5

Enter Mandal and Sawhney

The law of caste-based reservations started evolving with bewildering complexity in

subsequent decades. A cluster of questions occupied the docket in the cases that came
before the Supreme Court. For instance, the court was often asked the age-old question:
how will the ‘Other Backward Classes’ really be identified? Can entire castes be
considered backward? What about well-off families in castes that are backward—
would they get to benefit from reservations as well? How far could quotas go—would
Ambedkar’s vision that reservations were only for a minority of seats survive vote-bank
politics? In the colonial period, though the word ‘class’ was used in terms like
‘depressed classes’ and ‘backward classes’, ‘class’ was really a euphemism for caste.
However, in independent India, would reservations be permissible on the basis of
caste? This chapter examines the intense debates that took place among the judges of the
Supreme Court—judges like Justice Gajendragadkar, Justice Subba Rao, and Justice
Chinnappa Reddy—as the court attempted to answer these questions.

The 50 Per Cent Rule Is Born

In January 1960, the State of Mysore in south India set up a committee headed by a
member of the legislative assembly, Dr R. Nagan Gowda, to specify the criteria to
decide who the ‘socially and educationally backward classes’ were.1 The committee
opined that social backwardness would be defined on the basis of caste or community.2
For instance, if a caste was considered to have a generally low status in society, it was
identified as backward.3 For determining ‘educational backwardness’, the committee
figured out that in 1959–60, there were 6.9 students for every 1,000 people in Mysore
in the last three classes in high schools in the state. If a caste or community had a lower
average than this, then it was considered educationally backward.4
For instance, the Voddar caste in Mysore had only 1 such student for every 1,000
people in the state—way below the state average. They were obviously educationally
backward. By contrast, Brahmins had 38.8 such students for every 1,000 people in
Mysore—much higher than the state average.5 For this purpose, it did not matter how
much lower than the state average a community’s literacy levels were. For example,
though Muslims had 5 students for every 1,000 people in the state, not substantially
lower than the state average, they were still considered to be backward. The Nagan
Gowda committee also grouped Other Backward Classes into two categories—
backward and more backward. Those who had literacy levels that were less than 50 per
cent of the state average were considered ‘more backward’.6
The Nagan Gowda committee submitted its report in 1961. The following year,
Mysore State passed an order reserving 50 per cent of the seats in medical and
engineering colleges for Other Backward Classes (28 per cent were for backward
classes and 22 per cent for ‘more backward’ classes). This was over and above the
reservations for Scheduled Castes (15 per cent) and Scheduled Tribes (3 per cent). In
other words, the total reservation in the state was 68 per cent.7
This reservation policy was challenged in the Supreme Court in the case of M.R.
Balaji v. State of Mysore. Speaking for the court, Justice Gajendragadkar found fault
with the approach of the Nagan Gowda committee. He said that though caste could be a
factor in deciding social backwardness, it could not be the only factor.8 Its importance,
said the court, should not be exaggerated,9 and caste could not be the ‘sole or the
dominant test’ for deciding whether a community was socially backward.10 The
economic status of a community, the court felt, was as important, and those who are
‘deplorably poor automatically become socially backward’.11 Other factors, like
occupation or place of habitation, were also relevant.12 On the other hand, the Nagan
Gowda committee had adopted caste as ‘the predominant, if not the sole, test’, which
was wrong.13
The Supreme Court said that the committee ought to have only considered those
communities that were ‘well below the State average’ in terms of literacy to be
‘educationally backward’, e.g., those with less than 50 per cent of the state’s literacy
rate.14 It felt that the literacy levels of Muslims in the state were not so low as would
justify their being included in the category of educationally backward classes.15 It said
that in order to be considered eligible for reservations, a community had to be both
socially and educationally backward, not socially or educationally backward.16 The
backwardness of these communities, said the court, also had to be comparable to that of
Scheduled Castes and Scheduled Tribes,17 which is what the colonial Bombay
government had opined in 1933, as we have seen in Chapter 2. The Supreme Court
found that the categorization of communities into ‘backward’ and ‘more backward’ was
not permissible.18
However, the court was particularly alarmed by the fact that a very large chunk of
seats in Mysore had been reserved for backward communities. Nearly 90 per cent of the
state was now treated as backward.19 Article 15(4) of the Constitution, which allowed
the government to reserve seats in educational institutions, was considered at that time
to be an exception to the principle of equality of opportunity. The exception cannot
‘completely [exclude] the rest of the society’, said the court.20 This would lower the
standards of higher education in universities and was against the ‘national interest’
because the country’s need for ‘technicians, scientists, doctors, economists, engineers
and experts’ was far too great.21
Justice Gajendragadkar then went on to do something quite unexpected. He said that
reservations should be lower than 50 per cent of the total number of available seats.
How much lower depended on the ‘prevailing circumstances in each case’,22 adding that
reservations had to be within reasonable limits.23 He concluded by finding that
Mysore’s 68 per cent reservations were a fraud on the Constitution.24

