The Supreme Court’s Recusal Order: Glaring Conceptual Flaws


, ,

[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

Previously on this blog, I had written (here and here) about the problematic composition of the Constitution Bench in the ongoing case involving the interpretation of the Land Acquisition Act. As readers know, parties before the Court had argued last week asking for the recusal of Justice Arun Mishra from the bench. Orders had been reserved after hearing, and a judgment was delivered yesterday, refusing recusal. Now, given that five-judge bench of the Supreme Court had already transformed the proceedings from a legal hearing into theatrics, with Justice Mishra remarking (without contradiction) that it would be “the blackest chapter in history”* if the bench would “succumb”, it is unclear why the bench needed to undertake the farcical exercise of reserving an order than writing out a judgment over the weekend, when minds had apparently been made up at the hearing itself. However, the fact is that a judgment has now been delivered, and must therefore be analysed.

As a prelude to the analysis, I had noted in my previous post that various accusations had been made in court that a “lobby” was writing articles about recusal in order to malign the Chief Justice and the Supreme Court. Unfortunately, these accusations have now found their way into a recorded judgment. At paragraph 11 of the judgment, Justice Mishra notes the Solicitor-General’s submission that ” a tendency is growing in that as soon as important matters are listed, particular articles are written in the newspapers concerning the Constitution of the bench or to influence the decision on merits of a case. Newspaper articles are written to influence Court. The very independence of the judicial system is at stake…” In paragraph 36, he writes that “affronts, jibes, and consciously planned snubs should not deter us from discharging our onerous responsibility (emphasis supplied) In paragraph 39, he writes that “in Bal Kishan Giri v. State of Uttar Pradesh, (2014) 7 SCC 280, this Court has considered derogatory remarks and efforts to destroy the system.” And in their concurring opinion, four other learned Justices write that they agree with Justice Mishra’s “reasoning and conclusion.” One also notes that during oral arguments, Justices Mishra and Saran had made similar remarks, and none of the other judges on the bench had disagreed.

To this one can only say: if indeed there is a “lobby” engaging in “consciously planned snubs” to “destroy the system”, then this is a very serious matter indeed. Nobody can be permitted to “destroy the system.” Attempts to do so would constitute contempt of court. Now presumably, evidence for the existence and actions of this “lobby” exists, because five learned judges of the Supreme Court would surely not make or endorse such claims without evidence. That being the case, the problem would swiftly be resolved by the issuance of a formal notice of contempt to these unnamed “lobbyists”, so that the evidence for their “lobbying” could be exposed in a public hearing.

Now, on to the law: the facts leading up to these proceedings have been set out in my two previous posts, and I will not repeat them here. The key conceptual error that Justice Arun Mishra’s recusal judgment makes can be found in paragraph 11:

The first question before us is whether a Judge who has expressed an opinion in a smaller Bench and the case has been referred to a larger Bench, because of the conflict of the opinion or otherwise, can hear the matter in a larger bench.

It is almost trite to say that the answer you get depends upon the question you ask. The framing of the issue conflates three distinct situations that are, nonetheless, distinct in critical ways. The first situation is when, after the passage of time, a court is prompted to reconsider an earlier verdict that it had delivered. The second is a situation where a smaller bench, on a prima facie examination of binding precedent, makes a “referral” to a larger bench to reconsider it. And the third is a situation where within the same court, there are contrary judgments existing at the same time that require to be settled one way or another. These three situations – as I have argued in my previous posts – also need to be understood within a broader institutional framework: does the court normally sit en banc (all the judges sitting together), and therefore – for all practical purposes – speak at all times within one institutional voice, or does it (like our court) sit in multiple different panels of judges? Note that the first two situations outlined above could occur within either institutional form, while the third is limited to a court (like ours) where thirty-four judges divide themselves into numerous panels of two or three.

The problem with Justice Mishra’s framing, then, is that all his examples that follow after paragraph 11 correspond to either situation A or situation B. For example, his first example – the overruling of the five-judge bench decision in United Motors (1953) by a seven-judge bench in Bengal Immunity (1955) is an example of situation A. At that time, the Supreme Court had eight judges. An issue was decided one way in 1953, and then decided another way in 1955, with two of the judges who had sat in the first case also sitting in the second; in the first case, a majority of the court participated in the judgment, whereas in the second case, it was effectively an en banc judgment. This, therefore, was an example of the Court as an institution changing its mind about the correctness of its earlier ruling. Similarly, his second example – that of the two-judge bench order in Ujagar Prints, which doubted the correctness of the judgment in Empire Industries – is one of referral: there is existing precedent, another bench – on a prima facie examination – proposes a second look at the issue, and the question is referred to a larger bench for resolution. Justice Mishra points out that Justice Sabyasachi Mukherji, who was party to the judgment in Empire Industries, was also on the five-judge bench that reheard the issue. The issue there, however, was that Empire Industries held the field as good law: a referral does not displace or set up contrary precedent that needs to be adjudicated upon. In that sense, therefore, examples under situation B come within the same broad framework as situation A – that of an institution, acting as an institution, reviewing one of its earlier decisions.

A number of examples follow that – as discussed above – fall either within situation A or situation B. In each of these examples, Justice Mishra’s analysis is limited to pointing out that there were a few common judges in both the earlier decision, and the later decision. This is evident from his observation in paragraph 21, where he notes:

This Court has observed that law should be settled permanently and that it should be settled correctly. There may be circumstances where public interest demands that the previous decision be reviewed and reconsidered. Thus, it is apparent that this is the consistent practice of this Court that Judges who had rendered the earlier decision have presided over or been part of the larger Bench.

It is telling that in all of Justice Mishra’s list of relevant precedent, which ends at paragraph 22, there is not one case that is actually on all fours with the present situation: a situation where at the same time there existed two conflicting judgments delivered by different judges, sitting in benches of equal strength, and where the latter had declared the former to be per incuriam. In my previous posts, I have explained why this distinction matters fundamentally – an explanation that I will get back to a little later.

After paragraph 22, Justice Mishra cites a range of judgments on judicial bias that have no application to the present issue. It is unclear, for example, what relevance an American judgment stating the same judge should complete a trial despite showing signs of predisposition through the its course has to do with the controversy here – or another judgment that states that it is “desirable to have the same judge in successive causes.” In any event, after having cited these cases, Justice Mishra then comes to the substantive part of his judgment. In paragraph 27, he notes that:

There may not be even one Judge in this Court who has not taken a view one way or the other concerning Section 24 of the Act of 2013, either in this Court or in the High Court. If the submission is accepted, no Judge will have the power to decide such a matter on the judicial side. We have to deal with the cases every day in which similar or somewhat different questions are involved concerning the same provision. For having taken a view once, if recusal is to be made, it would be very difficult to get a Judge to hear and decide a question of law. We have to correct the decision, apply the law, independently interpret the provisions as per the fact situation of the case which may not be germane in the earlier matter. A judgment is not a halting­place, it is stepping stone. It is not like a holy book which cannot be amended or corrected. It may also work to the advantage of all concerned if a Judge having decided the matter either way is also a member of the larger bench. A Judge who had rendered any decision in a smaller combination is not disqualified from being part of a larger Bench when a reference is made to the larger bench. Rather, it is a consistent practice prevailing in various High Courts as well as of this Court to include the same Judge/Judges in larger Benches.

But it should now be clear that this is a mischaracterisation of the issue, because the dispute is not about whether a judge has “taken a view” on what Section 24 of the Land Acquisition Act says, or the parade of horribles that Justice Mishra trots out in this paragraph, and the ones that follow: bench-hunting, forum-shopping, etc. etc. The issue is a very narrow and simple one, that deserves to be spelt out once more: this is not a case where the institution is reviewing its earlier judgment, or deciding on a referral where a judge – on a prima facie reading – calls for a reconsideration. This is a case where a three-judge bench ruled one way on Section 24. Then another three-judge bench – headed by Justice Mishra and over the dissent of Justice Shantanagouder – held it to be per incuriam and set up its own contrary reading of Section 24. Note that Justice Mishra in that judgment did not simply express doubt about prior precedent and refer it for a fuller reconsideration (which is the normal practice). He overruled that judgment by effectively holding that it had no force in law. Not only did this go against established principles of stare decisis and judicial discipline, where you are not supposed to overrule a judgment that is binding upon you, but it also set up two conflicting lines of precedent within the same court at the same time – and it was that precise issue that the third three-judge bench was considering the day after Justice Mishra’s ruling, when the question was referred (by Justice Mishra) to the Chief Justice. It is telling that in a sixty-two page long judgment, Justice Mishra is unable to provide a single precedent – from India or from abroad – that has similar facts, and in his reasoning, he makes no reference whatsoever to this crucial point.

