NALSA v. UoI and Koushal v. Naz: Acts against the order of nature?

In the last post, I argued that after NALSA v. UoIKoushal v. Naz ought to be reconsidered, since in rejecting the distinction between sexual conduct and sexual identity, NALSA removes the intellectual foundation upon which the entire Koushal decision was based. In conversation with Danish since then, it seems to me that we might not even need to go that far. Here is how.

While Koushal v. Naz is commonly understood to have recriminalized homosexual intercourse in India, it did not actually do quite that. The Delhi High Court, in Naz Foundation v. NCT, held that insofar as S. 377 criminalizes same-sex intercourse between consenting adults, it is unconstitutional. Recall, however, that S. 377 does not criminalize homosexual intercourse in so many words. It criminalizes “carnal intercourse against the order of nature.” The conceptual middle-step, of course, requires associating homosexual intercourse with carnal intercourse against the order of nature.

And, interestingly, that is a finding that the Supreme Court in Koushal never made. Let us go back to the notorious Paragraph 42:

“Those who indulge in carnal intercourse in the ordinary course and those who indulge in carnal intercourse against the order of nature constitute different classes and the people falling in the later category cannot claim that Section 377 suffers from the vice of arbitrariness and irrational classification. What Section 377 does is merely to define the particular offence and prescribe punishment for the same which can be awarded if in the trial conducted in accordance with the provisions of the Code of Criminal Procedure and other statutes of the same family the person is found guilty. Therefore, the High Court was not right in declaring Section 377 IPC ultra vires Articles 14 and 15 of the Constitution.”

So there is nothing in Paragraph 42 that indicates what “carnal intercourse against the order of nature” is, and the Court makes no finding on the point. True, the Court refers to precedents on what constitutes “sodomy”, and so on, but it does not incorporate any prior definition into its verdict. Now consider Paragraph 51:

“Respondent No.1 attacked Section 377 IPC on the ground that the same has been used to perpetrate harassment, blackmail and torture on certain persons, especially those belonging to the LGBT community. In our opinion, this treatment is neither mandated by the section nor condoned by it and the mere fact that the section is misused by police authorities and others is not a reflection of the vires of the section.”

Here again, it’s not so much as what is stated, but what is omitted, that is significant. The Court specifically refers to the LGBT community, and again refuses to rule on whether the LGBT community is covered by the ambit of 377.

Indeed, if there is one thing that emerges out of all the precedents that the Court cites, it is that there’s no consensus on what the term means. One case holds that all non-procreative sex is against the order of nature, while another holds that that particular theory is outdates. Another finds oral sex to fall within the ambit of 377. Indeed, in paragraph 36, the Court specifically observes:

“The understating of acts which fall within the ambit of Section 377 has changed from non-procreative (Khanu v. Emperor) to imitative of sexual intercourse (Lohana Vasantlal v. State AIR 1968 Guj 352) to sexual perversity (Fazal Rab v. State of Bihar AIR 1963, Mihir v. Orissa 1991 Cri LJ 488).”

So, is homosexuality “perverse”? Again, the Court refuses to make a finding on point.

Consequently, the upshot of Koushal v. Naz is that although it is no longer the case that S. 377 expressly excludes homosexuals, it is also not the case that S. 377 includes them. That judgment is one to be made by later courts.

NALSA v. UoI is the judgment that does so.

The NALSA Court understands this, because in paragraph 18, it holds that because of Koushal, it will not rule on the constitutionality of S. 377. That, however, leaves it entirely free to interpret 377.

Now, in paragraph 20, the NALSA Court notes:

Each person’s self-defined sexual orientation and gender identity is integral to their personality and is one of the most basic aspects of self-determination, dignity and freedom.”

Notice that this paragraph is not limited to transsexuals. It refers to sexual orientation as such (especially because it uses the phrase in conjunction with “gender identity”). Now if sexual orientation is integral to personality, and a basic aspect of self-determination (and nobody – presumably not even the Koushal bench – would deny that same-sex intercourse is expressive of sexual orientation) – in what sense can we say that expressing your personality and self-determination is “against the order of nature”?

The Court then cites a number of international conventions and legal instruments, all of which prohibit discrimination on the basis of gender identity as well as sexual orientation. In paragraph 55, it holds that discrimination on grounds of sexual orientation or gender identity violates the equal protection of laws. And in paragraph 77 – the last paragraph of Justice Radhakrishnan’s judgment, he holds that:

We, therefore, conclude that discrimination on the basis of
sexual orientation or gender identity includes any discrimination, 
exclusion, restriction or preference, which has the effect of
nullifying or transposing equality by the law or the equal protection
of laws guaranteed under our Constitution.”

Now admittedly, Justice Radhakrishnan makes it clear that the scope of his judgment is limited to the rights of transsexuals. But surely it defies logic to argue that discrimination on the basis of sexual orientation is prohibited in the case of transsexuals, but allowed in all other cases? In other words, the logic of the NALSA judgment seems to make it clear – even if it does not say so in express terms – the “sexual orientation”, per se, is a protected category. And once again, recall that Koushal made no finding on that specific point.

In this sense, NALSA v. UoI can be taken to be a clarification of KoushalKoushal holds that it is constitutional to criminalise carnal intercourse against the order of nature. NALSA holds that sexual acts that are characterised purely in terms of the sexual orientation are not acts against the order of nature. In effect, this is the same as the holding of the Delhi High Court. However, while the Delhi High Court held that 377 is unconstitutional insofar as it criminalises same-sex intercourse between consenting adults – the logic of NALSA requires the conclusion that 377 is constitutional only because it does not criminalise same-sex intercourse between consenting adults. In that case, then – because NALSA is not a case about homosexuals, and therefore not precedent on that point - perhaps the next step is to approach the Court for a judgment clarifying the scope of S. 377 as applied to homosexuals.

None of this, of course, is an endorsement of anything in Koushal, or anything that Koushal stands for. The act/identity distinction in Koushal remains flawed and indispensable. Its cavalier treatment of the “so-called rights of a minuscule minority” remains indefensible. But insofar as we’re considering the present state of law, it is perhaps arguable that, reading Koushal and NALSA together, same-sex intercourse between consenting adults can no longer be criminal in India.