The Creamy Layer and the Cake

A year later, the Supreme Court decided another case25 involving a principle called the
‘carry-forward rule’. At this time, around 12.5 per cent and 5 per cent of the posts in
Central government jobs were being reserved for Scheduled Caste and Scheduled Tribe
candidates respectively. However, it often happened that the government was not able to
find enough candidates to fill up the reserved seats. In such cases, general candidates
were recruited to the reserved seats, and the reservation was ‘carried forward’ to a
subsequent year. Reservations could be carried forward for a total of two years. As a
result of this, 65 per cent of the positions that were filled up in 1961 went to reserved
candidates. In T. Devadasan v. Union of India, the question was whether this violated
the 50 per cent rule articulated by Justice Gajendragadkar in M.R. Balaji’s case.
The carry-forward rule can be explained with an interesting example that the court
came up with. Let us assume, it said, that there are 100 posts which become vacant in
Year 1, that another 100 positions are available in Year 2, but only 50 posts are
advertised in Year 3. Since reservations for Scheduled Castes and Scheduled Tribes are
12.5 per cent and 5 per cent respectively, this means that 18 posts will be set apart for
them in Year 1 and another 18 posts in Year 2. If these posts are all carried forward into
Year 3, this will mean that there will be 18 + 18 + 926 posts reserved for Scheduled
Castes and Scheduled Tribes in Year 3, adding up to 45 posts. In short, there will be
only 5 posts vacant for general candidates in Year 3.27
This, said the court, violated the 50 per cent rule and was unconstitutional. The
reservations were ‘so excessive’ that they ‘practically [denied] a reasonable
opportunity for employment to members of other communities’.28 Reservations had to be
within reasonable limits,29 especially since Article 16(4), the provision dealing with
reservations, was an exception to Article 16(1), i.e., the principle of equality of
opportunity.30
The government’s lawyers tried to argue that in computing 50 per cent, the court
should look at the total strength of the cadre and not the total number of available posts
in any given year. For example, let us say that there are 1,000 posts in a government
department, out of which only 100 are presently occupied by Scheduled Castes and
Scheduled Tribes. Let’s say that 100 vacant posts are now advertised in that department
this year, but 75 of those new seats are reserved for Scheduled Castes and Scheduled
Tribes. Though 75 per cent of the advertised seats have been reserved (much higher than
the 50 per cent cap laid down in M.R. Balaji’s case), only 17.5 per cent of the total
cadre strength (i.e., 175 out of 1,000 posts) would be occupied by reserved category
candidates at the end of the year. The state argued that if one looked at the total number
of employees in the organization, not the total number of seats that became vacant in the
year, the 50 per cent rule was not being violated. However, the court rejected this
argument. It held that ‘each year of recruitment will have to be considered by itself’31
for the purposes of the 50 per cent rule.
However, in a dissenting judgement, one of the 42 solo dissents that he would write
in his career,32 Justice K. Subba Rao said that the carry forward rule was valid. He
wrote that India’s reservation policy was like a race with two horses in it—one, a ‘first
class race horse’, the other, an ‘ordinary’ one. To make both horses start from the same
position would be unfair—the first-class horse would easily defeat the ordinary one.
Allowing the ordinary horse to run a shorter distance or fixing a weight on the first-
class horse makes the race a level playing field.33 Of course, this was very unfair
towards the first-class horse. However, ‘injustice to individuals’, he said, was ‘inherent
