Tag Archives: strict scrutiny

Guest Post – Difficult Conversations: On why the Supreme Court’s judgment in the Kerala Liquor Ban Case represents a lost opportunity to examine tough questions on discrimination

(This is a guest post by Karan Lahiri and Vrinda Bhandari. A condensed version of this piece appeared previously on Scroll.)

As a bleak year drew to a close, the Supreme Court delivered another distinctly underwhelming judgment, in The Kerala Bar Hotels Association & Anr. v. State of Kerala & Ors.. This judgment is disappointing not so much for its outcome, but because of a glaring omission.

 This case was about whether the State of Kerala’s statutory amendment restricting the grant of FL-3 licenses (i.e. bar licenses that allow the sale of Indian Made Foreign Liquor or IMFL to the public) to Five Star Hotels was constitutional. Many may feel that in upholding this measure, the Court has justifiably approved the State Government’s legitimate attempt to curb alcoholism, which has statistically been shown to be an “acute social problem”.

The most important issue, however, is unrelated to the legitimacy of the Government’s ultimate objective, or whether there is a right to trade in liquor (which the Court dwells on at length). The key question relates to the exception that has been made in favour of (20 or so) Five Star hotels in the State, and, consequently, those who can access such hotels, and whether this amounts to discrimination based on wealth and social class, violating the right to equality guaranteed by Article 14 of the Constitution. In fact, this issue was specifically raised, and is recorded in Paragraph 17 of the judgment: –

The classification at hand is based on social and economic class, as there is a clear distinction between the expense and resultantly the clientele of the hotels that have been allowed FL-3 licenses and those that have not. Therefore, a strict scrutiny test must be applied, and the Government must be asked to provide a rigorous, detailed explanation in this classification… [W]hen discrimination is based on class, it is more pernicious and needs careful judicial enquiry.”

Unfortunately, this argument is not dealt with at all. Instead, the Court quickly accepted the explanation that this exception was directed at encouraging tourism. This omission represents a lost opportunity for the Supreme Court to re-examine fundamental questions about equality and discrimination.

One Size Fits All

Are there certain kinds of discrimination that are worse than others, demanding a more searching judicial examination? As of now, when a law differentiates or discriminates between two sets or “classes” of persons, the constitutionality of this line-drawing process is examined on the basis of a two-step test. First, a Court looks at whether there is an “intelligible differentia”. In other words, the line dividing two or more groups must be clear, based on discernible characteristics shared by members of each group. Second, it examines whether this “differentia” bears a rational relation to what the impugned law seeks to achieve.

The problem is that this is a one-size-fits-all test. All laws, essentially, differentiate between two or more groups of individuals. The Delhi Government’s odd-even formula to curb pollution differentiates between different groups (by exempting CNG vehicles, for instance), but it is, intuitively, different from a law that, let us say, penalizes begging on the road by the indigent. Can the same test be used for both sets of cases?

The “strict scrutiny test”, referred to in the judgment, is a concept that is making inroads into our jurisprudence, and is based on a concept we have borrowed from the U.S. When a law in the U.S. is tested on the anvil of equality, the default rule is that it will be upheld if it is rationally related to a legitimate government purpose. However, where the law makes a suspect classification (i.e. discrimination on the basis of race, religion, nationality etc., or analogous “discrete and insular minorities”), the more rigorous “strict scrutiny” test is applied. To survive strict scrutiny: –

  1. A law must be narrowly tailored to achieve a compelling government interest.
  2. It must be the least restrictive means for achieving that interest.

Notice that this test weighs in the importance of the State’s ultimate objective, and ensures that the means used to reach that objective are closely fitted to the ends, and that these are the least onerous means. Further, the fact that “suspect classifications” triggers strict scrutiny signals that the nature of the rights being affected by a discriminatory law are factored in, where some kinds of discrimination (i.e. on the basis of race, nationality etc.) are worse than others (e.g. a law exempting CNG vehicles from Delhi’s odd-even rule).

