I apologize for my delayed comments on this issue. It took me quite a while to review and internalize these judgments. This judgment had created more cobwebs than it had cleared. Ascertaining the ratio itself was an exercise in and of itself. I was appalled at the quality of judgment writing which has hit a new low. The judgment of Justice Pasayat was very confusing.

Ashok Kumar Thakur is not about whether quotas should continue. It was not about quotas v. merit. Affirmative action for the disadvantaged classes is a constitutional mandate. The only question is who gets them. If the program is designed in a manner which undermines equality and reinforces caste (which Ambedkar termed “anti-national”), it upsets the entire constitutional scheme of promoting equality and the judiciary has to scrutinize it. Unfortunately the judiciary did not live up to its job. The judgment suffers from eight fatal flaws listed below. Each of them deserves an independent post. But I do want to present them when the issue is still alive.

First, and the most fatal flaw of this judgment is that it deepens the institution of caste based on an improper understanding of the text and the context of the Supreme Court ruling in Indra Sawhney. Indra Sawhney does not permanently embed caste as a starting point for identifying backward classes. Indra Sawhney only blessed caste, it did not mandate it. Caste might have been a legit imitate starting point for identifying backward class in the 1970 and 1980’s when the Backward Class list was prepared, but whether caste is still a proxy for backward class in 2008 can only be decided based on an empirical independent investigation. Not only did Indra Sawhney preclude a more scientific method for identifying backward classes, it mandated it by requiring a periodic independent investigation. The judges had a rare opportunity to go beyond caste by mandating the government to prepare a BC list, which was not based on caste. That would have been consistent with Indra Sawhney and would have taken the politics out of the entire business of affirmative action.

Second, the judgment chants, but does not enforce or substantiate the empirical investigation requirement laid down in Indra Sawhney. Indra Sawhney laid down several requirements regarding a periodic investigation of backward classes to ensure that only the deserving get the benefits. Indra Sawhney contemplated that every ten years the Backward Classes Commission would lay down the criterion for backwardness and based on those criterion, it would identify classes who could be called “ backward classes”. Between each survey period, if any person/group felt that it was unfairly excluded or included, it could approach the BC Commission which would decide based on the criterion laid down to identify backward classes. A similar process is not contemplated for SC and ST because “backward class” is a shifting class and not a fixed class like Scheduled Caste and Scheduled Tribes. “Backwardness” is a relative state of affairs based on the general advancement of the society and therefore the need for a periodic review. But over the last two decades very little action has been taken. Caste has become a “one-way ticket” to the backward class list. Once a caste is identified, very rarely has it gone out of the list. The BC has not been revised in many states from the seventies. The judgment lost focus on these requirements and instead gives gave some broad unenforceable guidelines on investigation. The judges, Justice Balakrishnan in particular, brushed aside this entire issue by saying that the identification of BC list was not based solely on caste and therefore it was proper. He equated the power of the Backward Class Commission to decide the function of inclusion and exclusion with the requirement to conduct an independent investigation. They were never meant to be the same.

Third, the judgment confuses “backward class” under Article 15(5) (and 15(4)) and Article 16(4). Under 15(5) (and 15(4)), numerical majority is irrelevant. A section of the society can comprise 70% of the population and still be backward. But under Article 16(4), a section of the society must not only be backward, but must also be “inadequately represented.” It is only in Article 16(4) that population and its representation in state services acquires significance and not under Article 15(4) and Article 15(5).

Fourth, the judgment messed up judicial review standard in affirmative action by dumping strict scrutiny. India never had complete strict scrutiny, but only partial strict scrutiny. Dumping the entire strict scrutiny standard is like dumping the baby with bath water. And the court only confused the existing standard. Justice Balakrishnan mentioned, but did not flesh out the alternative “ex facie unreasonable” standard. Instead, he did the unthinkable by putting the burden on petitioners to show that the backward classes do not constitute 27%. Although that’s an irrelevant question because of reasons pointed above, the burden is always on the state to justify affirmative action and not the petitioners. That was the holding in Indra Sawhney and in the cases before that.