Leave a comment

Filed under Equality, Non-discrimination, Sex Equality, Sexuality

NALSA v. UoI: The Supreme Court on transsexuals, and the future of Koushal v. Naz

Today, in NALSA v. UoIthe Supreme Court issued a landmark ruling recognising transsexuals as a third gender, and upholding their rights to equality (Article 14), non-discrimination (Article 15), expression (Article 19(1)(a) and autonomy (Article 21). The judgment involves a wide-ranging discussion of international law and domestic legislation in other countries, engages reams of evidence of actual discrimination against transsexuals in Indian society, and discusses the idea of human rights. It also, as I shall argue, entirely destroys the foundation of Koushal v. Naz, last December’s decision on LGBT rights.

In Paragraph 11 of the case, Justice Radhakrishnan defines “transgender” as an “umbrella term for persons whose gender identity, gender expression or behavior does not conform to their biological sex.” After a brief historical excursion into the history of the transgender community in India, he observes, in Paragraph 17, that S. 377 was brought in at at time when transgenders were thought to come within its ambit, and then, in paragraph 18, he notes that “Section 377, though associated with specific sexual act, highlighted certain identitiesand was used as an instrument of harassment and physical abuse…” In the same paragraph, he also holds that in light of Koushal v. Naz, the Court will here “express no opinion on [its constitutionality].” As we will see, however, the matter is not quite so simple.

Indeed, issues of gender identity and sexual orientation are inextricably bound up with each other through the judgment, and the Court – wisely – makes no effort to separate them. Immediately after his analysis of S. 377, in a section titled “Gender Identity and Sexual Orientation”, Justice Radhakrishnan begins by noting that “gender identity is one of the most fundamental aspects of life… it refers to each person’s deeply felt internal and individual experience of gender… including the personal sense of the body which may involve a freely chosen modification of bodily appearances or functions by medical, surgical or other means and other expressions of gender, including dress, speech and mannerisms.” (Para 19)

Notice here how gender identity and expressing that identity through conduct, such as dress and speech, are inseparable. This is a point we shall return to.

Justice Radhakrishnan then observes that “each person’s self-defined sexual orientation and gender identity is integral to their personality and is one of the most basic aspects of self-determination, dignity and freedom.” (Paragraph 20) This sets the tone for the rest of his judgment, where the two concepts – although distinct – are run together for the purposes of claims to rights.

After referring to international legal principles and foreign judgments, that are deeply solicitous of transgender rights (paras 21 – 42), Justice Radhakrishnan cites evidence (ironically, of a similar nature to that cited in Koushal) of widespread oppression against transgenders in India. (Paragraphs 45 – 46) He follows this up with an account of India’s obligations to follow international human rights law – and, in the absence of Indian legislation, engages in a kind of incorporation by reference, to argue that the international conventions ought to be read into Articles 14, 15, 19 and 21 (Paragraph 53). Armed with this, he has no trouble in holding that the “non-recognition of Hijras/transgender persons denies them equal protection of law… thereby leaving them extremely vulnerable to harassment, violence and sexual assault.” (Paragraph 55) Similarly, he argues that the point of Article 15′s non-discrimination guarantee is to prevent differential treatment of persons “for the reason of not being in conformity with stereotypical generalizations of binary genders… thereforethe discrimination on the ground of sex under Articles 15 and 16 includes discrimination on the ground of gender identity.” (Paragraph 59) Because of the historic discrimination against transsexuals, he also holds that the State must provide them with affirmative action under Article 16(4) of the Constitution.

The most interesting part of the judgment, however, is Justice Radhakrishnan’s analysis of Article 19(1)(a). He holds that “Article 19(1)(a) of the Constitution states that all citizens shall have the right to freedom of speech and expression, which includes one’s right to expression of his self-identified gender.” (Paragraph 62) After citing a few American cases on point, we come to the heart of the judgment, that is, Paragraph 66:

Gender identity… lies at the core of one’s personal identity, gender expression and presentation and therefore, it will have to be protected under Article 19(1)(a) of the Constitution. A transgender’s personality could be expressed by the transgender’s behavior and presentation. State cannot prohibit, restrict or interfere with a transgender’s expression of such personality, which reflects that inherent personality.” 


The values of privacy, self-identity, autonomy and personal integrity are fundamental rights guaranteed to the members of the transgender community under Article 19(1)(a)… and the State is bound to protect and recognize those rights.” 

A standard Article 21 analysis follows (Paragraph 67 onwards), but this would be the ideal point of departure to discuss NALSA v. UoI and Koushal v. Naz.

Recall that in Koushal v. Naz, it was argued that S. 377, insofar as it criminalises same-sex intercourse between consenting adults, violates their rights under Articles 14 and 15. In Paragraph 42 of Koushal, the Court rejects that argument. Let us excerpt the paragraph in full:

“Those who indulge in carnal intercourse in the ordinary course and those who indulge in carnal intercourse against the order of nature constitute different  classes and the people falling in the later category cannot claim that Section 377  suffers from the vice of arbitrariness and irrational classification. What Section  377 does is merely to define the particular offence and prescribe punishment for  the same which can be awarded if in the trial conducted in accordance with the  provisions of the Code of Criminal Procedure and other statutes of the same family the person is found guilty. Therefore, the High Court was not right in declaring Section 377 IPC ultra vires Articles 14 and 15 of the Constitution.”

The Court’s argument here rests upon a tight conceptual distinction between conduct and identity. Under a normal Article 14 analysis, the Court would have had to examine the classification, examine the rational nexus to a State purpose, and then examine the legitimacy of that purpose. The Court dodged that entire chain of argumentation by holding that all S. 377 did was to classify not persons, but acts – acts of carnal intercourse against the order of nature, and those in accordance with the order of nature. Hence, Article 14 was never attracted. This also allowed the Court to dodge the Article 15 argument, and engage with the Delhi High Court’s analysis of “sexual orientation” coming within the meaning of the word “sex”, because there was no classification of persons at all. In one fell swoop, therefore, the Supreme Court saved itself the trouble of analysing S. 377 on the touchstone of either Article 14 or 15, and this entire edifice rested upon its distinction between a person’s acts/conduct, and her identity.