Just to clarify, this is not about whether the “strict scrutiny” test is the best test out there. What we are saying is that this was an opportunity for the Court to re-examine what goes into making a robust litmus test in equality cases.

Conversations on Equality and Class Discrimination

First, the Court should have begun a useful conversation on whether the dominant two-step test used in India is adequate to analyze all equality-related cases. As Dr. Tarunabh Khaitan points out, the test used in India is inadequate, because it is blind to the impact on, and the nature of the group being affected, as it does not “balance” the State’s interests against the rights of affected persons. Further, it does not look at the importance of the State interest involved, or whether there is a less onerous alternative. Putting it simply, as per the dominant two-step test, described above, if measure “X” is challenged on the basis that it is discriminatory, it is, simply put, based on whether it is reasonable and suitable to achieve objective “Y”, namely the Government’s end goal. The law in India, as it exists looks only at whether X and Y are reasonably connected, without looking at who measure X is impacting, or looking closely at the nature of objective Y.

Secondly, the Court should have entered into whether certain classifications are inherently “suspect”, justifying closer scrutiny compared to other more benign cases of differentiation. A good textual starting point would have been Article 15(1), which enjoins the State from discriminating on the basis of religion, race, caste, sex and place of birth. It could then have gone on to examine whether discriminating on the basis of wealth and social class is analogous.

We believe that this analysis, on discrimination based on social class, should have gone something like this.

  1. Given that both the Preamble and Article 38(2), have woven equality of status into the Constitution, there is a constitutional justification for treating discrimination based on status and social class as a “suspect classification”, warranting a higher degree of scrutiny, as opposed to cases where an underlying constitutional norm is not violated.
  2. Further, the Court ought to have looked at the additional question of whether the poor in India constitute what Dr. Khaitan calls a “vulnerable group”. In our opinion, poverty is about more that the lack of money or assets. It is characterized by a lack of social membership, meaningful citizenship, and dignity. Such a structural conception of poverty helps explain why the poor, as a separate class – independent of their caste or religious identities – are marginalized actors with little influence in the political process,[1] deserving of judicial protection by means of more rigorous standard of scrutiny.
  • Applying strict scrutiny standard, we believe the Kerala amendment ought to have been struck down. This is because even a universal ban on public drinking would have been a more narrowly tailored solution to the State’s objective of reducing alcoholism, since alcoholism afflicts rich and poor alike.

We must remember here that the Kerala amendment is about more than the price of alcohol. It is fundamentally different from a law raising the MRP of liquor, or imposing a tax on liquor across the board. What makes the Kerala amendment “pernicious” is that it singles out only those with the means to purchase liquor in a Five Star hotel, as also the status to enter and access a Five Star hotel, thereby giving them the exclusive privilege of drinking in public. At the same time, poorer sections of the population, who do not have the status to access such elite establishments (let alone afford the prices), are deprived of this privilege. Surely, the rich are not immune to alcoholism? However, this is almost exactly what the Court seems to suggest, when it says – “There is also little scope for cavil that the guests in Five Star hotels are of a mature age; they do not visit these hotels with the sole purpose of consuming alcohol.” The Supreme Court has, previously, in the Maharashtra dance-bars case, frowned on such logic, stating: –

 “Our judicial conscience would not permit us to presume that the class to which an individual or the audience belongs brings with him as a necessary concomitant a particular kind of morality or decency.”

The Supreme Court itself is not blind to fact that the Five Star hotel rating implies both a higher price and higher status for access. In dealing with allegations that Five Star hotels have opened out some of their premises for consumption of liquor at depressed rates in less salubrious surroundings, the Court encourages the State to end such “malpractice”, because, according to it, Five Stars are “violating the ambiance which they portray by enabling drinking in specially created bars at lower prices.”

What we have outlined above is what we believe to be the correct line of reasoning. We believe that gradualism and experimentation cannot be used as arguments to defer to legislative judgment (as the Court has done in this case), when groups are singled out based on social class and status. You may not agree either with the reasoning or the result we have proposed. That’s what makes Constitutional Law exciting – the fact that it inspires healthy and vibrant debate. Unfortunately, our Supreme Court has chosen to steer clear of such debate – a strange choice for an institution in a democratic polity, given that conversation and debate lie at the heart of democracy.