in any scheme of reservation’.34
For perhaps the first time in the court’s history, Justice Subba Rao said that Article
16(4) was not an exception to Article 16(1). In Chapter 3, we saw that Ambedkar had
said that reservations were an exception to the rule of equality of opportunity. Justice
Subba Rao, however, felt that this was not so.35 He thought that the 50 per cent rule laid
down in M.R. Balaji’s case was ‘intended only to be a workable guide but not an
inflexible rule of law’.36 In any event, he believed that the rule applied to the total cadre
strength as the state had argued and not to each recruitment cycle.37
Thereafter, in 1975, the Supreme Court was called on to decide whether a service
rule in Kerala was valid. Lower division clerks employed in that state had to pass a test
within two years of getting promoted in order to keep their promotions. The rule
allowed the government to give Scheduled Caste and Scheduled Tribe clerks who had
been promoted a longer period of time to pass the test.38 In State of Kerala v. NM
Thomas,39 the Supreme Court found that this rule was fine.
However, while arriving at that conclusion, one judge, Justice Krishna Iyer, said
something quite interesting. In his judgement, he wrote that reservations had many
dangers. One of them was that its benefits were, ‘by and large … snatched away by the
top creamy layer of the “backward” caste or class’. This was the first time that the
words ‘creamy layer’ had been used in a Supreme Court judgement. Justice Krishna
Iyer believed that the well-off, affluent members of the backward classes were the ones
who were getting to ‘consume the whole cake’, while those genuinely in need, the
‘weakest among the weak’, were left with nothing.40
There were ‘vocal groups’, said Justice Krishna Iyer, who had ceased to be
backward thanks to employment and education. These groups, he said, wished to ‘wear
the “weaker section” label’ even though they were nearly as advanced as socially
forward communities.41 He felt that it was necessary for the government to constantly
reevaluate the progress made by backward classes in order to ensure that reservations
would not become ‘reverse discrimination’ against forward communities.42 In fact, he
wrote, research conducted at an institute of social studies in Patna ‘revealed a dual
society among harijans’—‘a tiny elite gobbling up the benefits and the darker layers
sleeping distances away from the special concessions’. For the latter, the provisions of
India’s constitution were a ‘noble romance’. The reservations ‘bonanza’, the study said,
went ‘to the “higher harijans”’.43 In another case, Justice Krishna Iyer wrote that India
had a somewhat ‘dubious obsession with “backwardness”’ and that the ‘politicking’
that went on with ‘castes labelled backward’ was something which demanded ‘judicial
examination’.44
Many judges who decided the N.M. Thomas case said things that were contrary to the
principles that had been laid down previously. For instance, some said that in deciding
whether reservations were reasonable, the entire cadre strength must be looked at, and
not each recruitment cycle.45 Adopting Justice Subba Rao’s reasoning in T. Devadasan’s
case, some said that the 50 per cent cap was not a hard-and-fast rule46 at all and that
Article 16(4) of the Constitution was not an exception to Article 16(1).47 For instance,
Justice Fazal Ali said that if the backward classes constituted 80 per cent of the
population, reserving 80 per cent of all posts and seats was reasonable.48
The state of Kerala soon adopted a creamy layer test for backward classes. Though
the Ezhava community was recognized as a backward class in the state, its members
who had an income above Rs 10,000 could not seek admissions in medical colleges.
The Supreme Court found that this rule was perfectly valid.49