And the distinction matters. It matters in a polyvocal Court of thirty-four judges, where small panels hear issues on the same legal point, and return conflicting findings. All of Justice Mishra’s precedents and arguments are based upon the vision of the court acting as a unified institution, with internal mechanisms for reviewing and rethinking its previous decisions. However, you cannot eat your cake and have it too: a unified institution does not have a situation where oppositional lines of precedent are generated at the same time, and require further resolution between them. As someone pointed out: this situation is akin to Examiner A failing a student, Examiner B passing him, and the answer-script being sent back to Examiner A to “resolve” the conflict. Yes, Examiner A might be persuaded to change his mind. But that, it should be obvious, is hardly the point. It is for this reason that I had noted in the previous post that:

In India, however, we have a situation where within the highest judicial body, the existence of a multiplicity of judicial panels undermines institutional coherence, and creates a situation where the apex Court is effectively disagreeing with itself. This is what has happened in the present case: abstracting for a moment from the thicker context, what has happened is that two three-judge benches of the Supreme Court have taken diametrically opposite views on the same issue. Now the existence of thirty-four judges on the Supreme Court means that there is a ready solution to hand: send the issue to a bench that has a higher number of judges, in order to “resolve” the conflict.

Shorn of the legalese, what this effectively means is that within the highest judicial body, there is an internal appellate mechanism to deal with the problem of institutional incoherence, flowing from the Court’s unique structure. I use the word “appellate” in its ordinary sense; it is, obviously, not an “appeal” as that word is defined under Indian law, but it is basically a sui generis response to a situation where even within the apex Court, there are situations when conflicting views require a resolution in the interests of institutional coherence.

Once we understand this, however, it becomes clear that if the same judge is going to be present at both stages of the process, then this form of resolution becomes pointless. This is why it is important to understand Mr. Shyam Divan’s argument that what this effectively amounts to is a judge adjudicating upon the correctness of his own judgment in “collateral” proceedings; it is not formally an appeal, but in every significant respect, these latter proceedings are doing the work of an appeal. The source of the confusion is that – for understandable reasons – we continue to think of the Supreme Court as a unified body that speaks in one institutional voice, while the reality has moved very far away from this. A more accurate analogy would be with the European Court of Human Rights, where the same Court is divided into a “Chamber” and a “Grand Chamber.” Chamber judges and Grand Chamber judges are drawn from the same overall pool of ECHR judges: “The Grand Chamber is made up of the Court’s President and Vice-Presidents, the Section Presidents and the national judge, together with other judges selected by drawing of lots.” But: “When it hears a case on referral, it does not include any judges who previously sat in the Chamber which first examined the case.” (Emphasis Supplied)

Needless to say, this point finds no mention in Justice Mishra’s judgment. It also finds no mention in a brief concurring opinion penned by the other four judges – Benerjee, Saran, Shah and Bhat JJ – who repeat the same fallacy when they observe in paragraph 5 that: “We notice that his order has cited several previous instances where judges who rendered decisions in smaller bench compositions, also participated in larger bench formations when the reasoning (in such previous decisions) was doubted, and the issue referred to larger benches, for authoritative pronouncement.”

A final point needs to be noted. As Suhrith Parthasarathy noted at the time, there was a crucial issue of judicial discipline and respect for stare decisis in the manner in which Justice Mishra delivered the Indore Development Authority judgment. Holding a binding three-judge bench judgment to be per incuriam, effectively overruling it, and also overruling at one stroke seventeen Supreme Court judgments that had followed it – instead of doing things the normal way and making a referral – all raised serious questions about the functioning of a polyvocal court. It was precisely those issues that the three-judge bench intended to consider when this judgment was brought to its attention the next day – a process that was stalled when the case was referred to the Chief Justice. The recusal order makes it clear that those issues now stand buried. If henceforth, however, it is open to coordinate benches to overrule each other through declarations of per incuriam – and then for the Chief Justice acting as the Master of the Roster to in effect assign the case to judges who have been on one side through his powers of bench composition – then what we are looking at in the future is a factionalised court, where judicial decisions fall hostage to internal power struggles and bench-formation imperatives. There is indeed a potential parade of horribles in store – but it has nothing to do with bench hunting and forum shopping, and everything to do with the institutional integrity of the Supreme Court.


*”Blackest chapter”? Blacker than ADM Jabalpur? Blacker than Koushal v Naz? There is a lot of blackness down that particular road, and such overwrought hyperbole from the bench does nobody any favours – least of all “the system.”

The Bombay High Court on Illegal Surveillance



In a judgment handed down earlier today, a two-judge bench of the Bombay High Court found that the Central Bureau of Investigation’s [“CBI”] surveillance of a businessman accused of paying bribes was illegal. The Court quashed the surveillance orders, and directed destruction of the call records. The judgment is important for three reasons: (i) it is an model of principled constitutional adjudication, following both the letter of the law and applicable precedent; (ii) it extends the logic of the Supreme Court’s privacy judgment [“Puttaswamy”] in the context of unconstitutionally obtained evidence; and (iii) it provides a genuine and enforceable remedy upon a finding that fundamental rights have been breached.

Principled Constitutional Adjudication

The facts of the case were simple. There were three surveillance orders (29th October 2009, 18th December 2009, and 24th February, 2010), which constituted the legal bases for the interception of the petitioner’s phone calls. The petitioner argued that these orders were both ultra vires the Telegraph Act, as well as the Puttaswamy judgment. As the call records had been used in the charge-sheet against the petitioner, it was also argued that those records should be destroyed, and could not be part of the case against him.

The Bombay High Court (Jamdar and More JJs) went straight to the text of the Telegraph Act. Justice More noted that Section 5(2) of the Act made it clear that telephone interception was permissible only under one of two circumstances: the existence of a public emergency, or a threat to public safety. This was obviously not a public emergency, so the key phrase was “public safety.” More J. then referred to the judgment of the Supreme Court in PUCL v Union of India (1997), where “public safety” had been clearly defined as “the state or condition of freedom from danger or risk for the people at large… [that would be] apparent to a reasonable person.” He then noted that the PUCL judgment had also set out procedural safeguards in cases of surveillance, that were to be treated as enforceable rules under Section 5 of the Telegraph Act.

More J. went on to observe that the judgment in PUCL had been endorsed by the nine-judge bench of the Supreme Court in Puttaswamy. This also meant that contrary precedent in R.M. Malkani v State of Maharashtra – that had held that telephone interception did not violate any fundamental rights – now stood overruled. Puttaswamy had also held that any restriction upon the right to privacy had to satisfy the proportionality standard. More J. then held that the directions in PUCL conformed to the proportionality standard, and were evidently binding upon the Bombay High Court.

Applying the law to the facts, More J. found that it was abundantly clear that there was no threat to “public safety”, following PUCL’s definition. Therefore, “it was impermissible to take resort to telephone tapping.” (para 16) He noted that “even at this stage, from the affidavits filed by the Respondents or the charge-sheet, the Respondents could not justify any ingredients of risk to the people at large or interest of the public safety.” (para 17) Consequently:

We are satisfied that in peculiar fact of the instant case, the impugned three interception orders neither have sanction of law nor issued for legitimate aim, as sought to be suggested. The impugned three interception orders could not satisfy the test of “Principles of proportionality and legitimacy” as laid down by the nine judges’ constitution bench decision in K. T. Puttaswamy (supra). We, therefore, have no hesitation in holding that all three impugned orders are liable to be set aside. Accordingly, we quash and set aside the same. (para 19)

Now this may look like a straightforward application of law – and it is – but there are two points that I want to highlight here. The first is the unambiguous understanding of the legal standard laid out in Puttaswamy, and its application. Puttaswamy makes clear that in cases where fundamental rights are infringed, it is not enough for the government to cite law and order, and for the Court to engage in an undefined balancing exercise (that somehow ends up being resolved in favour of the government). Rather, Puttaswamy’s proportionality standard is a four-step test that requires, inter alia, government to show that there is a legitimate aim, that the infringing action is rationally connected to that aim, that it is necessary (i.e., the least restrictive option to achieve that aim), and that it is proportionate (i.e., the benefits outweigh the costs of infringing rights). The Bombay High Court makes it clear that the government cannot avoid justifying its actions under that standard.