Karan Lahiri and Vrinda Bhandari are both practicing lawyers. Karan Lahiri assisted in representing one set of Appellants in the Kerala liquor ban case before the Supreme Court.

[1] There is empirical evidence to suggest that the poor, India, rarely come together as a class in a way that makes democratic institutions respond to their preferences. This ties is with the writings of certain American Constitutional scholars, who speak of the poor as an “anonymous and diffuse” group that cannot organize itself effectively.

Blog Editor’s Note: Previously, I’ve made the case for a higher threshold of Article 14 scrutiny in election cases here (unfortunately, we know how that turned out!), and Mihir wrote a guest post examining how the arbitrariness doctrine serves an alternative to the unsatisfactory one-size-fits-all nature of the rational review standard. To the persuasive arguments made in this post, I have a quick addition: a higher threshold of scrutiny would insist not only on the showing of a compelling State interest and a narrowly-tailored law, but would also place the evidentiary burden upon the State to demonstrate the connection. For instance, in this case, instead of making the State’s case for it by the blanket observation that people going to 5-Star Hotels are “more mature”, the Court would require the State to bring hard evidence forward showing that drinkers in 5-Star Hotels are less prone to be alcoholics, or indulge in violence, or something of that sort. Needless to say, it’s rather unlikely that there would be evidence of this sort forthcoming, and the law would have to fail. The purpose of this imposing this evidentiary burden is precisely to forestall the State from relying upon gross and invidious stereotypes about people (“uneducated persons can’t distinguish between right and wrong”, “5-Star alcohol drinkers are mature”) which, in itself, is a profound denial of the principle of equal concern and respect.



Filed under Article 14, Equality

Some thoughts on Article 14 Post-DSPE

Previously on this blog, Manish has criticised the Supreme Court’s opinion in Subramanian Swamy v. CBI, which was handed down earlier this month (I have done the same here). To recap: S. 6A of the Delhi Police Special Establishment Act (“DSPE”) required governmental sanction for investigation of offences under the Prevention of Corruption Act, in case the allegations concerned senior government officials (Joint Secretary or above). The Court struck down this provision on Article 14 grounds, holding that different rules for different classes of officials with respect to investigating corruption were unconstitutional, but – as I’ve tried to demonstrate – got its reasoning hopelessly entangled. The case, however, provides us with an opportunity to reflect upon the current state of Article 14 jurisprudence.

One particularly worrying trend – exemplified in the DSPE case – is the increasing reliance that the Court is placing on the “legitimate state interest” prong of its Article 14 analysis. Traditionally, Article 14 scrutiny of governmental classification consisted of two prongs: an intelligible differentia between the classes that the challenged law had created, and a rational nexus with a governmental purpose. In cases like the Delhi High Court’s Naz Foundation judgment, and now in DSPE, the Court has also required that the State purpose itself be legitimate, and this requirement has played a pivotal role in the outcomes of these cases.

This, however, is extremely problematic. Who judges whether a State purpose is legitimate or not? Traditionally, the Court has the power to strike down laws enacted by a democratically elected legislature under two circumstances: first, if the legislature lacks the competence to pass that law (for example, if it is a State legislature passing a law upon a central subject, like defence); and secondly, if the law violates a fundamental right. In other words, the touchstone for legal legitimacy is the Constitution itself – which lays down rules of competence, and enshrines fundamental rights. As long as those two requirements are satisfied, it is not for the Court to examine the merits of the law, its wisdom or its desirability, or its efficiency – that is within the domain of the parliament, and the remedy for bad laws lies at the ballot box. Lack of competence and violation of fundamental rights, however, are independent grounds for striking down legislation – so on what basis then is the Court creating this extra criterion of legitimate state purpose, which is neither related to competence nor to fundamental rights, as part of its Article 14 analysis?