‘The Constitution Does Not Permit Us to Be Arbitrary’

In the years that followed, the Supreme Court started diluting the principles laid down
by Justice Gajendragadkar in the M.R. Balaji case. For instance, the court found that it
was okay for governments to identify backward classes using caste as the primary
criterion.50 The 50 per cent cap began to be seen as a rule of thumb, not an inflexible
norm.51 The idea that reservations in educational institutions affected the excellence of
colleges started being questioned.52
For instance, in one case decided in 1981, the Supreme Court had to consider the
legality of a decision taken by Madhya Pradesh concerning medical admissions. Around
216 out of 720 seats in government medical colleges in that state were reserved for
Scheduled Castes and Scheduled Tribes. However, no such student could seek
admission into one of those colleges without securing the minimum number of marks: 40
per cent in all subjects and 30 per cent in each subject.53 Due to this rule, hardly any
reserved category students got admission into medical colleges. The state of Madhya
Pradesh therefore decided to do away with the rule altogether. In other words, it was no
longer necessary for reserved category students to secure a minimum score in their
exams in order to get admitted into a medical college in the state. As a result of this, all
the 216 reserved seats were likely to be filled up by Scheduled Caste and Scheduled
Tribe candidates.
The Supreme Court said that there was nothing wrong with what Madhya Pradesh had
done. The court did not accept the argument that this would lower the standards of the
medical profession in the state. After all, the court said, ‘the curriculum remains the
same for all’.54 The government had not made it any easier for backward class students
to become doctors—it had only relaxed the requirements for getting admission into a
college. The court was cognizant of the fact that many such students who had not even
secured the minimum number of marks would probably not be able to survive in college
and might drop out.55 This was a risk that the court was willing to take. It was no longer
acceptable to argue, as Justice Gajendragadkar had done in M.R. Balaji’s case, that
giving seats away to unmeritorious students would result in a waste of the state’s
resources. Madhya Pradesh might get fewer doctors because many of the reserved
category students might not be able to graduate, but that was okay.
A few years later, in another case, Justice Chinnappa Reddy, a judge of the Supreme
Court, launched an attack on the very idea of merit. ‘[W]hat is merit?’ he asked. A
backward class child ‘brought up in an atmosphere of penury [and] illiteracy … looked
down upon by tradition and Society’ could not fairly compete with others. Such a child
had ‘no books and magazines to read at home, no radio to listen [to], no TV to watch, no
one to help him with his homework’, and had parents who were ‘either illiterate or so
ignorant and ill-informed that he cannot even hope to seek their advice’. Such a boy, the
judge said, had to ‘trudge to the nearest public reading room to read a newspaper to
know what is happening in the world’. How could such children be said to have no
merit if they, with all their disadvantages, were able to score 40 per cent or 50 per cent
marks in an examination, though the ‘children of the upper classes who have all the
advantages, who go to St Paul’s High School and St. Stephen’s College, and who have
perhaps been specially coached for the examination may secure 70, 80 or even 90 per
cent’, he asked.56
According to Justice Reddy, merely securing high marks in an exam did not
necessarily make a person a good administrator. Efficient administrators, he thought,
needed other qualities like ‘the capacity to understand with sympathy’ and to ‘tackle
bravely the problems of a large segment of [the] population’. Though efficiency in
government service was important, the argument that reservations would breed
inefficiency in government was being used as a ‘camouflage’ by the upper classes ‘to
monopolise the services’, he added.57
In his judgement, Justice Reddy contradicted virtually every principle which had
been laid down by Justice Gajendragadkar in the M.R. Balaji case. So, for instance, he
found that caste could be used as the ‘primary index’ to determine social
backwardness.58 Mere poverty was not a criterion for determining backwardness,59 he
felt, though the poor certainly needed help. He found the idea of poor Brahmins being
given reservations ‘too grotesque even to be considered’.60 Since agricultural income
was not taxed, he believed that rich landowners would be able to establish that they
were backward if a person’s finances were made an indicator of backwardness.61
 Contrary to M.R. Balaji’s case, Justice Reddy said that Other Backward Classes
need not be as backward as Scheduled Castes and Scheduled Tribes.62 Any other rule,
he said, would ‘practically nullify’ the reservations in favour of Other Backward
Classes, since it would be hard to find communities that were as backward as
Scheduled Castes and Scheduled Tribes.63 Unlike what Justice Gajendragadkar had
said, Justice Reddy thought it was okay to classify Other Backward Classes into
backward and more backward.64 Perhaps most importantly, he said that the court could
not put any cap on the quantum of reservations. Justice Gajendragadkar’s 50 per cent
rule, he said, was arbitrary, and ‘the Constitution does not permit us to be arbitrary’.65
Though reservations for backward classes might be used as a ‘vote-catching device’ by
unscrupulous political parties, he thought that this was a price that India would have to
pay. ‘[O]ut of evil cometh good’, he said.66
However, all the five judges in that case wrote separate judgements, and Justice
Reddy’s views did not have the support of the majority. Interestingly, unlike Justice
Gajendragadkar, who was a high-caste Hindu, Justice Reddy was the son of a fifth-
generation Roman Catholic. An atheist, Reddy refused to ‘be classified to any religion
or any community’ preferring to be considered ‘a human being and an Indian’.67