Only, this distinction is deeply flawed, and NALSA v. UoI exposes the flaw in stark and glaring terms. When it comes to sex and sexual orientation, your “identity” means nothing if you can’t express it. A law that targets conduct, conduct that is the very expression of identity, thereby targets identity itself. When, therefore, S. 377 outlaws homosexuals from engaging in same-sex intercourse, it doesn’t just criminalise a set of acts – in outlawing the most basic expression of one’s sexuality, it criminalises sexuality – and thereby, identity – itself. As Justice Kennedy observed in Lawrence v. Texas:

“When sexuality finds overt expression in intimate conduct with another person, the conduct can be but one element in a personal bond that is more enduring. The liberty protected by the Constitution allows homosexual persons the right to make this choice.”

Similarly, in Elane Photography, the Supreme Court of New Mexico observed, just last year, that:

“… when a law prohibits discrimination on the basis of sexual orientation, that law similarly protects conduct that is inextricably tied to sexual orientation.”

Once the conduct/identity distinction collapses, the entire edifice upon which Koushal v. Naz was raised collapses along with it. The Court cannot now dodge either Article 14 or Article 15. It must, if it wants to uphold S. 377, provide a legitimate state purpose and a rational nexus for a law criminalising homosexuals as a class, and it must expressly engage with the Delhi High Court’s argument.

And lastly, as NALSA v. UoI shows, there is a further issue of Article 19(1)(a) to be considered (incidentally, 19(1)(a) was argued before the Delhi High Court in Naz, but the Court felt it sufficient to decide the case on grounds of 14, 15 and 21). To the extent that Article 19(1)(a) protects core expressions of our identity – including our sexual identity – as the Court holds today, it must necessarily protect homosexuals in expressing their identity. So even if the Court doesn’t wish to collapse conduct and identity – even if it wishes to hold the two to be separate – the logic of NALSA v. UoI leads inexorably to the conclusion that at the very least, in criminalising conduct, S. 377 criminalises the expression of homosexual identity, and therefore suffers from a 19(1)(a) problem.

In sum: if the conduct/identity distinction dissolves, S. 377 violates Article 14 and 15, because in criminalising conduct, it criminalises identity. If the conduct/identity distinction remains, S. 377 violates Article 19(1)(a), because it criminalises conduct that is the expression of identity. Either way, under the logic of NALSA, it is unconstitutional.

All this, of course, does not touch the Court’s holding that gender identity is “integral to the dignity of an individual and is at the core of “personal autonomy” and “self-determination”, and is therefore also protected by Article 21. (Para 74) This, as well, is inconsistent with the analysis in Koushal v. Naz.

The operative paragraph – with its directions – can be found at the end of the judgment, Paragraph 129. Today’s judgment is progressive in the best sense of the term, and is to be lauded. Equally important, however, today’s judgment is based upon reasoning that is fundamentally at odds with the reasoning in Koushal v. Naz. If NALSA is rightly decided, then Koushal is wrong. Surely, then, the time is now right to rehear Koushal before a Constitution Bench, redeem the promise of full moral membership for all persons, regardless of their sexual orientation, that the Delhi High Court affirmed so eloquently five years ago, and turn the page on one of the darker chapters in the Supreme Court’s civil rights history.






Filed under Non-discrimination, Sexuality

Net Neutrality, Free Speech and the Indian Constitution – I

(This piece first appeared on the CIS website, here)

Net neutrality is rapidly becoming one of the most important issues facing internet governance and internet freedom today, and it is quite likely that it will soon raise issues of law and legal policy in India. In this post (and the next), I will discuss net neutrality, free speech and the Indian Constitution.

I will not here go into the debates surrounding the multiple meanings of the concept of “net neutrality” but take, for the purposes of this post, the following definition:

The idea that all Internet traffic should be treated equally is known as network neutrality. In other words, no matter who uploads or downloads data, or what kind of data is involved, networks should treat all of those packets in the same manner.

In other words, put simply, net neutrality requires the extant gatekeepers of the internet – such as, for instance, broadband companies – to accord a form of equal and non-discriminatory treatment to all those who want to access the internet. Examples of possible discrimination – as the quote above illustrates – include, for instance, blocking content or providing differential internet speed (perhaps on the basis of a tiered system of payment for access).

Net neutrality has its proponents and opponents, and I do not have space here to address that dispute. This post – and the next – are premised on the assumption that net neutrality is both an important and a desirable goal (this brief article in the Times of India provides a decent, basic primer on the stakes involved).

An example of net neutrality in practice is the American Federal Communications Commission’s Open Internet Order of 2010, which was the subject of litigation in the recently concluded Verizon v. FCCThe Open Internet order imposed obligations of transparency, no blocking, and no unreasonable discrimination, upon internet service providers. The second and third requirements were vacated by a United States Court of Appeals. The rationale for the Court’s decision was that ISPs could not be equated, in law, to “common carriers”. A common carrier is an entity that offers to transport persons and/or goods in exchange for a fee (for example, shipping companies, or bus companies). A common carrier is licensed to be one, and often, one of the conditions for license is an obligation not to discriminate. That is, the common carrier cannot refuse to carry an individual who is willing and able to pay the requisite fees, in the absence of a compelling reason (for example, if the individual wishes the carrier to transport contraband). Proponents of net neutrality have long called for treating ISPs as common carriers, a proposition – as observed above – was rejected by the Court.

With this background, let us turn to India. In India, internet service providers are both state-owned (BSNL and MTNL), and privately-owned (Airtel, Spectranet, Reliance, Sify etc). Unlike many other countries, however, India has no network-neutrality laws. As this informative article observes:

The Telecom Regulatory Authority of India (TRAI), in its guidelines for issuing licences for providing Unified Access Service, promotes the principle of non-discrimination but does not enforce it… the Information Technology Act does not provide regulatory provisions relating to Internet access, and does not expressly prohibit an ISP from controlling the Internet to suit their business interests.”

In the absence of either legislation or regulation, there are two options. One, of course, is to invoke the rule of common carriers as a common law rule in court, should an ISP violate the principles of net neutrality. In this post (and the next), however, I would like to analyze net neutrality within a constitutional framework – in particular, within the framework of the constitutional guarantee of freedom of speech and expression.