Secondly, in recent times, we have seen far too often that courts have not only given such overriding importance to the government’s stated goals (“public interest”, “national interest”, “national security” etc.), but have also let the government define what the content of these phrases is. This has been most prominently visible in the Kashmir Cases (discussed on this blog), where the invocation of “national security” has acted as an impenetrable shield against any further judicial scrutiny. But it has also been visible elsewhere – in the courts’ recent jurisprudence on anticipatory bail, for example, or in its jurisprudence on bans upon organisations, bail under the UAPA, the concept of “custodial interrogation”, and of course the Supreme Court’s recent use of Article 142 to provide legal cover to the compulsory taking of voice samples.

In short, courts have increasingly become fixated on the governmental imperatives of law and order, treating procedural rights and safeguards as nuisances that need to be swept aside in order to ensure that “public interest” is satisfied. In that context, the Bombay High Court’s decision to straightforwardly apply the law as well as constitutional precedent, leading to the inescapable conclusion that surveillance was illegal, is most refreshing. Notice that the Court could have decided the other way as well: it could have held, for instance, that corruption is so grave a problem that adequately prosecuting it is indeed in the “interests of public safety.” This, of course, would have twisted the meaning of “public safety” out of all recognition – but it is not a reasoning technique that we are unfamiliar with (recall, for example, the Supreme Court’s interpretation that “migration” fell within the meaning of “external aggression”, in Sarbananda Sonowal’s Case).

Instead, the Bombay High Court’s judgment is a reminder that the “balance” between governmental goals and individual rights has already been struck in the text of statutes (providing procedural safeguards) and in constitutional decisions that lay out doctrines like proportionality. The job of a court now is to straightforwardly apply text and precedent, and stand by the legal conclusion that follows from that application.

The Remedy

Equally important is the Court’s remedy. The Bombay High Court noted that in PUCL, it had been made clear that if the Review Committee found that telephone interception was illegal, copies of the intercepted material would have to be destroyed. This had also been adopted by Rule 419(A)(17) of the Telegraph Rules. Consequently, More J. noted:

… having held that the impugned interception orders have been issued in contravention of the provisions of section 5(2) of the Act, we have no option but to further direct the destruction of intercepted messages. (paragraph 21)

The Court also pointed out that the CBI itself had been taking conflicting stands in court about the legal foundation of the orders, and the actions of the Review Committee – a practice that the Court strongly deprecated (paragraph 28). More J. found, in addition, that the successive interception orders were carbon copies of each other, and had evidently been passed without application of mind – a clear breach of the statute.

Unconstitutionally Obtained Evidence

The Respondents argued, however, that even if there had been illegality in the collection of the evidence, they should be entitled to use it in the course of the criminal prosecution. They relied upon several judgments that had held that as long as evidence was relevant, it could be introduced in a trial, regardless of the legality of how it was obtained.

The Court’s response to this argument is perhaps the most remarkable part of the judgment. After distinguishing the relevant precedent (including the judgment in Pooran Mal, More J held:

We may also add here that if the directions of the Apex Court in PUCL’case (supra) which are now re-enforced and approved by the Apex Court in K. T. Puttaswamy (supra) as also the mandatory rules in regard to the illegally intercepted messages pursuant to an order having no sanction of law, are permitted to be flouted, we may be breeding contempt for law, that too in matters involving infraction of fundamental right of privacy under Article 21 the Constitution of India. To declare that dehorse the fundamental rights, in the administration of criminal law, the ends would justify the means would amount to declaring the Government authorities may violate any directions of the Supreme Court or mandatory statutory rules in order to secure evidence against the citizens. It would lead to manifest arbitrariness and would promote the scant regard to the procedure and fundamental rights of the citizens, and law laid down by the Apex Court. (paragraph 39)

This is a hugely important holding. I have argued elsewhere that in Selvi v State of Karnataka, a three-judge bench of the Supreme Court had drawn a crucial distinction between illegally obtained evidence (admissible if relevant) and unconstitutionally obtained evidence (inadmissible under all circumstances). Evidence obtained in breach of fundamental rights, in other words, could not then be used in court against the citizen. Although the Bombay High Court did not cite Selvi, it did cite Puttaswamy (which endorsed Selvi), and more importantly, it provided a strong rationale for this: that to hold on the one hand that a certain method of collecting evidence was unconstitutional, while also allowing the State to use the evidence so collected, would not only be arbitrary, but would also ensure that fundamental rights and individual safeguards remained parchment barriers against the State.


In each of its three conclusions – on illegality, on the remedy, and on evidence – the Bombay High Court, as I have shown, engaged in a textbook application of law and precedent. It faithfully applied existing standards balancing individual procedural safeguards against the imperatives of law and order, found that under those standards surveillance was clearly illegal, and then went through with the consequences that followed. What makes it remarkable, of course, is the number of recent judgments that appear to have given a go-by to these standards by invoking the imperatives of the State.

It is also remarkable, I think, for another reason: the Puttaswamy judgment was hailed as a new dawn for constitutionalism (Constitutionalism 3.0, as someone commented) precisely because of its categorical assertion that the individual was at the heart of the constitutional order, that infringements of individual rights must be limited to the strict confines of what was constitutionally permitted, and must be justified as such by the State. Puttaswamy was not just a simple finding that privacy was a fundamental right under the Constitution, but the reasoning of the nine judges brought hope that it would lay the foundation for transforming how civil rights adjudication happens: adjudication that would stop treating – as K.G. Kannabiran once said – the “restrictions” as fundamental and the rights as contingent, and that would stop providing automatic normative priority to the reasons of State. And at the time of Puttaswamy, I had written that:

But we must all be equally clear about the fact that the real task will begin now: it will begin with the first bench that is asked to apply Puttaswamy to a concrete case where privacy runs up against reasons of State, and it will continue in the months, years, and decades to come. The task is not simply to apply Puttaswamy, but to use Puttaswamy to craft a genuinely progressive civil rights jurisprudence, where the original constitutional compact – that individual rights are not subordinate to “public good”, “social good”, “public interest” (or any other variant of the phrase) – is restored. And that, now, is the responsibility of citizens, of lawyers, and of course, of the judges who will be called upon to adjudicate privacy and liberty claims in the wake of this judgment. For judges, indeed, it is a challenge: to be true to the animating spirit of Puttaswamy, and make the hard decision to tell the State that although its aim may be laudable, its motives unimpeachable, and its method beneficial, under the Constitution of India, it nonetheless cannot have what it wants.

I think it is fair to say that the last few months have severely tested this cautious optimism, expressed more than two years ago. From that perspective, the Bombay High Court judgment was beautiful to read, because it indicated the difference that Puttaswamy can and could make, and it provided a glimpse of Puttaswamy’s promise to initiate a transformative civil rights jurisprudence, a promise that had so enchanted us at the time.

The enchantment may have worn off, but the hope remains!


Guest Post: The Abortion Petition – Some Key Questions


, , ,

(This is a guest post by Gauri Pillai.)

This post examines the recent writ petition filed in the Supreme Court of India, challenging the constitutionality of certain provisions of the Medical Termination of Pregnancy Act, 1971 (‘Act’). The Act permits termination up to a period of 20 weeks. It requires the approval of one medical professionals for termination prior to 12 weeks, and two medical professionals for termination between 12 and 20 weeks. Termination is permitted only when continuation of pregnancy would cause grave injury to the physical or mental health of the pregnant woman, or where there is substantial risk that the fetus, if born, would suffer from such physical or mental abnormalities so as to be seriously handicapped (Section 3(2)). If the pregnancy is caused due to rape, or failure of contraceptive device used by a ‘married woman or her husband’, it is presumed that there is grave injury to the mental health of the woman (Explanation, Section 3(2)). Beyond 20 weeks, the Act permits termination only if ‘immediately necessary to save the life’ of the pregnant woman (Section 5).

The Challenge 

The Act, the petition argues, imposes a severe restriction on women’s right to ‘reproductive choice’, that is, ‘the right to choose whether to conceive and carry pregnancy to its full term or to terminate, [which lies] at the core of one’s privacy, dignity, personal autonomy, bodily integrity, self-determination and right to health, recognised by Article 21 of the Constitution’. This is because termination, even within the first trimester, is permitted only upon fulfilling certain conditions, set out above; termination on account of fetal abnormality is allowed only up to 20 weeks; and finally, termination beyond 20 weeks is restricted to instances ‘immediately necessary’ to save the ‘life’ of the pregnant woman.