Mandal and Sawhney

Over twenty years after the Kalelkar Commission submitted its report attempting to
define who the Other Backward Classes were, in January 1979,68 Prime Minister
Morarji Desai appointed a commission headed by a former Chief Minister of Bihar,
Bindhyeshwari Prasad Mandal,69 to objectively identify Other Backward Classes. The
new commission noted that the Kalelkar committee had ‘not formulated any objective
criteria for classifying Other Backward Classes.’70 Determined not to make the same
mistake, the Mandal Commission came up with eleven criteria for figuring out who the
Other Backward Classes were.71
There were four criteria for social backwardness: castes that were considered to be
socially backward by others, those that depended primarily on manual labour for their
livelihood, those whose members got married below 17 years at a rate higher than the
state average,72 and, interestingly, castes whose women entered the workforce at a rate
25 per cent higher than the state average were socially backward.
The Mandal Commission identified three factors for educational backwardness:
castes whose children between the ages of 5–15 years never attended school at a rate
25 per cent more than the state average, those whose students of that age dropped out of
school at the same rate, or castes who had a proportion of ‘matriculates’ (i.e., those
who finished high school) 25 per cent lower than the state average were educationally
backward.
Similarly, there were four criteria for economic backwardness: castes whose average
value of family assets was 25 per cent below the state average, those whose families
lived in ‘kuccha houses’ 25 per cent more often than the state average, castes whose
source of drinking water was beyond half a kilometre away for more than 50 per cent of
their households, and those where the number of families that had ‘consumption loans’
was 25 per cent above the state average were economically backward.
The commission came up with a points system73 for its eleven criteria, in which
social backwardness was given the highest weightage—12 out of 22 points were
allotted to this category. On the other hand, 6 points and 4 points were allocated to
educational and economic backwardness respectively. Any caste that earned 11 out of
22 points on this scale was considered to be socially and educationally backward.
According to the commission’s calculations, the Other Backward Classes accounted for
52 per cent of the population of India.74
The Mandal Commission avoided the mistakes made by the Nagan Gowda
committee. Now, only communities that were well below or above the state average in
certain indices were considered backward. Further, though caste was an important
factor in its calculations, it was not the only measure of social backwardness. The
Nagan Gowda committee had said that castes that were thought to have a low status
were, for this reason alone, socially backward. For the Mandal Commission, this was
only one of many factors that went into social backwardness. However, though Justice
Gajendragadkar in M.R. Balaji’s case had said that economic criteria must be as
important as social criteria for determining social backwardness, the Mandal
Commission gave economic criteria the least importance.
The Mandal Commission recommended 27 per cent reservation for Other Backward
Classes. This figure, 27 per cent, was calculated in order to avoid violating the 50 per
cent rule laid down by the Supreme Court in M.R. Balaji’s case. Reservations for
Scheduled Castes and Scheduled Tribes at this time were 22.5 per cent of the total
available seats. Though the Mandal Commission believed that Other Backward Classes
constituted 52 per cent of India’s population, the highest representation that could be
given to them was therefore 27 per cent.75 Thus, though the reservations given to