In order to do so, two questions become important, and I shall address them in turn. First, given that most of the ISPs are privately owned, how does the Constitution even come into the picture? Our fundamental rights are enforceable vertically, that is, between individuals and the State, and not horizontally – that is, between two individuals, or two private parties. Where the Constitution intends to depart from this principle (for instance, Article 15(2)), it specifically and expressly states so. As far as Article 19 and the fundamental freedoms are concerned, however, it is clear that they do not admit of horizontal application.

Yet what, precisely, are we to understand by the term “State”? Consider Article 12: 

“In this part, unless the context otherwise requires, the State includes the Government and Parliament of India and the Government and the Legislature of each of the States and all local or other authorities within the territory of India or under the control of the Government of India.”

The key question is what, precisely, falls within the meaning of “other authorities”. The paradigmatic example – and this is something Ambedkar had in mind, as is evidenced by the Constituent Assembly Debates – is the statutory corporation – i.e., a company established under a statute. There are, however, more difficult cases, for instance, public-private partnerships of varying types. For the last fifty years, the Supreme Court has struggled with the issue of defining “other authorities” for the purposes of Part III of the Constitution, with the pendulum swinging wildly at times. In the case of Pradeep Kumar Biswas v. Indian Institute of Chemical Biology, a 2002 judgment by a Constitution bench, the Court settled upon the following definition:

The key question is what, precisely, falls within the meaning of “other authorities”. The paradigmatic example – and this is something Ambedkar had in mind, as is evidenced by the Constituent Assembly Debates – is the statutory corporation – i.e., a company established under a statute. There are, however, more difficult cases, for instance, public-private partnerships of varying types. For the last fifty years, the Supreme Court has struggled with the issue of defining “other authorities” for the purposes of Part III of the Constitution, with the pendulum swinging wildly at times. In the case of Pradeep Kumar Biswas v. Indian Institute of Chemical Biology, a 2002 judgment by a Constitution bench, the Court settled upon the following definition:

“The question in each case would be whether in the light of the cumulative facts as established, the body is financially, functionally and administratively dominated by or under the control of the Government. Such control must be particular to the body in question and must be pervasive. If this is found then the body is a State within Article 12. On the other hand, when the control is merely regulatory whether under statute or otherwise, it would not serve to make the body a State.”

Very obviously, this dooms the ISP argument. There is no way to argue that ISPs are under the pervasive financial, functional and administrative domination or control of the State. If we step back for a moment, though, the Pradeep Kumar Biswas test seems to be radically under-inclusive. Consider the following hypothetical: tomorrow, the government decides to privatize the nation’s water supply to private company X. Company X is the sole distributor of water in the country. On gaining control, it decides to cut off the water supply to all households populated by members of a certain religion. There seems something deeply wrong in the argument that there is no remedy under discrimination law against the conduct of the company.

The argument could take two forms. One could argue that there is a certain minimum baseline of State functions (ensuring reasonable access to public utilities, overall maintenance of communications, defence and so on). The baseline may vary depending on your personal political philosophy (education? Health? Infrastructure?), but within the baseline, as established, if a private entity performs a State function, it is assimilated to the State. One could also argue, however, that even if Part III isn’tdirectly applicable, certain functions are of a public nature, and attract public law obligations that are identical in content to fundamental rights obligations under Part III, although their source is not Part III.

To unpack this idea, consider Justice Mohan’s concurring opinion in Unnikrishnan v. State of Andhra Pradesh, a case that involved the constitutionality of high capitation fees charged by private educational institutions. One of the arguments raised against the educational institutions turned upon the applicability of Article 14’s guarantee of equality. The bench avoided the issue of whether Article 14 directly applied to private educational institutions by framing the issue as a question of the constitutionality of the legislation that regulated capitation fees. Justice Mohan, however, observed:

What is the nature of functions discharged by these institutions? They discharge a public duty. If a student desires to acquire a degree, for example, in medicine, he will have to route through a medical college. These medical colleges are the instruments to attain the qualification. If, therefore, what is discharged by the educational institution, is a public duty that requires… [it to] act fairly. In such a case, it will be subject to Article 14.

In light of Pradeep Kumar Biswas, it is obviously difficult to hold the direct application of the Constitution to private entities. We can take Justice Mohan, however, to be making a slightly different point: performing what are quintessentially public duties attract certain obligations that circumscribe the otherwise free action of private entities. The nature of the obligation itself depends upon the nature of the public act. Education, it would seem, is an activity that is characterized by open and non-discriminatory access. Consequently, even private educational institutions are required to abide by the norms of fairness articulated by Article 14, even though they may not, as a matter of constitutional law, be held in violation of the Article 14 that is found in the constitutional text. Again, the content of the obligation is the same, but its source (the constitutional text, as opposed to norms of public law) is different.

We have therefore established that in certain cases, it is possible to subject private entities performing public functions to constitutional norms without bringing them under Article 12’s definition of the State, and without the need for an enacted statute, or a set of regulations. In the next post, we shall explore in greater detail what this means, and how it might be relevant to ISPs and net neutrality.

Leave a comment

Filed under Free Speech, Internet Freedom

New Pages on the Blog: A Summary

Over the last few weeks, I’ve tried to expand the reach of the blog by adding some new pages. Most of these are under construction, and constantly being updated. Here is a quick summary:

The book reviews page brings together all the book reviews on this blog. So far, we’ve reviewed Agamben’s State of Exception, Waldron’s The Harm in Hate Speech, and Narrain & Suresh (eds.), The Judicial Nineties.

The Civil Liberties Database aims to provide a comparative constitutional law resource for civil liberties lawyers. It provides links to summaries and full texts of important civil liberties judgments from the United States, South Africa, Canada, the United Kingdom, the European Court, international tribunals (where applicable) and so on. The Database is classified by liberties (e.g., free speech, discrimination law etc.), and by sub-categories (pornography, hate speech, commercial speech etc.). The free speech section is the most detailed so far, and others are under construction.

Free Speech Watch aims to record free speech violations throughout India, as they happen.