Fifth, the only achievement of this judgment was its strong emphasis on creamy layer, but it confused this area by lack of clarity on educational creamy layer. Mr. Venkatesan in his blogs said that chief justice has clarified that there was no such ruling. The Chief Justice has no power to issue such a clarification. Only a Supreme Court bench can issue such a clarification. Once a judge delivers a judgment, he does not have the power to issue a clarification. (unless there is a review petition or another bench interprets that judgment)

The absence of any mention of this aspect in the final order – whether signed by all the judges or not – is of no consequence. It has been a settled principle that what matters is the reasoning of the judgment and not the final conclusion. (The SC had ruled that the final conclusion in Kesavananda Bharti has no legal significance.)
There is a strong justification for creating an educational creamy layer exception. The justification given in Indra Sawhney for creating a creamy layer is because they as members of the class need to share common characteristics and a creamy layer no longer shares those characteristics and therefore the connection ceases with the backward class. The same logic applies to an educational creamy layer. Once a person from a backward class graduates, he is no longer “educationally backward”. He loses that connection with the rest of the class and he does not share those characteristics. So even though he might come from “socially and educationally backward class”, he is no longer educationally backward by virtue of his educational attainments. His connection with that class ceases.

Sixth, the judgment stamped the poor constitutional advice given to the HRD Ministry by reconciling Article 15 (5) and Article 15(4). Before Article 15(5), reservation for OBC’s in state aided institutions could be provided under Article 15(4). Several states used that mechanism. In case of unaided or private institutions, the judgment in T.M.A. Pai and Inamdar precluded reservation and admission was to be only on merit. The court said that reservation would be an unreasonable restriction under Article 19(1)(g). The primary purpose of Article 15(5) was only for private institutions. For state institutions, Article 15(4) was already there. Instead of recognizing this simple constitutional history, the judgment went into this elaborate discussion about reconciling Article 15(4) and Article 15(5). There was no such need. The government could have provided reservation in IIT and IIM and other central government educational institutions even without Article 15(5).

Seventh, the Court deviated from precedent when it declined to hear the constitutional challenge of affirmative action in unaided educational institutions. The reason – there were no private institutions before the court. The court could not have thought of a poorer excuse! This issue came up extensively. The entire point of Article 15(5) was about private unaided institutions. The court assumed that only private institutions had the right to question caste-based reservations in Article 15(5). Both the teacher and the students and even the general public have the right to question an amendment which mandates caste based quotas in private institutions which do not take support from the state. It is obvious that standing was just an excuse to defer this issue to a future court.
Surprisingly Justice Balakrishnan – who declined to decide the constitutional validity of reservations in private colleges since the Act in question only provided reservation in central government institutions – decided the issue of creamy layer for SC’s and ST. Even that question was also not in issue since Act was only confined to providing reservation for OBC’s in central institutions.

Eighth, the judges never considered the issue of proportionality. In India, the affirmative action discourse has been focused only on quotas. Unlike the United States, the primary purpose of affirmative action in India is to compensate for past discrimination and not promote diversity or any other objective. In such a case, the compensation must be proportional to the deprivation suffered. Quotas stand at the end of the chain. Some forms of deprivation may not justify even a full quota, but possibly a milder form of affirmative action in the form of increase in marks or similar measures. Again, this can happen only though an independent investigation which the judges simply did not focus on.


  1. Vivek, You have written a crisp review of the judgment.You have touched upon issues that ought to have been dealt with the Court.Basically I find that the judgments go by the verdict in Mandal case instead of taking that
    forward or breaking a new ground.
    Justice Bhandari tries to bo beyond the Mandal judgment in his
    view but acknowledges the contraint there.It is a pity that
    the Court did not use this case to revisit the issue once again taking into account Mandal judgment
    and post Mandal developments.The judgments leave many issues unaddressed and there are many
    unanswered questions, particularly whether the lack of data is an issue or not in justifying reservations or the question of
    identifying which classes are
    not ‘adequately represented’.
    I am in the process of writing an
    article on the judgment and this
    blog post will be cited and discussed therein. If you have not read it already, read the judgment of the Andhra Highcourt
    on reservations for Muslims.

  2. Thanks for comments. I agree with you. the Court should have gone ahead, but it locked itself in precedent. Do send me the article when you are done. I am working on a legal article on the Muslim reservation case. I have worked extensively on that case.

  3. I was just wondering about your eighth point – the proportionality argument. In Indra Sawhney, B.P. Jeevan Reddy J. cited the American case of Metro Broadcasting Inc. v. Federal Communications Commission, wherein the Court had opined that “benign race conscious measures are constitutionally permissible even if they are not designed to compensate victims of past governmental or societal discrimination so long as they serve important governmental objectives and are substantially related to achievement of those objectives.” Reddy J. agreed with this holding, and said that the same applies to India. If, then, we were to accept that the rationale for quotas is not compensation, would that affect your proportionality argument?

Leave a Reply

Your email address will not be published. Required fields are marked *