Scheduled Castes and Scheduled Tribes of 15 per cent and 7.5 per cent was roughly
proportionate to their numbers in the population, Other Backward Class reservations
fell substantially short of proportional representation.76
 By the time the Mandal Commission submitted its report, the Janata Government had
been defeated by Indira Gandhi’s Congress in the elections. The report was shelved for
ten years, until 13 August 1990, when the government of Prime Minister V.P. Singh
issued an office memorandum which, for the first time in India’s history, gave
reservations to Other Backward Classes in Central government jobs. Until then, only
Scheduled Castes and Scheduled Tribes had posts reserved for them in Central
government services. After this office memorandum was issued, 27 per cent of the
vacancies in civil posts and services under the Central government were reserved for
socially and educationally backward classes.77 It was perhaps no small coincidence that
at this time, regional backward caste parties entered national politics, about a fourth of
India’s parliament had Other Backward Class MPs, and 20 per cent of Prime Minister
V.P. Singh’s cabinet had backward caste members.78
Thereafter, once Prime Minister Narasimha Rao came to power, another office
memorandum was issued on 25 September 1991. This provided that preference in
reservations would be given to the poorer sections of the socially and educationally
backward classes. It also said that 10 per cent of the vacancies in Central government
jobs would be reserved for economically backward communities.79
Several people filed cases in the Supreme Court challenging these office memoranda
issued by the V.P. Singh and Narasimha Rao governments. Included among them was a
lawyer by the name of Indra Sawhney.80 These cases, which were decided by a bench of
nine judges of the Supreme Court,81 would be known by her name.82
The Supreme Court in Indra Sawhney’s case said that the criteria formulated by the
Mandal Commission were okay.83 While caste could be an important factor to determine
backwardness, it could not be the only factor.84 It was always open to a government to
identify Other Backward Classes without reference to caste altogether, by looking at
factors like occupation, income, and other criteria.85 Social backwardness was
essential.86 However, Justice Jeevan Reddy relied on colonial history and found that in
pre-independence India, ‘the expressions “class” and “caste” were used
interchangeably’ and ‘caste was understood as an enclosed class’.87 It was held that a
caste could be a class as well.88
Disagreeing with Justice Gajendragadkar’s view in M.R. Balaji’s case, the view
which was also taken by the government of colonial Bombay in 1933, the court held that
the Other Backward Classes did not have to be as backward as Scheduled Castes and
Scheduled Tribes.89 In fact, the phrase ‘backward classes’ in Article 16(4) of the
Constitution was wider than the phrase ‘socially and educationally backward classes’
used in Article 15(4).90 In other words, said the court, a caste could qualify for
reservations in government jobs even if it was educationally forward, though it would
not qualify for reservations in educational institutions. It was okay for a government to
classify other backward communities into backward and more backward.91
However, the court held that backwardness could not be decided on the basis of
economic criteria alone.92 For instance, Justice Sawant felt that the high-caste poor
were not socially backward and if they were given reservations along with Other
Backward Classes then all the seats would be monopolized by them.93 The court
therefore struck down the 10 per cent reservations for economically backward
communities that had been provided by Prime Minister Narasimha Rao’s government.94