It is easy to see why the Court feels the need to bring in this analytical construct: a traditional Article 14 enquiry is very easy to satisfy. In most cases, the government will be able to construct some purpose that can bear a rational connection with the classification (in the DSPE case, for instance, the government’s stated purpose was administrative efficiency). Consequently, the legitimate purpose prong is one way for the Court to ensure that Article 14 doesn’t become almost entirely toothless. Yet this is not the right way to go about things, because it essentially transforms the Courts into super-legislatures, passing judgment on the “legitimacy” of laws, based upon standards of its own creation.

There is, however, another ways in which Article 14 can be given some bite. The first is to create a tiered structure of scrutiny, based on the model adopted by the United States Supreme Court. In the US, equal protection violations are judged on three levels: rational review, intermediate scrutiny and strict scrutiny. Rational review – like a standard Article 14 analysis – requires only a “rational connection” between the classification and the purpose. In certain cases, however, which involve “fundamental rights” (for example, if the classification is based on race), the Court requires the government to show a compelling interest, and also that the challenged legislation is the narrowest method of achieving that interest (strict scrutiny has been used to strike down affirmative action policies). Government almost never succeeds under this standard. In between rational review and strict scrutiny, there is “intermediate scrutiny”, which the Court uses in its sex-discrimination cases. Intermediate scrutiny requires the government to show a substantial interest (in between compelling and any interest), and a reasonable connection (in between rational and narrowly tailored) between the law and the purpose. The government’s success rate, unsurprisingly, lies somewhere in between rational review (almost always successful) and strict scrutiny (almost never successful).

The key question, of course, is to determine which tier of scrutiny will apply to which set of cases. Judgments of institutional competence and the gravity of the interests involved play a part in this determination, which is ultimately the task of the judiciary. For instance, adjudication upon the merits of economic policy will – given concerns of institutional competence – necessarily involve the weakest form of review (rational review). The DSPE case, however, involved setting up two different legal procedures for different classes of persons, involving fundamental rule of law concerns (as the Court itself acknowledged). Given that the classification raised rule of law concerns, there is a clear argument in favour of ratcheting up the level of scrutiny to intermediate or strict. Under intermediate scrutiny, for example, the Court would have required the government to come up with evidence to demonstrate that administrative efficiency was actually served by the classification in question (as opposed to taking the government’s claims on face value, as it would do in a rational review case).

This leads us to another important point: tiers of scrutiny also involve questions of burdens and standards of proof. Under rational review standards, the Court will not itself look into whether the government has shown beyond doubt that is classification actually serves its stated purpose. As we move up the levels, however, given the importance of the interests involved, the government has a heavier burden of making such a demonstration to the satisfaction of the Court. The basic idea here is to prevent the government from invoking legislative purposes as a colourable method for screening otherwise illegitimate classifications. So, for example, in the DSPE case, the idea behind requiring the government to show – by evidence – that there is a connection between its classification and administrative efficiency, is to prevent the efficiency argument from acting as a screen, shielding corrupt high officials from investigation.

The tiered scrutiny framework does away with the problems of the legitimate-purpose enquiry. Of course, a full version of the argument will need to be developed, over time, by the judiciary. The present state of the law, however, is untenable: “legitimate purpose” is an entirely free-floating concept that will, in the end, become a plaything of individual judges. Tiered scrutiny, I suggest, is one coherent method by which to bring a degree of order to Article 14, while performing the same function that legitimate purpose does – prevent the government from doing an end run around its constitutional obligations by coming up with false or colourable purposes.

Leave a comment

Filed under Article 14, Equality

Surveillance and Privacy in India – II: Gobind and the Compelling State Interest test

After its judgment in Kharak Singh, the Court was not concerned with the privacy question for a while. The next case that dealt – peripherally – with the issue came eleven years later. In R.M. Malkani v State of Maharashtra, the Court held that attaching a recording device to a person’s telephone did not violate S. 25 of the Telegraph Act, because “where a person talking on the telephone allows another person to record it or to hear it, it can-not be said that the other person who is allowed to do so is damaging, removing, tampering, touching machinery battery line or post for intercepting or acquainting himself with the contents of any message. There was no element of coercion or compulsion in attaching the tape recorder to the telephone.” Although this case was primarily about the admissibility of evidence, the Court also took time out to consider – and reject – a privacy-based Article 21 argument, holding that:

“Article 21 was invoked by submitting that the privacy of the appellant’s conversation was invaded. Article 21 contemplates procedure established by law with regard to deprivation of life or personal liberty. The telephonic conversation of an innocent citizen will be protected by Courts against wrongful or high handed interference by tapping the conversation. The protection is not for the guilty citizen against the efforts of the police to vindicate the law and prevent corruption of public servants. It must not be understood that the Courts will tolerate safeguards for the protection of the citizen to be imperiled by permitting the police to proceed by unlawful or irregular methods.

     Apart from the fact that it joined Kharak Singh in refusing to expressly find a privacy right within the contours of Article 21, there is something else that unites Kharak Singh and R.M. Malkani: they hypothetical in Kharak Singh became a reality in Malkani – what saved the telephone tapping precisely because it was directed at “… a guilty person”, with the Court specifically holding that the laws were not for targeting innocent people. Once again, then, the targeted  and specific nature of interception became a crucial – and in this case, a decisive – factor. One year later, in another search and seizure case, Pooran Mal v Inspector, the Court cited M.P. Sharma and stuck to its guns, refusing to incorporate the Fourth Amendment into Indian Constitutional law.

It is Gobind v State of MP, decided in 1975, that marks the watershed moment for Indian privacy law in the Constitution. Like Kharak Singh, Gobind also involved domiciliary visits to the house of a history-sheeter. Unlike Kharak Singh, however, in Gobind the Court found that the Regulations did have statutory backing – S. 46(2)(c) of the Police Act, which allowed State Government to make notifications giving effect to the provisions of the Act, one of which was the prevention of commission of offences. The surveillance provisions in the impugned regulations, according to the Court, were indeed for the purpose of preventing offences, since they were specifically aimed at repeat offenders. To that extent, then, the Court found that there existed a valid “law” for the purposes of Articles 19 and 21.

By this time, of course, American constitutional law had moved forward significantly from eleven years ago, when Kharak Singh had been decided. The Court was able to invoke Griswold v Connecticut and Roe v Wade, both of which had found a “privacy” as an “interstitial” or “penumbral” right in the American Constitution – that is, not reducible to any one provision, but implicit in a number of separate provisions taken together. The Court ran together a number of American authorities, referred to Locke and Kant, to dignity, to liberty and to autonomy, and ended by holding, somewhat confusingly:

the right to privacy must encompass and protect the personal intimacies of the home, the family marriage, motherhood, procreation and child rearing. This catalogue approach to the question is obviously not as instructive as it does not give analytical picture of that distinctive characteristics of the right of privacy. Perhaps, the only suggestion that can be offered as unifying principle underlying the concept has been the assertion that a claimed right must be a fundamental right implicit in the concept of ordered liberty… there are two possible theories for protecting privacy of home. The first is that activities in the home harm others only to the extent that they cause offence resulting from the mere thought that individuals might he engaging in such activities and that such ‘harm’ is not Constitutionally protective by the state. The second is that individuals need a place of sanctuary where they can be free from societal control. The importance of such a sanctuary is that individuals can drop the mask, desist for a while from projecting on the world the image they want to be accepted as themselves, an image that may reflect the values of their peers rather than the realities of their natures… the right to privacy in any event will necessarily have to go through a process of case-by-case development.

But if no clear principle emerges out of the Court’s elucidation of the right, it was fairly unambiguous in stressing the importance of the right itself. Interestingly, it grounded the right within the context of the freedom struggle. “Our founding fathers,” it observed, “were thoroughly opposed to a Police Raj even as our history of the struggle for freedom has borne eloquent testimony to it.” (Para 30) The parallels to the American Fourth Amendment are striking here: in his historical analysis Akhil Amar tells us that the Fourth Amendment was meant precisely to avoid the various abuses of unreasonable searches and seizures that were common in England at the time.