By critiquing these restrictions for being excessive, harsh and disproportionate, the petition mounts a challenge to the Act under Article 21. However, I argue, the petition does not interrogate, and dislodge, the assumptions that lie behind these restrictions. Instead, it takes these assumptions as given and works within them, to argue for more expansive rights.

The Presumption of Motherhood

The Act is premised on two fundamental assumptions.

First, the Act views women primarily as mothers, and pregnancy as natural and inevitable. This is indicated by provisions of the Act which allow abortions only under exceptional, adverse circumstances, suggesting that the default option for women is to continue with their pregnancy. It could be argued that women’s right to an abortion is restricted to exceptional situations in order to balance the interests of the woman against interests of the fetus. If so, the present framing of the Act implies that in exceptional, adverse circumstances, the interests of the woman take precedence over that of the fetus. However, in the absence of such circumstances, the interests of the fetus override that of the woman.

This framing assumes one of two things (or both): (a) in the absence of exceptional circumstances, women are happy to continue their pregnancy; and/or (b) in the absence of exceptional circumstances, women should be expected to continue their pregnancy. In case of (a), it is assumed that there would be no harm to women’s interests because women, under ordinary circumstances, would want to continue their pregnancy (and be mothers). For instance, in the landmark decision of Suchitra Srivastava v Chandigarh Administration, the Supreme Court observed, ‘the termination of pregnancy has never been recognised as a normal recourse for expecting mothers’. Since there is no harm to women’s interests, fetal interests are given priority. In case of (b), it is assumed that the harm to women’s interests from continuation is lesser than the harm to fetal interests from termination. This is premised on the assumption that even if the individual woman does not desire the pregnancy, pregnancy is natural and inevitable, something all women go through. Therefore, in the absence of exceptional circumstances, the harm from continuation of an unwanted pregnancy is minimal. As a result, in these circumstances, the harm to women’s interests from continuation of pregnancy is considered to be lesser than the harm to fetal interests from termination. Thus, the Act starts from a position where women are seen first and foremost as mothers, and pregnancy as natural, inevitable, and desired by all women. This assumption then influences the assessment of harm and balancing of interests carried out by the Act. This is not to say that no other considerations influence the balancing, but to highlight that the current framing of the Act suggests that presumptions regarding women’s role as mothers is one such consideration.

The petition appears to resist this narrative by asking for abortion on demand within the first trimester. However, a close reading of the petition reveals that the reason behind this claim is not the recognition that it is ‘normal’ for women not to be mothers, and opt for termination. Instead, the claim is made because the health risk to women from continuation of pregnancy is more than the health risk accompanying termination during the first trimester. Thus, even in asking for abortion on demand within the first trimester, the petition fails to dislodge the gendered assumption underlying the Act which views women as mothers, and pregnancy as natural and inevitable; instead, it merely works to expand the instances of termination permitted within this narrative.

A Strand in a Gendered Web

Second, the Act views abortion as an ordinary medical procedure, with no larger social import. This is obvious in the very title of the Act, which refers to ‘medical termination of pregnancy’, rather than ‘abortion’; the Statement of Reasons or Objects of the Act which describes abortions as ‘health measures’ to alleviate ‘danger to the life or risk to the physical or mental health of the woman’, and prevent ‘wastage’ of her life; the predominance given by the Act to the opinion of medical professionals, viewing them as co-decisionmakers along with the women; and, the reliance placed by courts on the decisions of Medical Boards—set up to weigh the risk of continuation of pregnancy against the medical risk of termination—in allowing abortions beyond 20 weeks, under Section 5 (see here, here and here). Women’s decisions to undergo an abortion are overridden if the Medical Board opines that the continuation of pregnancy is ‘less hazardous’ than termination at that stage (see here). Thus, the medical risk of termination becomes the primary consideration while making a decision under the Act.

This tendency to prioritise the medical risk of termination over other considerations, including the woman’s choice, is evident in the petition as well. For instance, the petition constructs the right under Article 21 in the following terms: ‘Where the termination of pregnancy itself does not involve risk to the physical life of the woman, her right to choice…[has] to be respected’. Similarly, as mentioned above, the petition argues for abortion on demand within the first trimester only because ‘there is no dispute [that] the risks involved when pregnancy is carried to full term far outweigh the minimal and negligible risks involved when pregnancy is terminated in the first trimester. Keeping this in view, the State cannot make any law restricting the right of the woman seeking abortion’ during the first trimester. In this sense, the petition follows the Act in viewing abortion primarily as a medical procedure, by respecting women’s choice only when medical opinion favours it.

However, this understanding of abortion is reductive, and ignores that at the centre of abortion lies a group of persons—women—who have been historically oppressed on account of their reproductive ability. The fact that women can reproduce is translated into the essentialist, universal assumption that women must. These gendered expectations, however, do not end at the moment of birth. Women are not only expected to bear children but also assume primary responsibility for their care. This largely unilateral responsibility of child-care has confined women to the domestic sphere, and maintained the distinction, and the hierarchy, between the public and private spheres. Motherhood and paid employment are constructed as incompatible. Several empirical studies demonstrate that the presence of young children in the house is associated with lower female workforce participation in India, with women quitting work after childbirth (see here and here). At the same time, childcare—women’s contribution in the home—has a low status in society. Though ‘reproduction entails incredible liabilities and workload, [it] is still considered to be of lesser value compared to men’s engagement in production that yields market value’. In this sense, pregnancy is not an isolated nine-month episode in the life of a woman; it is instead located within gendered structures of power, which require women to bear children and raise them, compelling them to sacrifice other opportunities they value, while, at the same time, devaluing their labour. Articles 14 and 15 provide constitutional grounding to this perspective on pregnancy. However, the role of these provisions is outside the scope of the specific argument being made by this post, and is thus not developed.

Against this context, abortion cannot be viewed as just another medical procedure. Instead, it should be seen as allowing women to take back control over their bodies and lives—control they have been historically denied, and on the basis of which they have been disadvantaged. This recognition is absent within the petition, which argues for permitting women to undergo termination only as long as there is no threat to their life or health. In this manner, the petition places abortion on the same page as other medical procedures, where the opinion of the medical professional is given overriding priority. Through this, women seeking abortions are seen merely as patients seeking medical care, stripping away the gendered social context of reproduction in India.

Thus, the Act is deficient because it is premised on two underlying assumptions: women as mothers, and abortion as an ordinary medical procedure. The petition, in challenging the Act as restrictive and unconstitutional, does not dislodge these assumptions, and shift abortion outside these narratives. Instead, it retains the framing offered by these assumptions, and therefore constructs a limited right to medical termination of pregnancy. This might be an intentional strategy, to ensure that the claim has a greater chance of being accepted by the court. However, it is important to question whether in making this concession and failing to interrogate these assumptions, the petition, even if granted, will result in real and effective reproductive control for women.

The Remedy

The remedies sought by the petition include striking down as unconstitutional Section 3(2)(a) to the extent it makes termination during the first trimester conditional; Section 3(2)(b) to the extent that it imposes a 20 week limit on termination in case of grave mental or physical injury to the woman, or fetal abnormality; and, Section 5 to the extent that termination beyond 20 weeks is permitted only when ‘immediately necessary’ to save the ‘life’ of the pregnant woman. If the approach I propose is adopted, and the fundamental assumptions underlying the Act are challenged, certain remedies—such as abortion on demand within the first trimester—would remain the same. Others could, however, differ. If the balancing exercise does not start with viewing women as mothers, and pregnancy as natural and inevitable, then the balance struck between the woman’s and fetal interests could change. Similarly, if abortion is not viewed as an ordinary medical procedure, a case could be made for why the choice of a woman who wishes to undergo termination even after being informed of the risk to health or life needs to be respected. These would change the very structure of the Act, by questioning the imposition of conditions and time limits. However, even if the specific remedies do not change, by challenging the assumptions underlying the Act, the petition, if accepted, could shift the discourse on abortion, and set us on the path towards meaningful reproductive choice for women.

Kashmir: Fundamental Rights and Sealed Covers


, , , ,

In a previous post, we discussed one of the peculiar features of the ongoing litigation regarding the communications shut-down and other restrictions in Kashmir. One of these features is the absence – in court – of the government’s orders that constitute the basis for the restrictions (whether under the Telecom Suspension Rules or Section 144 of the CrPC). As we discussed, one of the basic requirements for a restriction upon fundamental rights is the existence of a law, and its publication (i.e., the law being made available to the citizens whose freedoms it seeks to restrict). There can be no restriction of fundamental rights in the absence of law, or on the basis of secret laws.