Far-Flung and Remote Areas

Perhaps the greatest disagreement among the nine judges in Indra Sawhney’s case took
place over the question of whether to accept Justice Gajendragadkar’s 50 per cent cap
on reservations. The judges held, contrary to what Ambedkar had said in the Constituent
Assembly, that Article 16(4) of the Constitution was not an exception to Article 16(1)—
in other words, that reservations were not an exception to the principle of equality of
opportunity.95 However, the judges disagreed with each other about whether this meant
that there could be a cap on reservations.
The majority of the court agreed that the 50 per cent rule would be sacrosanct and
could only be violated in ‘certain extraordinary situations’ to provide for backward
communities in far-flung or remote areas of the country.96 In other words, reservations
could not be provided to backward classes in proportion to their population—the
Constitution prescribed ‘adequate’, not ‘proportional’ representation in government
jobs.97 On the carry-forward rule, the judges agreed that the 50 per cent cap would
apply to each year and would not be fixed on the basis of the strength of the cadre.98
In perhaps the most significant move, the court decided that the ‘creamy layer’ would
have to be mandatorily excluded from the Other Backward Classes. How were
members of the creamy layer to be identified? According to Justice Jeevan Reddy,
economic criteria alone would not do.99 A migrant worker who goes to the UAE might
have more money than the other members of his caste, but he might still be backward.
Further, money could not be used as a uniform measure of the creamy layer throughout
the country either—a sum of Rs 10,000 would go much further in rural India than in the
cities. Justice Reddy said that people like factory owners (who were in a position to
employ many people) and government servants were a part of the creamy layer.100
 Justice Sawant wrote that a family was a part of the creamy layer if it had the
capacity to compete with the forward castes, not just if it were relatively better off
compared to the other members of its own caste. In other words, a creamy layer family
was one which had ‘cross[ed] the Rubicon of backwardness’. He said that it could
hardly be argued that ‘once a backward class, always a backward class.’101 Justice
Pandian was perhaps the only one who felt that even the creamy layer should be
included in the Other Backward Classes.102 However, the court agreed that Scheduled
Castes and Scheduled Tribes were presumed to be backward103—courts could not
investigate whether the castes included in those categories were backward or not.104
Finally, the court decided, by a majority, that reservations could not be allowed in
promotions,105 reversing decades of jurisprudence on this point.106 Justice Jeevan Reddy
felt that if reservations were allowed in promotions, the members of reserved classes
would feel as though they ‘need not have to compete with others but only among
themselves’. As a result of this, they will not feel any need to ‘work, compete and
excel’, as they would think that ‘their promotion is assured’. This would also create a
sense of frustration amongst those from the general categories. ‘Crutches cannot be
provided throughout one’s career’, he said. He added that this would not prevent
backward classes from reaching the upper echelons of government service, since direct
recruitments often took place in those positions, where reservations were permitted.
District judges, for instance, are often directly recruited and not promoted from the
lower judiciary.107
In short, the court in the Indra Sawhney case accepted the Mandal Commission report
but asked the government to exclude the creamy layer from the Other Backward Classes.
It accepted the 50 per cent rule but said that it could be exceeded in rare cases. It
refused to allow 10 per cent reservations for economically backward communities and
rejected reservations in promotions.108

Ambedkar, Gajendragadkar, or Chinnappa Reddy?