The parallels with the United States become even more pronounced, however, when the Court examined the grounds for limiting the right to privacy. “Assuming that the fundamental rights explicitly guaranteed to a citizen have penumbral zones and that the right to privacy is itself a fundamental right, that fundamental right must be subject to restriction on the basis of compelling public interest.” “Compelling public interest” is an interesting phrase, for two reasons. First, “public interest” is a ground for fundamental rights restrictions under Article 19 (see, e.g., Article 19(6)), but the text of the Article 19 restrictions do not use – and the Court, in interpreting them, has not held – that the public interest must be “compelling”. This suggests a stricter standard of review for an Article 21 privacy right violation than Article 19 violations. This is buttressed by the fact that in the same paragraph, the Court ended by observing: “even if it be assumed that Article 19(5) [restrictions upon the freedom of movement] does not apply in terms, as the right to privacy of movement cannot be absolute, a law imposing reasonable restriction upon it for compelling interest of State must be upheld as valid.” The Court echoes the language of 19(5), and adds the word “compelling”. This surely cannot be an oversight.

More importantly – the compelling State interest is an American test, used often in equal protection cases and cases of discrimination, where “suspect classes” (such as race) are at issue. Because of the importance of the right at issue, the compelling state interest test goes hand-in-hand with another test: narrow tailoring. Narrow tailoring places a burden upon the State to demonstrate that its restriction is tailored in a manner that infringes the right as narrowest manner that is possible to achieve its goals. The statement of the rule may be found in the American Supreme Court case of Grutter v Bollinger:

Even in the limited circumstance when drawing racial distinctions is permissible to further a compelling state interest, government is still constrained under equal protection clause in how it may pursue that end: the means chosen to accomplish the government’s asserted purpose must be specifically and narrowly framed to accomplish that purpose.”

 To take an extremely trivial example that will illustrate the point: the State wants to ban hate speech against Dalits. It passes legislation that bans “all speech that disrespects Dalits.” This is not narrowly tailored, because while all hate speech against Dalits necessarily disrespects them, all speech that disrespects Dalits is not necessarily hate speech. It was possible for the government to pass legislation banning only hate speech against Dalits, one that would have infringed upon free speech more narrowly than the “disrespect law”, and still achieved its goals. The law is not narrowly tailored.

Crucially, then, the Court in Gobind seemed to implicitly accept the narrow-tailoring flip side of the compelling state interest coin. On the constitutionality of the Police Regulations itself, it upheld their constitutionality by reading them narrowly. Here is what the Court said:

Regulation 855, in our view, empowers surveillance only of persons against whom reasonable materials exist to induce the opinion that they show a determination, to lead a life of crime – crime in this context being confined to such as involve public peace or security only and if they are dangerous security risks. Mere convictions in criminal cases where nothing gravely imperiling safety of society cannot be regarded as warranting surveillance under this Regulation. Similarly, domiciliary visits and picketing by the police should be reduced to the clearest cases of danger to community security and not routine follow-up at the end of a conviction or release from prison or at the whim of a police officer.”

    But Regulation 855 did not refer to the gravity of the crime at all. Thus, the Court was able to uphold its constitutionality only by narrowing its scope in a manner that the State’s objective of securing public safety was met in a way that minimally infringed the right to privacy.

Therefore, whether the Gobind bench was aware of it or not, its holding incorporates into Indian constitutional law and the right to privacy, not just the compelling State interest test, but narrow tailoring as well. The implications for the CMS are obvious. Because with narrow tailoring, the State must demonstrate that bulk surveillance of all individuals, whether guilty or innocent, suspected of crimes or not suspected of crimes (whether reasonably or otherwise), possessing a past criminal record or not, speaking to each other of breaking up the government or breaking up a relationship – every bit of data must be collected to achieve the goal of maintaining public security, and that nothing narrower will suffice. Can the State demonstrate this? I do not think it can, but at the very least, it should be made to do so in open Court.

In the next post, we shall continue with the development of privacy law post-Gobind.


Filed under Privacy, Surveillance