In the hearing of 16th October, this question was (finally) put to the State by the Supreme Court Justices. It is reported that Solicitor-General Tushar Mehta stated that he had no objection to showing the orders to the Court, but considerations of national security may require him to withhold them from the petitioners – and that the petitioners had no “right” to claim access to the orders. Accordingly, the Court’s Order records that if the Solicitor-General wants to claim “privilege” over the orders, then the Court “requests him to file an affidavit indicating the reasons for claiming such privilege.” 

While we wait for the government’s affidavit, it is important to note that what is at stake here is a creeping expansion of the “sealed cover”, which we have seen so often in recent times. It is also important to note that it is entirely unjustifiable: executive orders – passed under cover of law – restricting rights of citizens are not and cannot be subject to legal privilege, or submitted to the Court in a sealed cover. 

At one level, it is questionable whether a legal order revealed only to the Court, and hidden from citizen, counts as “publication” in the relevant sense. But there is a more basic reason why this is unconstitutional. If I – as a citizen – do not have access to the legal order that purports to restrict my rights, I have no effective way of challenging it in Court and demonstrating it to be unconstitutional. I cannot show that it is disproportionate and fails the reasonableness standard under Article 19(2). What this means, in turn, is that effectively, that I have no remedy to enforce my fundamental rights. And a right without a remedy is, of course meaningless. 

Effectively, therefore, denying the order on the basis of which rights are infringed amounts to a suspension of the rights themselves. As explained in a previous post, this can only be done – and that too, partially – through a formal declaration of Emergency; in other words, the government’s arguments are entirely based upon the logic of an Emergency, without the courtesy of a formal declaration of Emergency. 

It is important to remember the last time the contrary argument was made. The last time it was made – unsurprisingly – was in ADM Jabalpur. There, the argument made on behalf of the detenus was that the suspension of the right to move the Court to enforce habeas corpus amounted to the denudation of Article 21 itself. To this, the Court said that the mere fact that you could not move the Court did not mean that the rights ceased to exist – it just meant that you had no way of enforcing them. But if ADM Jablpur has been buried “ten fathoms deep” by Puttaswamy, then surely there is no remaining scope for the Government, in 2019, to make this argument.

Consequently, therefore, no “affidavit” can justify keeping the communications suspension and S. 144 orders secret, and there is no justification for handing them over to the Court in a “sealed cover”. They must be made disclosed, and not only to the petitioners, but to the general public, so that affected parties are in a position to seek remedies before the courts of law. Any other outcome would only amount to a justification of the logic of Emergency.   

Addendum: The Concept of “Bias” in a Polyvocal Court


, , ,

[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

It is important to begin this post with a clarification. Yesterday, I wrote a blog post pointing out that the composition of the Constitution Bench scheduled to hear the case concerning the interpretation of S. 24 of the Land Acquisition Act raised some important issues, specifically concerning the powers of the Chief Justice as the “Master of the Roster.” The post was one among many public interventions that raised similar questions about the Bench. During the hearings of the case today, it is reported that Justice Mishra observed that “there are attempts in the media and social media to “malign” the institution by raising questions about the bench and CJI’s decision.” Solicitor-General Tushar Mehta agreed, and stated that “there is a pattern here, where a few days before a major case is heard, things are written on social media or web magazines raising certain objections, that are intended to influence the proceedings.” Another judge on the bench, Justice Vineet Saran, agreed with him.

As two Justices of the Supreme Court, and the Government’s second-highest ranking Law Officer have seen fit to hold forth from the bully pulpit against nameless “critics”, a few points need to be made in response, in addition to what has already been pointed out by Nitin Sethi. The first is that the composition of the Bench was made public this Saturday. The hearing was scheduled for today. Unless the Solicitor-General believes that critics of the Court had divined the composition of the Bench in advance, and were only keeping their powder dry until it was formally announced, the only time that these objections could possibly have been raised was between Saturday and Tuesday, i.e., “a few days before the hearing.” Secondly, the Supreme Court is a public institution in a democratic republic, that is committed to the principle of open justice. That the Court’s conduct will be subjected to rigorous public scrutiny is exactly how it should be. Critics of the Court put their names to what they write, take responsibility for their writing, and set out their arguments (along with the grounds on which they are based) openly and in the public domain, which is exactly how it is supposed to happen in a democracy. The critics are, in fact, considerably more transparent than the subject of discussion, which is the allocation of benches by the Chief Justice – a process that is entirely opaque and discretionary. And thirdly, the only pattern that is visible here is Supreme Court judges’ reaction to public scrutiny by shielding themselves in a cloak of defensiveness and self-righteousness, where we go straight from criticism to a “maligning” of the institution (and, by extension, an equation of individual judges with “the institution”). Readers will recall that this was exactly the reaction at the time of sexual harassment allegations against the Chief Justice – a reaction that, again, was supported by the highest law officers of the government.

With these preliminary points out of the way, I want to focus on one specific issue that was raised during oral arguments today: the issue of deciding bias. The issue arose out of the petitioners’ request that Justice Arun Mishra recuse himself from hearing the case. Much of the questioning on this point was led by Justice S. Ravindra Bhat; in particular, drawing on precedent from common law countries, Justice Bhat made a distinction between “pecuniary bias”, “personal bias”, and “intellectual bias”, and questioned whether “intellectual bias” alone was sufficient ground for a recusal; he also observed that in Indore Development Authority, Justice Mishra had only expressed a “view”, and could – in theory – be persuaded to change his mind. Senior Counsel Shyam Divan’s response to these questions can be read here, but in this post, I want to make a separate point: questions of bias and recusal cannot be adjudicated without paying close attention to the unique nature of the Indian Supreme Court as an apex Court: that is, its polyvocal character. To directly copy standards of bias that have been evolved in common law jurisdictions – as Justice Bhat’s line of questioning suggests – without paying attention to the Indian Supreme Court’s institutional character, is to essentially be making a category mistake.

Almost uniquely among the major English-speaking Apex Courts of the world, there exists in the Indian Supreme Court a massive numerical difference between the strength of the Court (34 judges) and the strength of benches hearing day to day cases (2 or 3 judges). The US Supreme Court, for example, sits en banc (as a full Court) of nine Justices. The South African Constitutional Court sits en banc for the most part as well. The UK Supreme Court has twelve judges, that often sit in panels of five (but can also sit en banc or close to it for important cases, such as the recent prorogation judgment). Similar situations exist for the apex courts of Kenya, Canada, New Zealand, Australia. What this means is that these apex Courts possess an institutional coherence: judgments can be fairly said to express a “view of the Court” (as an institution). And if you want a judgment of the apex Court to be set aside, you have to – effectively – convince the same body to go back on its earlier view.

In India, however, we have a situation where within the highest judicial body, the existence of a multiplicity of judicial panels undermines institutional coherence, and creates a situation where the apex Court is effectively disagreeing with itself. This is what has happened in the present case: abstracting for a moment from the thicker context, what has happened is that two three-judge benches of the Supreme Court have taken diametrically opposite views on the same issue. Now the existence of thirty-four judges on the Supreme Court means that there is a ready solution to hand: send the issue to a bench that has a higher number of judges, in order to “resolve” the conflict.

Shorn of the legalese, what this effectively means is that within the highest judicial body, there is an internal appellate mechanism to deal with the problem of institutional incoherence, flowing from the Court’s unique structure. I use the word “appellate” in its ordinary sense; it is, obviously, not an “appeal” as that word is defined under Indian law, but it is basically a sui generis response to a situation where even within the apex Court, there are situations when conflicting views require a resolution in the interests of institutional coherence.