At the time that the Indra Sawhney case was decided, there were three different points
of view on reservations. The first approach, a conservative one, was that of India’s
founding fathers like Nehru and Ambedkar. Nehru frowned on quotas while Ambedkar
wanted them to be restricted to a minority of seats. The supporters of this approach
believed that reservations were an exception to the principle of equality of opportunity
and that an exception cannot swallow the rule.
 The second viewpoint, a centrist one, was of judges like Justice Gajendragadkar,
who suggested that reservations should extend to no more than 50 per cent of the
available seats or posts. They drew on the arguments made by Ambedkar in the
Constituent Assembly and said that Article 16(4) was an exception to Article 16(1).
Giving backward classes too large a share of the pie, they suggested, would be unfair to
the other communities. The Constitution itself only spoke of ‘adequate’ (not
‘proportional’) representation for the backward classes, they pointed out. Allowing
sub-standard students to get reserved seats in educational institutions, they said, would
mean that many of them would be unable to complete their courses and India would be
deprived of the doctors, scientists, and economists that she desperately needed. Yet,
they recognized that reservations could not be restricted to a tiny minority of seats.
The third viewpoint, a liberal one, was of judges like Justice Subba Rao, Justice
Fazal Ali, and Justice Chinnappa Reddy who said that there should be no hard-and-fast
rule restricting reservations to 50 per cent. They argued that backward classes should
be entitled to a share commensurate with their numbers in the population. Where did the
50 per cent cap come from, they asked. It was an arbitrary formulation picked out of a
hat by an over-zealous court. What is merit, they asked. A boy from a socially
disadvantaged community who scored 40 marks had as much merit as someone from St
Stephen’s college who scored 70. Allowing them to enter medical colleges, they said,
did not lower the standard of medical education, since the curriculum remained the
same for one and all.
The solution that the Supreme Court eventually came up with in the Indra Sawhney
case attempted to balance these different viewpoints. Reservations would not be
restricted to a minority of seats as Ambedkar had said. They could not extend to 70 per
cent either, contrary to what liberals like Justice Subba Rao would have preferred. Yet,
at the same time, they recognized that Justice Gajendragadkar’s 50 per cent cap required
exceptions in extraordinary cases. On the other hand, in deciding that reservations
would not be permissible in promotions, the court swung towards the conservative
view adopted by the founding fathers. Even Justice Gajendragadkar had not taken such
an extreme position. In fact, one of the first judgements of the Supreme Court
recognizing reservations in promotions was authored by him.109
When Justice Krishna Iyer first spoke of the ‘creamy layer’, he applied it to
‘Harijans’. The word ‘Harijan’, as we have seen, was originally used for the
‘depressed classes’, those who had suffered from the stigma of untouchability, whom
the Constitution now called Scheduled Castes. However, the Supreme Court in Indra
Sawhney’s case did not apply the concept of the creamy layer to Scheduled Castes or
even the Scheduled Tribes. It was meant to weed out well-off or forward sections of the
Other Backward Classes. The question which then arises is: why should the creamy
layer concept not apply to Scheduled Castes and Scheduled Tribes as well?110 One
answer might be that reservations were meant to compensate castes that had suffered the
stigma of untouchability for centuries. However, as we saw in the first chapter,
Scheduled Tribes were usually not untouchable communities. Why should the creamy
layer doctrine not apply to them? The reason Scheduled Tribes were considered to
deserve reservations was because of their backwardness. If that is so, then why should
those who have come out of their backwardness continue to benefit from reservations,
especially when their counterparts in the Other Backward Classes cannot do so? The
creamy layer concept was eventually taken to its logical conclusion by the Supreme
Court in the Jarnail Singh case, which we will see in the next chapter.
We now know with some more clarity who the ‘Other Backward Classes’ are. The
answer depends on the kind of reservation that is being spoken about. The Constitution
grants reservations in educational institutions to ‘socially and educationally backward
classes’. Unlike in government jobs, here, both social and educational backwardness
are essential. Reservations in government jobs, on the other hand, are provided to those
who primarily suffer from some kind of social disability—typically castes (Hindu or
otherwise) that are backward (e.g.,manual labourers, those who marry early, castes
considered low), minus their creamy layer. The Mandal Commission had said that a
caste whose women are in the workforce is socially backward for that reason, because
upper-caste women would usually not want to work out of choice. Perhaps such criteria
now need revision.