchability on the agenda of racial discrimination for the simple reason that I do not
believe that untouchability is a form of racial discrimination. It is a form of

Page 22
Debate
17
discrimination, and it may be worse than racial discrimination, but one must not put it
into an agenda on racial discrimination. That would be to act in bad faith and under
false pretences. I think that those who were doing it knew perfectly well that caste is
not a form of race but they felt that in a good cause it does not matter if you stretch a
point a little. I have no objection to the atrocious practice of untouchability being
discussed here in Geneva or in Buenos Aires or anywhere. I have myself discussed the
practice of untouchability and the violation of the rights of untouchables in papers
published both within and outside India, but I think it is wrong to include it in a
conference on racial discrimination. I sincerely believe that caste is not a form of race
and the overwhelming weight of professional opinion on the subject would agree. Race
is a different thing from caste and therefore, if you want to discuss caste or
discrimination on the basis of caste, or the practice of untouchability, do so by all
means and don’t confine the discussion to Delhi or Bombay or Bangalore. Do it
anyway, wherever you want to. But don’t do it under false pretences. It has various
other kinds of political implication. Once you say that because these groups are more
or less endogamous therefore they are races, there is nothing to prevent religious
minorities from claiming that they are victims of racial discrimination. There is nothing
to prevent the Catholics in Northern Ireland from claiming that they too are victims of
racial discrimination. What is the difference? There is nothing to prevent the French
Canadians in Canada from claiming that they are victims of racial discrimination. It
cannot stop at caste. It has to cover the entire spectrum and I think it makes nonsense
of our conception of race. I’ll put before you an observation which I have quoted from a
very distinguished British biometrician and geneticist, J.B.S. Haldane. Haldane wrote a
short piece in which he said that the term “race” has so many different meanings as to
be virtually useless in scientific discussion, though it is very useful in getting members
of the same nation to hate one another. Using any argument to keep the temperature
up is something to which I cannot consent.
Q-
I have seen a lot of frustrations among the upper castes about this question
of affirmative actions or quotas. Do you feel that it might become a reason for more
social divisions in India?
AB -
You know, it can. There is a certain amount of frustration although I don’t
have all that much sympathy for the upper castes and their complaints. My main worry
is that there are so many different groups. The logic of reservation or of numerical
quotas undercuts the very basis of the functioning of a whole range of modern
institutions. I find it very difficult to think of a university department functioning
successfully if the general impression is accepted, that Dalit students will not be taken
care of unless there is a Dalit professor. Only a Dalit professor will take care of Dalit
students properly. Tribal students will not be taken care of properly unless there is a
tribal professor, and so on. In some states in India, where quotas have been in
operation for some time, faculty positions are earmarked not only for specific subjects
but also for specific castes or groups of castes. This creates an atmosphere of distrust
which makes it extremely difficult for universities, hospitals and other modern
institutions to function properly. Such an institution can carry quotas up to a certain
point, but when they invade the entire structure of the institution, it becomes extremely
difficult for it to function effectively. So quite apart from the grievances of the upper
caste students, there are other reasons why comprehensive quotas are likely to create
problems.