The Regressive Laws Database page, which is very much in its infancy, aims to compile a list of particularly regressive Indian laws (such as the marital rape exception, S. 377 etc.), with brief explanations of how they violate essential constitutional guarantees, and progressive political thought more generally.

Lastly, Theorising Rights provides links to classical texts in political philosophy, that are the foundations of how we think about our civil rights and liberties. We have recently added some brilliant feminist pieces by Nancy Fraser, Catherine MacKinnon and Carole Pateman, Marx’s original critique of liberal rights in On the Jewish QuestionRonald Dworkin on liberalism, and two full-length books: Judith Butler’s Precarious Life (on national security and emergency powers), and Sheldon Wolin’s Democracy Incorporated (on democracy and the economic basis of society).

As mentioned above, all the pages are works-in-progress – so check back for continuing updates!

Leave a comment

Filed under Uncategorized

Reservations, Equality and the Constitution – X: Untidy Endnotes

Let us try to sum up the long and tortuous judicial history of reservations in India. Conceptually, I suggest that we can divide this history into three distinct phases. The first phase, that lasted from 1950 (Champakam Dorairajan to N.M. Thomas), can be called the “colour-blind phase”. The judges held that the vision of equality that the Constitution subscribes to is “colour-blind”, that is, it refuses to take any account of caste, race, sex, religion etc. in the distribution of benefits and burdens. Articles 16(4) and – after Champakam, 15(4) – were constitutionally mandated exceptions to the rule and, as such, were to be construed narrowly, being departures from the norm. Cases like Balaji, which imposed a 50% cap on reservations, best embodied this view of the Constitution.

In N.M. Thomas, the Court shifted to a group-subordination view of equality, according to which groupings such as caste must be taken into account to achieve genuine equality, since historically, these groups have been the site and the locus of sustained discrimination. In holding that Article 16(4) is not an exception to, but a facet of 16(1), the Court effectively held that 16(1) itself – and thus, the equality code as a whole – is committed to the group-subordination vision. As we have seen, it is unclear whether this view ever commanded a majority of the judges in a decision.

Post Indra Sawhney, in the 90s and the 2000s, the pendulum swung back towards the middle. Now the Court held that while 16(1) was about colour-blindness, 16(4) was about group subordination, and both were independent guarantees of equality that had to be balanced against each other. So 16(4) was no longer an exception to 16(1), but neither was 16(1) just a more abstract way of expressing 16(4) – rather, both constituted different elements of the overall 14-15-16 equality code.

This view of two conflicting vision of equality, to be held and balanced against each other, ultimately permitted Parliament a wide degree of latitude both in devising schemes of reservations and justifying them by amending the Constitution itself. This was done through the introduction of Articles 16(4A), 16(4B) and 15(5), the first two of which were expressly introduced to overcome adverse Court decisions. By constitutionalising the carry-forward rule, the catch-up rule and educational reservations for OBCs, Parliament repeatedly shifted the balance towards empowering identified groups, at the cost of formal individual equality. These amendments survived basic structure scrutiny because the Court, having already held that the Constitution was committed to both visions, could hardly forbid Parliament from shifting the balance this way or that. Presently, the “private, unaided educational institutions” aspect of 15(5) is under basic structure scrutiny. The outcome of the case will be interesting, because it will demonstrate how far the Court is willing to let Parliament go with modifying the existing balance that it first invoked in Indra Sawhney.

We have also seen that group-based reservations are justifiable on two distinct grounds. One theory holds that the ultimate goal of the society is to get to a point where group identity becomes irrelevant; but that point can be reached only by using group identity as a locus for remedial programs, aimed at eradicating a present inequality that has been caused by historical circumstances. This theory continues to value the individual as the ultimate subject of its solicitude, and views group-based action as a necessary evil. On the other hand, the opposite theory values the group for itself, and views reservations as, ultimately, designed to achieve equality not between individuals, but between groups.

The two theories, as we have seen, clashed in the Court’s last major reservation case, Ashoka Kumar Thakur, and the Court emphatically asserted that the Constitution was concerned, ultimately, with bringing about a casteless society. The judicially-crafted doctrine of the creamy layer demonstrates this commitment. The creamy layer denies reservations to those individuals belonging to identified beneficiary groups (on the basis of caste) who do not share the social and educational backwardness that has provided the ground for singling out that caste in the first place. Notice that if ultimate group equality was the goal of the Constitution, then the creamy layer doctrine would be entirely uncalled for.

This, however, raises the following interesting point. In Ashoka Kumar Thakur, the Court expressly rejected the creamy layer doctrine when it comes to Scheduled Tribes and Scheduled Castes. This suggests that within the group subordination vision of equality, both subsidiary visions – the one of an eventually casteless society, that seeks to dissolve groups, and the one of a society in which groups exist based on equality – are at play. Now, this conceivably makes sense in the case of Scheduled Tribes. Arguments from multiculturalism and pluralism provide a plausible justification for preserving and strengthening the identity of defined sub-groups with their own particular ways of life (although, admittedly, arguments by feminists focus on inter-se inequalities within such groups – see, e.g., Benhabib, The Claims of Culture).  But on what basis – if the aim is a casteless society – is there a distinction between Scheduled Castes and other castes? Do Scheduled Castes have a claim on group preservation that non-Scheduled Castes don’t? The Court, in refusing to extend the creamy layer doctrine to Scheduled Castes, cites Constitutional provisions that define the group – but in so doing, it creates a doctrinal and conceptual incoherence, where different standards of equality are applied to analytically identical groups.

The selective application of the creamy layer doctrine represents a deeper conceptual problem with the Court’s reservations jurisprudence. The Court has never clearly stated what the ultimate goal is; if the goals are multiple (substantive equality? casteless society? equivalent representation?), then how do they relate to one another? Most importantly, how does the methodology (such as the 21-point Mandal Commission) relate to – or even, help in identifying – the goals at stake? A failure to address and answer these questions is largely responsible for the profusion of constitutional amendments that have brought in highly concrete administrative regulations into the domain of constitutional law, as well as the conflicting signals that have repeatedly issued from the bench, over the last two decades.