Once we understand this, however, it becomes clear that if the same judge is going to be present at both stages of the process, then this form of resolution becomes pointless. This is why it is important to understand Mr. Shyam Divan’s argument that what this effectively amounts to is a judge adjudicating upon the correctness of his own judgment in “collateral” proceedings; it is not formally an appeal, but in every significantrespect, these latter proceedings are doing the work of an appeal. The source of the confusion is that – for understandable reasons – we continue to think of the Supreme Court as a unified body that speaks in one institutional voice, while the reality has moved very far away from this. A more accurate analogy would be with the European Court of Human Rights, where the same Court is divided into a “Chamber” and a “Grand Chamber.” Chamber judges and Grand Chamber judges are drawn from the same overall pool of ECHR judges: “The Grand Chamber is made up of the Court’s President and Vice-Presidents, the Section Presidents and the national judge, together with other judges selected by drawing of lots.” But: “When it hears a case on referral, it does not include any judges who previously sat in the Chamber which first examined the case.” (Emphasis Supplied)

The issue, therefore, is not – as Justice Bhat’s line of questioning suggests – about how we are to understand “bias” from common law precedent. The issue is an institutional one: if what we now have is a situation where the Supreme Court has thirty-four judges, hears cases in panels of two or three that end up disagreeing with each other, and the resolution to that disagreement is by a panel of larger strength, then it is clear that the only way if this system is to work at all is if the latter panel has fresh judges. The alternative – that the same judge (or judges) sit on the larger bench borrows from two incompatible worlds: the problem is caused because of our poly-vocal Court with its thirty-four judges, but the solution comes from a world in which there is still one institution that automatically speaks with one voice.

And of course, it is here that the role of the Chief Justice – as “Master of the Roster” – is a crucial one, as discussed in the last post; and it is here that the need for standards that guide that discretion in the establishment of benches become so crucial. The establishment of these standards, it bears repeating, is not because people are out to “malign” the Chief Justice, but to ensure that the uncanalised discretion that opens up the Chief Justice to malignant accusations is actually subjected to public and democratic norms.

[Disclosure: The author clerked with Justice Bhat (as he then was) at the High Court of Delhi, in 2014.]


The Land Acquisition Bench and Continuing Issues around the “Master of the Roster”


, ,

[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

On October 12, it was reported that a five-judge bench of the Supreme Court would start hearing a set of cases about the interpretation of Section 24 of the Land Acquisition Act (2013). The Bench would be led by Arun Mishra J., and would commence hearings on October 15 (tomorrow). The composition of the Bench, however, has immediately raised eyebrows. To understand why, it is important to set out a brief history of the dispute.

The legal controversy itself – that involves the fate of huge tracts of land across the country – has been ably summarised by Suhrith Parthasarathy and Namita Wahi. For our purposes, the point is this: in 2014, a three-judge bench of the Supreme Court (Pune Municipal Corporation) interpreted Section 24 of the Land Acquisition Act in one way. The Supreme Court’s interpretation (as is normal) was followed by High Courts across the country, and also by multiple two-judge benches of the Supreme Court itself, for the next four years. However, in December 2017, a two-judge bench of the Supreme Court took a contrary view, and asked for a larger bench to consider the matter. In accordance with convention, the Chief Justice constituted a three-judge bench to look into the issue. Very swiftly after that (in February 2018, within two months of the December 2017 judgment) that three-judge bench (Indore Development Authority) also took the opposite view on Section 24 from Pune Municipal Corporation.

Now, in the normal course of things, the judgment of a three-judge bench is binding on all coordinate benches; consequently, if a three-judge bench disagrees with another three-judge bench, the correct thing to do is to refer the issue to the Chief Justice, so that a higher bench can lay down the position of law authoritatively.* However, instead of doing this, a majority of the three-judge bench – over a dissent by Justice Shantanagouder – held that Pune Municipal Corporation was per incuriam (a judgment delivered without the authority of law), and therefore not binding. In one stroke, therefore, the Bench in Indore Development Authority exempted itself from being bound by Pune Municipal Corporation, overruled all the two-judge bench decisions that had followed it, and declared that its reading of Section 24 was now the law. Soon after that, pending land acquisition matters in the Supreme Court began to be disposed off in accordance with the new understanding.

As Suhrith Parthasarathy pointed out at the time, in a legal system that rests upon the principle of stare decisis (consistency and uniformity of interpretation), coordinate benches overruling each other – especially in order to upset a settled interpretation of law – is improper. To this it may be added: calling another judgment per incuriam is not something that is done in the normal course of things. Per incuriam means not simply that the prior judgment is wrong, but that it is so wrong (it missed a binding statute or ignored a binding judgment) that it has no legal force at all. Indore Development Authority’s understanding of per incuriam, therefore, is itself open to doubt; but what is not open to doubt is the problematic manner in which the Indore Development Authority bench acted.

Unsurprisingly, this led to immediate chaos at the Supreme Court. The issue was brought to the notice of another three-judge bench of the Supreme Court, headed by Lokur J., who had been one of the parties to the original Pune Municipal Corporation decision (full disclosure: the author was, at the relevant time, working in the chambers of one of the senior counsel involved in the petitions). Lokur J. passed an order noting that the question of whether a three-judge bench could hold the decision of another three-judge bench to be per incuriam needed to be considered. Until this question – and the question of whether there needed to be a reference to a larger bench to decide the issue – was decided, he also requested other Supreme Court benches not to continue with disposing off pending land acquisition matters. The very next day, however, two two-judge benches of the Supreme Court – that were hearing the land acquisition cases – referred the matter directly to the Chief Justice. That was how the matter ended up with the Chief Justice (at the time, Dipak Misra CJI), and how the present Constitution Bench came to be set up.

I discuss this history because it reveals that there existed two clear – and very entrenched – views in the Supreme Court about the interpretation of Section 24 of the Land Acquisition Act. These views were entrenched enough for coordinate benches to declare judgments per incuriam, to overrule a long line of settled precedent on one side, and to request benches within the same Supreme Court to temporarily suspend hearing land acquisition cases, on the other (for the avoidance of doubt, this does not imply, of course, an equivalence).

That a five-judge bench is needed to resolve this controversy and lay down the law on the meaning of Section 24 may, ultimately, have been inevitable. However, here is the problem: the December 2017 two-judge bench order doubting the correctness of Pune Municipal Corporation was authored by Arun Mishra J. The February 2018 three-judge bench decision in Indore Development Authority, holding Pune Municipal Corporation to be per incuriam, was authored by Arun Mishra J. The 22 February order referring the question to the Chief Justice – in the teeth of Lokur J.’s order – was also authored by Arun Mishra J. And the five-judge bench that has been set up now to resolve the “conflict” is headed by Arun Mishra J. In other words, the same judge, sitting in a two-judge bench, doubted the correctness of a three-judge bench; then, sitting in a three-judge bench, overruled that decision (over a dissent) in favour of his interpretation of the law; when this was questioned by another three-judge bench, referred the case to the Chief Justice; and is now heading the five-judge bench to decide who was correct.

On any conceivable understanding of natural justice and the rule of law, this is simply unsustainable, especially in a Supreme Court that has thirty-four judges. And this brings us to the root of the problem, which is not really about personalities, but is institutional: the institution of the “master of the roster.” Recall that the Chief Justice’s position as the “master of the roster” (as entrenched in a series of judgments early last year) vests in her absolute discretion to constitute benches and allocate matters in the Supreme Court. When the controversy was at its height last year, I had written that the principle of the “master of the roster” – that originated out of administrative needs, and the requirement of needing someone to administer the court – was very problematic in the context of the modern Supreme Court. The Court’s strength (at that time, 26 judges), combined with the gradual weakening of the gravitational force of precedent effectively means that the Office of the Chief Justice’s administrative power of selecting benches can at least potentially in some cases translate into the power to affect outcomes (if not to determine them).

The composition of the Land Acquisition Bench gives us a textbook example of this. Because of the absolute discretion of the Master of the Roster in constituting benches, there is no questioning why the Bench is the way it is; however, the effect is self-evident – the Bench, which has been set up to resolve an interpretive controversy within the Supreme Court, is led by a judge who has been a protagonist on one side of the controversy, expressing his views not once but on several occasions, and not in extra-judicial forums, but through judgments of the Court. What would a detached and objective observer conclude upon seeing this? They would conclude that in this five-judge bench, at least one vote – the vote of the senior-most judge – is more or less decided (and it is unsurprising that the All India Farmer Association has already written to the Chief Justice making exactly this point).

And the solution – it bears repeating – has to be institutional. As long as absolute power remains concentrated in a single individual – who happens to be occupying the Office of the Chief Justice – issues of this kind will continue to arise, especially when the stakes are as high as they are in this case (as this thread summarises). The Master of the Roster has created a single point of failure, something that – it is well-known by now – is the surest death knell for institutional integrity. And the damage is long-lasting: in this case, for example, even if the Bench is now to be reconstituted, questions will linger over why it was constituted this way in the first place, and the continuing trustworthiness of a process that has been vitiated so badly in the first instance.