One way of doing this, as was suggested in the last post, is to adopt a form of intermediate scrutiny in adjudicating reservation cases. If the ultimate goal is, indeed, a casteless society, then caste-based classifications that ground the distribution of benefits and burdens, are at least presumptively suspect. Because the Constitution expressly envisages group-based reservations, though, it is perhaps not appropriate to hold the government to the level of strict scrutiny. Intermediate scrutiny solves both problems. It requires the government to demonstrate a substantial interest (which, following Ashoka Kumar Thakur, would probably be the eventual achievement of a casteless society), and to then show that there is a reasonable fit between its scheme of reservations, and the ultimate goal. Indeed, as was suggested in the last post, the creamy layer – without the Court quite knowing it – has been serving as just this kind of intermediate scrutiny. Whether or not the Court expressly adopts intermediate scrutiny at any point, there is, at the very least, an urgent need for doctrinal and conceptual clarity in the reservations jurisprudence: what values are at stake, and what are the permissible means of achieving the values, that take into account individual rights as well as historic and continuing social discrimination? Perhaps the next basic structure case will be the case that achieves that.

Leave a comment

Filed under Equality, Reservations/Affirmative Action

Reservations, Equality and the Constitution – IX: Ashoka Kumar Thakur and Tiers of Scrutiny

The concurring opinions of Pasayat J. and Bhandari J. do not, in substance, depart from the holding in the Chief Justice’s opinion. Both opinions emphasize that reservations for socially and educationally backward classes must remove the creamy layer, if they are to be constitutional. This, as we discussed in our last post, shows that the ultimate subject of analysis, as far as SEBC equality goes, is not the backward class, but the individual. And indeed, Bhandari J. says as much in his opinion.

Two aspects of Bhandari J.’s opinion stand out. First, he severs “private, unaided institutions” from Article 15(5), holding that by a complete exclusion of Article 19(1)(g), subjecting private, unaided institutions to reservations violates the basic structure. Whatever the merits of this argument, it is important to note that no private institution appeared in this case, the question was not before the Court, and the issue was not argued. For a judge to rule on an issue that is not before the Court and is not argued by the lawyers, in an adversarial legal system, is judicial bad form.

Secondly, and more pertinently, Bhandari J. cites a raft of constituent assembly debates, as well a series of precedents, to argue that the ultimate goal of the Constitution is to achieve a casteless society, and the constitutionality of reservation policies must be judged by that standard. In so doing, he seeks to reject a caste-based identification of socially and educationally backward classes, and shift to an economic test, although he cannot do so because of the binding opinion of the nine-judge bench in Indra Sawhney.

The observations, however, are a good point of departure to examine the last remaining issue. As we discussed in the last post, it was argued before the Court that the American doctrine of strict scrutiny ought to be applied to reservations. As we saw, the Chief Justice rejected the proposition out of hand, on the ground that the American and Indian constitutions are drafted differently. This, however, is not reason enough in itself, because the American Constitution makes no mention of strict scrutiny – it is entirely a judicial creation. In order to understand the stakes, then, a brief digression into American doctrine would be appropriate.

The Fourteenth Amendment to the American Constitution, passed in the aftermath of the Civil War, guarantees to all citizens “the equal protection of laws”. The question then becomes: what constitutes constitutionally impermissible unequal treatment?

American courts answer this question by subjecting classifications to three different tiers of scrutiny, based on the manner of the classification in question. The first level is “rational review” scrutiny. A classification is permissible as long as the government can establish a goal with which it is rationally connected (notice the similarities with a standard Article 14 analysis).

The second level is “intermediate scrutiny”. This is primarily used in cases of sex discrimination. Intermediate scrutiny requires the government to show a “substantial interest”, that is “reasonably” served by the classification in question. In many of the sex equality cases, sex-based classifications have been invalidated because they have been found to be based on outmoded stereotypes about the role of men and women in society (such as, for instance, that women are required to be the primary care-givers) – the perpetuation of which is not taken to be a goal that the government has a substantial interest in.

The third level is “strict scrutiny”, which applies to “suspect classes”, such as race. A race-based classification, then, must demonstrate a compelling governmental interest, and must also be narrowly tailored - that is, it must be the least restrictive way of achieving the compelling interest at stake. There has been bitter controversy between the liberal and conservative wings of the American Supreme Court as to whether strict scrutiny ought to apply to race-based classifications that are remedial or beneficial in nature, such as affirmative action policies; as of now, because of a conservative majority on the Court, all race-based classifications, no matter what their aim, are subjected to strict scrutiny, and consequently, affirmative action programs have been severely cut-back, and quotas held unconstitutional altogether.

The basic idea behind heightened scrutiny is that certain classifications – such as those based on race – are inherently harmful because either they stigmatize individuals, or that they fail to respect individuals by treating them not as individuals, but as members of racial groups. They also come bearing a history of terrible repression and discrimination. Consequently, the government is held to a near-impossible standard when it aims to use that particular method of classification.

There are two points to be made here. The first is that in many ways, caste in India, and race in the United States, have had a relevantly similar history. In particular, they have been used as identifying labels to sort a group of people into a permanently subordinate category, excluding them from normal economic and social life, and placing them in a position of extreme inferiority. The site of discrimination has been the identifying label itself – a person’s race, or caste, marks them out as deserving or undeserving of equal concern and respect from society. This explains, then, why the framers were so concerned with the goal of achieving a casteless society – they understood that in order to eradicate discrimination, its most potent marker would also have to be eradicated.

Now, that doesn’t resolve the issue. As is repeatedly argued in the United States, to get beyond race, you need to take race into account. If, for centuries, black people were kept in conditions of extreme subordination, then ushering in a regime of formal equality will only mean that their subordination will continue, because they are in no position to compete on equal terms with people who have had a two-century head-start. Consequently, true equality will require the government to take race into account in crafting affirmative action and other socially beneficial programs that, in a sense, compensate for the historically-determined present-day subordinate position of a class that is only formally free and equal. Of course, it is crucial to craft such programs in such a way that classification on the basis of race is a tool to achieve a society in which race becomes entirely irrelevant to social life. That is to say, paradoxically enough, race-based programs are designed to eliminate race, and not to entrench it.