There are possible solutions: the establishing of Constitution Benches by a random draw of lots, or having a permanent Constitution Bench with the five senior-most judges (at any given time) occupying it; and so on. But what is clear is that the concept of the “Master of the Roster” has become – and will continue to be – a huge albatross around the neck of the Supreme Court. And in a poly-vocal Court of thirty-four judges, the present controversy has shown just how unsustainable it is.

*In this case, it is also interesting to note that until 2018, a very large number of Supreme Court judgments had followed Pune Municipal Corporation, without demurral about its correctness; the controversy actually began with the Indore Development Authority judgment, where a bench suddenly realised – after four years – that multiple Supreme Court judgments had all gotten it wrong).

Judicial Emergencies: Law and Practice


, , , ,

[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

One of the curious features around the ongoing litigation in the Supreme Court concerning the communications lockdown and executive detentions has been the absence of written documentation filed by the government, in court. For example, we still do not have access to the legal order imposing the communications lockdown, and in the habeas corpus cases, the Supreme Court’s “innovative” remedy of telling the petitioners to travel to J&K to meet the people who are allegedly detained, has effectively exempted the government from responding on questions of legality (so far).

In the absence of the government’s responses, it is impossible to know what legal and constitutional justifications it is offering for what are undoubtedly serious rights violations. In a previous post, I explained why this is a huge problem in a country – and a Court – that continues to (claim to be) governed by the rule of law. This situation, however, changed for the first time on October 1, when the State of Jammu and Kashmir (not the Indian Government) brought on record its reply (a “Limited Affidavit”) in Anuradha Bhasin v Union of India – the first of the petitions filed to challenge the communications lockdown.

From a survey of the Limited Affidavit (a copy of which is on file with the author of this post), the State’s rationale begins to appear. In paragraph 3, it states that the modus operandi of terrorists has been to infiltrate J&K from across the border, and then instigate local militants, with the help of separatist groups. In paragraph 5, the Limited Affidavit goes on to note that the government’s actions on Article 370 would jeopardise the position of these militants, as well as the separatist elements. And because of the influence that these groups enjoyed, there was an “imminent threat of deterioration of law and order.” In paragraph 11, the Limited Affidavit observes that in view of the “apprehension of misuse of Data Services”, requests were made to service providers pertaining to “different zones/areas” based on the threat perception, and which were ultimately confirmed in accordance with the Telecom Suspension Rules of 2017. In paragraph 12, it deals with the imposition of curfew orders under S. 144 CrPC on the basis of an apprehension of the deterioration of law and order, passed by various District Magistrates. In paragraph 14, the Limited Affidavit insists that the restrictions are temporary, and will be lifted based on an assessment of the situation in “each area.” Finally (for our purposes), in paragraph 16, the Limited Affidavit states that “the need-based restrictions were/are reasonable, had nexus with the purpose (pre-empt inflammation of passions and rumour-mongering).

Now for the purposes of this post, I want to bracket the issue of curfews and S. 144 (including issues around functioning of schools, hospitals, and markets), and focus exclusively on the issue of the communications lock-down. As I had discussed in the previous post, a communications lock-down – that infringes Article 19(1)(a) of the Constitution (at the very least) – can only be justified if the State’s measures meet the threshold of “proportionality”. One of the elements of “proportionality” requires the State to impose the least restrictive measure that is consistent with its legitimate purpose or goal. To take the very specific example of a communications lock-down: given the existence of “white lists” and “black lists” – which allow the State to block or disable mobile numbers in a targeted fashion – is it proportionate to shut off the communications of the entire state, instead of targeting individuals known to be – or even suspected to be – involved in or encouraging terrorist activities?

The Limited Affidavit, however, makes no argument at all on the question of proportionality. It argues that there exists a reasonable nexus between the measure (communications lock-down) and the goal (pre-empt inflammation of passions and rumour-mongering, presumably with a view to maintaining the 19(2) goal of “public order); now while even this connection is open to question (see, for example, recent research on the topic arguing against the belief that communications lock-downs prevent rumour mongering), what is clear is that no argument is offered in the Limited Affidavit for why more targeted measures cannot work (or were even contemplated). Surely it is not the State’s case that every individual in J&K is a potential inflamer of passions and a rumour-mongerer? That argument would be entirely at odds with our entire legal system’s focus on individual responsibility, and our aversion to collective punishment, or attributing collective criminality to entire groups of people.

But if that is not the argument, then what is? The only answer is to be found in the Limited Affidavit’s mention of “zones” or “areas” of potential disturbance, and also that the assessment is being made on the basis of the situation that these “zones” are in. That argument, however, falls into exactly the same problem discussed above: it is no longer about determining that there exists reasonable cause to (preventively) deprive an individual from exercising her Article 19(1)(a) rights, but that within an “area” (and “area” here means an entire state), everyone will be presumptively apprehended to be “misusing Data Services.”

Note, here, that this is not the same as Section 144 prohibitory orders in the physical world, where the State cordons off certain areas and prohibits assemblies there, ostensibly for the purpose of preventing riots. The justification for those prohibitory orders (and even they must meet the standard of proportionality, and cannot be perpetually extended) is that once the crowd is in place, you actually can’t separate the rioters from those caught up in the riot – and so you prevent the crowd from forming in the first place. That argument doesn’t translate into the digital world, especially when you already have a Section 144 order in place preventing assemblies on the ground.

That being the case, what justifies a departure from targeted shut-downs of identified mobile numbers to an assessment of what an “area” is like? I suggest that our Constitution allows for only one situation in which that approach is permitted: a declaration or “Proclamation” of Emergency, where (certain) rights may be suspended en masse, without the need for the proportionality assessment that may require individualised targeting. In other words, our Constitution recognises that there may exist rare and exceptional situations, where the situation is such that the a proportionate restriction of fundamental rights is (temporarily) impossible, and for the period of that impossibility, the State is exempted from adhering to the constitutional standard.

But here’s the crucial point: an Emergency must be declared formally. Not only must citizens be put on notice that their rights are suspended, but the existence of an exceptional situation – because of its very character – must strictly adhere to the legal formalities that are required to bring it into force. There’s been a lot of talk lately about an “undeclared Emergency”, but the point to note is that in legal terms, an “undeclared Emergency” is an absurdity. Without the legal form, it does not exist; and if it does not exist, it cannot be assumed in Court.

Or, to put the point more simply: there are two legal regimes. The normal legal regime, which requires the Court to rigorously apply its constitutional standards to violations of rights; and the Emergency regime, in which (some of) those rights stand suspended. But the two regimes cannot (legally) bleed into one another; where there is no Emergency, the State cannot “implicitly” invoke its logic in Court – and equally, the Court cannot, in its orders, act as if there was an Emergency.

The Limited Affidavit in Anuradha Bhasin’s Case, however argues as if there was an Emergency; but what is more worrying is that the Court’s orders (until now) appear to accept that. The most glaring example of this is the September 16th Order, which I discussed in my last post: the Court’s Delphic proclamation that the State ought to ensure that “national security” must be “balanced” with personal liberties is the logic of Emergency: not only does it impose no obligation upon the State, but in avoiding any mention of proportionality, it allows the State to determine exclusively how to achieve that balance – and thus, effectively, suspends those rights as legal rights.

This is what I would like to call a “judicial Emergency”: there is no proclamation of Emergency, but the Court – on its own initiative – acts as if there exists an Emergency, and its orders reflect judicial standards that are uniquely applicable to the Emergency regime. And this, in my submission, is the most appropriate conceptual framework within which to analyse the Court’s conduct on the Kashmir petitions over the last two months – a framework that is now fortified by the State’s first formal statement to come on the court record.

Of course, if we extend the analysis beyond communications shut-downs – and to habeas corpus – we find something even more troubling: because there the Court is acting not merely as if it was a judicial Emergency, but as if it was a judicial Emergency in 1976. Remember that the 44th Amendment – after Indira Gandhi’s fall – ensured that even during an Emergency, Article 21 – and therefore, the writ of habeas corpus – cannot be suspended. The Supreme Court’s conduct on habeas corpus, therefore, takes us into a world in which not only is there a judicial Emergency, but also, a world in which the 44th Amendment … no longer exists.

That is a somewhat concerning world to live in.

All Petitions are Equal, But Some Petitions are More Equal Than Others?