The analogy is now obvious. We aim at a casteless society, but also understand that to get there, we need to take caste into account in crafting present-day beneficial and remedial programs. Much of Bhandari J.’s judgment focuses precisely on this (and this perhaps explains why he is far more sympathetic to American doctrine than the other judges). His advocacy of an economic test to determine social and educational backwardness stems from his suspicion that we shall ever truly be able to get beyond caste if we use caste-based classifications to redistribute educational opportunities in society.

We can now see, with Bhandari J., that while strict scrutiny is perhaps an inappropriately high standard to hold the government to, given the inherent dangers in caste-based classifications, there is a good case to be made for an intermediate standard of review – one that requires the government to show something more than simply a rational connection between its classification and its goal, but something more. In other words, the government ought to be able to demonstrate how its caste-based identification of socially and educationally backward classes reasonably contributes to the goal of achieving a casteless society.

And indeed, the use of the creamy layer doctrine is an instantiation of this form of intermediate review. The creamy layer doctrine circumscribes the discretion of the government in selecting socially and educationally backward classes on the basis of homogenous categories such as caste. It requires the government to wean out those members of an identified caste who are not, as it turns out, socially and educationally backward. By doing so, the doctrine seeks to ensure that there really is a substantive connection between eventual castelessness, and the affirmative action programs being used to achieve that goal. Some form of such scrutiny also exists when the Court, for instance, holds that reservations cannot be used for posgraduate studies, or in technically demanding fields such as aviation.

Intermediate scrutiny, therefore, exists in our affirmative action doctrine, although we don’t call it that. If the creamy layer is one particular instantiation of intermediate scrutiny, though, there is no reason why the Court shouldn’t subject caste-based affirmative action programs to a more general and broad intermediate scrutiny. In other words, instead of according near-complete deference (barring creamy layer) to the government in its selection of SEBCs, the burden should lie on the government to demonstrate in a convincing way how its program – if it is caste-based – will avoid the peril of entrenching (and not removing) caste. Unlike strict scrutiny, the government would not be placed under the near-impossible burden of showing that its program is the narrowest possible way of achieving its goals, but it would still be required to show that there is reasonable connection.

Given its emphatic rejection of American doctrine in Ashoka Kumar Thakur,  it is unlikely that the Supreme Court would adopt this. As 15(5) is currently undergoing another basic structure challenge, however, it does provide an opportunity for the Court to do so, and require the government to more carefully circumscribe and shape its quota systems, than it is doing at the present.

Leave a comment

Filed under Equality, Reservations/Affirmative Action

Reservations, Equality and the Constitution – VIII: Ashoka Kumar Thakur, 15(5) and the Basic Structure

In the last post, we noted that in Nagaraj, a basic structure challenge to Articles 16(4A) and 16(4B) was rejected on the ground that the abstract concept of equality guaranteed by the basic structure was not “damaged or destroyed” by concrete tweaks to the mechanism of reservations. That was not, however, the end of basic structure challenges to affirmative action amendments. In 2005, via the 93rd Amendment, Article 15(5) was inserted into the Constitution. It read:

Nothing in this article or in sub-clause (g) of clause (1) of article 19 shall prevent the State from making any special provision, by law, for the advancement of any socially and educationally backward classes of citizens or for the Scheduled Castes or the Scheduled Tribes in so far as such special provisions relate to their admission to educational institutions including private educational institutions, whether aided or unaided by the State, other than the minority educational institutions referred to in clause (1) of article 30.

Brief background may be apposite. Textually, there is a degree of overlap between Articles 15(4) and 15(5). Recall that 15(4) permits the State to make special provisions for the advancement of socially and educationally backward classes, or for SCs/STs, notwithstanding anything in Article 29(2). 15(5) adds everything that follows after the “in so far as…”. The reference to Article 19(1)(g) is a response to the judgment in T.M.A. Pai Foundation, which held that the running of an educational institution falls within Article 19(1)(g)’s guarantee of the freedom of profession. In bringing private educational institutions within the ambit of reservations, the Amendment covers a possible 19(1)(g) challenge by excluding its application. And lastly, the Amendment excludes from its ambit, minority educational institutions, that have their own set of constitutionally guaranteed rights under Article 30.

Under the authority of Article 15(5), Parliament passed what came to be known as Act 5 of 2007. The Act laid out the specifics of reservations: simplifying greatly, it provided (subject to certain exceptions that need not concern us here), 15% reservation for SCs, 7% for STs, and 22% for OBCs (bringing it up to a total of 49%) per academic year. This was applicable to “Central Educational Institutions”, that is, institutions that had been set up under a statute, were financially aided by the government, or were in some other way under the control of the government. The Act also defined “other backward classes” as “socially and educationally backward classes of citizens, that were to be determined by the central government”.

The constitutionality of Act 5 was challenged, and Article 15(5) was challenged as a basic structure violation. The case was heard by a Constitution Bench, in Ashoka Kumar Thakur v. Union of IndiaA bewildering variety of arguments were made before the Court; although the Court unanimously upheld Article 15(5), four different opinions were issued, running into more than two hundred and fifty pages. Any analysis, therefore, will suffer from omission, but let us, in any event, start with the opinion of the Chief Justice.

Some of the major arguments attacking Article 15(5) focused on how it destroyed the scheme of equality running through Articles 14-15-16; how the use of caste to identify “other backward classes” was unconstitutional; how the exclusion of the creamy layer (following Indra Sawhney) was unconstitutional; and how the exclusion of Article 30 minority institutions also violated the principle of equality.

The Chief Justice first examined the basic structure argument. As a preliminary matter, since the challenge was in the context of Act 5, which was only applicable to State institutions, that part of Article 15(5) which referred to private, unaided educational institutions was excluded from the scope of the enquiry (it so happens that it is presently being litigated under another basic structure challenge). In other words, Ashoka Kumar Thakur was restricted to the question of whether Article 15(5), insofar as it applied to State institutions, violated the basic structure.

Much like in Nagaraj, the Chief Justice referred back to Kesavananda Bharati, and distinguished between the abstract concept of equality, and specific conceptions of it (although he termed them as “facets of equality”).  While an amendment that sought to damage or destroy the concept itself was void, one whose objective or effect was only that “the facets of the principle of equality… be altered…” was valid. Or:

“The larger principles of equality as stated in Article 14, 15 and 16 may be understood as an element of the “basic structure” of the Constitution and may not be subject to amendment, although, these provisions, intended to configure these rights in a particular way, may be changed within the constraints of the broader principle. The variability of changing conditions may necessitate the modifications in the structure and design of these rights, but the transient characters of formal arrangements must reflect the larger purpose and principles that are the continuous and unalterable thread of constitutional identity.” (Paragraph 93)

Clearly, then, what is at stake is the level of abstraction at which an amendment address and modifies equality. It would seem, from Nagaraj and from observations such as these, only an Amendment that brings about a state of affairs that could not be reasonably covered by any conception of equality, would fall foul of the basic structure. Since Article 15(5) was at best a “moderate abridgement or alteration” of the principle of equality, the basic structure challenge failed. (Paragraph 94)

Arguments that Articles 15(4) and 15(5) were contradictory, and that Article 15(5) unconstitutionally excluded minority institutions, were also swiftly rejected. That brought the Chief Justice to the constitutionality of Act 5 itself. It was argued, first, that the identification of socially and educationally backward classes had been done on the basis of caste, and that this was impermissible. After surveying a number of precedents on the issue (most of which have been discussed on this blog), the Chief Justice concluded: “”Caste” is often used interchangeably with “class” and can be called as the basic unit in social stratification.” (Paragraph 123) Nonetheless, after consulting various authorities on the subject, he went on to observe:

“A social class is therefore a homogeneous unit, from the point of view of status and mutual recognition; whereas a caste is a homogeneous unit from the point of view of common ancestry, religious rites and strict organizational control. Thus the manner in which the caste is closed both in the organizational and biological sense causes it to differ from social class. Moreover, its emphasis upon ritual and regulations pertaining to cleanliness and purity differs radically from the secular nature and informality of social class rules. In a social class, the exclusiveness would be based primarily on status. Social classes divide homogeneous populations into layers of prestige and esteem, and the members of each layer are able to circulate freely with it.” (Paragraph 137)

Specifically, the Chief Justice went on to hold that the identification of OBCs was done through a variety of socio-economic criteria, and thus wasn’t “solely” on the basis of caste. Or, to put it more simply, even if certain castes were inserted into the OBC category qua castes, they were a socially and educationally backward group, and not because they all were part of the same, defined caste. Caste, that is, to use the words of a previous case, was a “convenient shorthand” to identify the class.

The argument, however, is problematic, because it simply ignores the fact that there is a reason why Article 15(1) prohibits discrimination on the basis of caste. Think of the following hypothetical. A Constitution – like ours – prohibits discrimination on the basis of race, and has an enabling provision for the advancement of socially and educationally backward classes. The government establishes a quota for “blacks, Asians and Hispanics”, and argues that they’re all classes, and that they’re all backward – and that race is only a convenient shorthand for identifying them. This is a disingenuous argument. That is not to say, however, that that settles the issue: a prohibition against discrimination on the basis of race does not necessarily prohibit classification on the basis of race, motivated by benign objectives. That is the key question that the Chief Justice should have addressed. Instead, he dodged it, and held that although Other Backward Classes comprise of castes, yet, the castes are picked because they are classes, and not because they are castes. Again, this simply ignores the fact that the framers had something in mind when they wrote Article 15(1) into the Constitution.

The Chief Justice’s logic, however, led inexorably to his holding on the issue of the creamy layer. Excluding the creamy layer would make little sense if your analytical unit is the caste, because one’s caste identity exists as a matter of birth, independent of all other factors. If the unit is class, however, then the creamy layer – that is, persons belonging to an otherwise backward caste who are not themselves backward – do not form part of the backward class (which, by definition, must be backward), and therefore must be excluded. Hence, the Chief Justice observed:

It is to be understood that “creamy layer” principle is introduced merely to exclude a section of a particular caste on the ground that they are economically advanced or educationally forward. They are excluded because unless this segment of caste is excluded from that caste group, there cannot be proper identification of the backward class. If the “Creamy Layer” principle is not applied, it could easily be said that all the castes that have been included among the socially and educationally backward classes have been included exclusively on the basis of caste.” (Paragraph 149)

Or: Backward caste (unconstitutional selection) – creamy layer = backward class (constitutional selection).

The Chief Justice then observed:

By excluding those who have already attained economic well being or educational advancement, the special benefits provided under these clauses cannot be further extended to them and, if done so, it would be unreasonable, discriminatory or arbitrary, resulting in reverse discrimination. ” (Paragraph 150)

This casts further light upon the individual equality vs group subordination principle that we have discussed before. The Court seems to believe here (contrary to prior pronouncements) that notwithstanding the framing of Articles 15(4) and 15(5) as group rights, what is at stake is actually an individual right that one gains through contingent membership in a group.

The matter becomes somewhat complicated, however, because the Chief Justice went on to reject the operation of the creamy layer in the case of Scheduled Castes and Scheduled Tribes:

“Creamy layer” principle is one of the parameters to identify backward classes. Therefore, principally, the “creamy layer” principle cannot be applied to STs and SCs, as SCs and STs are separate classes by themselves.” (Paragraph 161)

This suggests – perhaps unsurprisingly – that as far as SCs and STs go, the principle of equality is indeed group-subordination: the historical site of injustice is the group, and it is the (undifferentiated, externally identified) group that is now the subject of affirmative action. This leads to the interesting conclusion that within the same constitutional provision, two different principles of equality (for SC/STs and OBCs) are at work.

The Chief Justice was also invited to incorporate the American affirmative doctrine of strict scrutiny, which he rejected. A full comparison would be beyond the scope of this post. We may note, in conclusion, Justice Raveendran’s brief concurrence, which corroborates the individual-centric analysis of the theoretical justification for OBC reservation: the ultimate objective, he observed, was a casteless society. (Paragraph 13) Caste was being taken into account for reservations only to hasten its ultimate eradication. The eradication of caste is an aim that is notably different from achieving an equality of castes. Likewise, the group subordination principle normally (though not exclusively) aims at equalising the relations between groups, and not necessarily eradicating group identity itself.


Leave a comment

Filed under Equality, Reservations/Affirmative Action