[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

On October 5, 2019, a Division Bench of the Bombay High Court delivered judgment in Vanakshakti v Union of India. The Petitioners had asked for a declaration that – among other things – the Aarey Colony be declared as a reserved forest or a protected forest under the Indian Forest Act, and a prohibition upon any development, construction or other non-forest work in the colony. The Division Bench dismissed the petitions on procedural grounds: it found that the two legal questions before it were pending adjudication in different forums. The question of whether the Aarey Colony was to be declared a forest, following the principles laid down in Godavarman’s Case, had been answered in the negative by the Bombay High Court in a judgement on 26th October, 2018. The Bombay High Court’s judgment had been carried in appeal to the Supreme Court, where it was pending. And a related issue – that of declaring the Aarey Colony as an eco-sensitive zone – was pending consideration before the National Green Tribunal. Notably, while dismissing the petition, the Division Bench also specified the procedural route that was open to the Petitioners. It noted in paragraph 25 that:

… pertaining to the issue of the area to be declared as a reserved forest or a protected forest since the proceedings before the Supreme Court are pending in T.N.Godavarman’s case (supra) petitioners should file an application before the Supreme Court and concerning the eco-sensitive zone Notification dated 5th December 2016 an application should be filed before the National Green Tribunal.

Subsequently, on 5th October, a public interest petition was moved before a separate Division Bench of the Bombay High Court. According to the petitioners, when the previous bench had rejected the Aarey Petition, an oral request had been made for a stay of the order – and thus, a restraint upon the cutting of any trees – until the petitioners could move the Supreme Court. This request had been rejected, but the Court had expressed an expectation that the Mumbai Metro Rail Corporation [“MMRC”] would not proceed with cutting trees. Perhaps unsurprisingly, the Division Bench refused to act upon the oral understanding, and rejected the petition.

Meanwhile, the MMRC – that had earmarked the site for the construction of a Metro rail shed – went full steam ahead with cutting the trees. There were widespread protests, arrests were made, various celebrities tweeted against the tree-cutting, and some political parties issued statements. Then, on the evening of the 6th of October, the Vacation Officer of the Supreme Court issued a notice stating that:

Take Notice that a Special Bench has been constituted to hear the matter tomorrow, i.e., 7th October, 2019 at 10.00 AM on the basis of a letter dated 6th October, 2019 addressed by Shri Rishav Ranjan with regard to feeling of trees in Aarey Forest, State of Maharashtra which has been registered as a Public Interest Litigation. (Emphasis Supplied)

As the concluding part of this story, today morning, the vacation bench of the Supreme Court passed an order stating that:

Shri Tushar Mehta, learned Solicitor General appearing for the State of Maharashtra has stated that they are not going to cut any further trees till the next date of hearing. In the circumstances, the statement is quite fair …  As prayed for jointly by the learned counsel for the parties, list these matters before the Forest Bench on 21st October, 2019, as the said Bench is hearing the matter pertaining to the similar issues in T.N. Godhavarman’s case. As undertaken, status quo be maintained till the next date of hearing with respect to cutting of trees.


Now in view of the fact that Shri Tushar Mehta, learned Solicitor General, was also heard to have said in Court that “whatever has to be cut is cut”, and in view of the fact that a destroyed forest cannot really be restored by a judicial order, it is unclear what precisely will be the impact of the hearing on 21st October, even if the petitioners were to succeed. That apart, however, the manner in which the Special Bench was constituted, and the hearing in question, raises some serious procedural concerns. The Supreme Court here acted in exercise of its “epistolary jurisdiction”, which basically translates to converting letters (“epistles“) to the Court into petitions, on the basis that it would facilitate justice in a country where not everyone is in a position to approach the Court with formally perfect pleadings. Now, leaving aside for a moment whether this was a fit case for the exercise of epistolary jurisdiction, the core problem is that the Supreme Court effectively registered – and then heard and passed orders on – a public interest litigation against a judicial order of a High Court. This, however, is entirely improper: public interest petitions are not – and cannot be – maintainable against judicial orders. To start with, public interest litigation must demonstrate a prima facie infringement of a fundamental right to be maintainable – and it has been settled at least since the days of Mirajkar that judicial orders, by definition, cannot violate fundamental rights. Therefore, Article 32 petitions (of which PILs form a subset) cannot be filed against judicial orders. The Vacation Bench was presumably bound by Mirajkar (a nine-judge bench judgment), but in issuing notice and directing status quo upon a PIL against a judicial order, it appears to have entirely ignored binding precedent.

This is not a hyper-technical point: there is a process by which High Court judgments can be challenged before the Supreme Court (following from a certificate of leave to appeal granted by the HC, through a Special Leave Petition etc.), and that process is extremely important if we are to preserve the judicial structure that presently exists. Allowing effectively anyone to challenge a High Court judgment in collateral PIL proceedings before the Supreme Court not only further dilutes the authority of High Courts, but also raises the potential of floodgate litigation, where just about anyone can show up with a PIL before the Supreme Court, on the basis that they are unhappy with a High Court judgment. In 2019, in the days of ambush PILs and serial PIL filers, this concern is a real and immediate one. And in this case, the use by the Supreme Court of its “epistolary jurisdiction” is particularly surprising, because the October 4 High Court judgement itself had provided the procedural path that had to be followed: recall once more that the Division Bench did not decide the case on merits, but because it was bound by the coordinate Bench decision from October 2018 – and that the correct way, therefore, was to file an application in the pending Godavarman case, where precisely this issue – declaring land as forest land – is being argued. If the urgency of the situation merited it – and I agree that it did – what was to stop the Supreme Court from immediately reconstituting the Godavarman bench, and hearing the matter as an urgent application?

Now of course, the argument will be that the situation was so pressing and so urgent, that these little procedural niceties must fall by the wayside, in order to achieve the overriding goal of saving the Aarey Forest. Unfortunately, in 2019, that position reflects a dangerous naïveté about judicial power and arbitrariness. The dilution of procedural constraints that limit the concentration of power within the higher judiciary, precisely on the basis of such ends-justify-means logic – is what has given us (for example) the destructive NRC process in Assam. Examples could be multiplied, but again, in 2019, one really shouldn’t need to.


There is, of course, a second rather large Kashmiri stag in the room (see above). As this blog post is written, the communications lock-down in the State of Jammu & Kashmir (a continuing threshold violation of Article 19(1)(a)) has entered into its sixty-second day, and multiple detentions remain un-adjudicated. When these issues were pressed before the Supreme Court, the Chief Justice retorted that the ongoing Ayodhya hearings left him with no time to hear the Kashmir petitions. Apart from the fact that the Supreme Court has thirty-four sitting judges, the Aarey Case shows that when the Supreme Court thinks that a matter is urgent enough, it does have time, Ayodhya notwithstanding: it has time enough to convene a special bench through its “epistolary jurisdiction” during the Dussehra Vacations, which sits at 10 AM (recall that the normal sitting time is 10 30 AM), ignores a nine-judge bench precedent, and passes status quo orders when (one wonders whether the undoubted “national interest” served by the Bombay Metro need not be “balanced” against environmental concerns?).

This is not some kind of a “gotcha” argument: rather, what this entire scenario shows, of course, is a deeper systemic problem that comes along with the expansion of judicial power, without the rigorous establishment of judicial standards to guide and – when necessary – check that judicial power. In the absence of those standards, what we get is – effectively – the rule of (mostly) men, rather than the rule of law. Aarey is urgent enough to merit a special vacation bench and status quo orders, but Kashmir is not: and the difference between the two rests solely in the discretion of the individual who, at the time, happens to be occupying the office of the Chief Justice. This is not a sustainable situation.

Nobody likes to see the environment damaged, and the MMRC’s undue haste in chopping down trees is indefensible. But I doubt very much whether weakening procedural constraints even further, and handing over even greater powers to the Supreme Court (and, in this case, in the Office of the Chief Justice) will end up doing very much for the cause of the environment in the future. Indeed, the history of environmental PILs warns us otherwise – but it is a history that appears to have gone largely unheeded.

Guest Post: Resigning MLAs, Constitutional Silences and a “Fraud on the Constitution”?: A Response to Mihir Naniwadekar


, ,

(This is a Guest Post by Goutham Shivashankar.)

This post is a response to Mihir Naniwadekar’s excellent and thought-provoking posts on the Bombay High Court’s decision in Vijay Namdeorao Wadettiwar v State of Maharashtra. Naniwadekar’s posts are available here and here. Naniwadekar argues that the High Court erred in failing to hold the recent induction of Radhakrushna Vikhe Patil (“RVP”) as a Cabinet Minister in Maharashtra’s BJP-led government to be a fraud on the constitution. As I understand it, his analysis is based on certain suspect premises. I hope to set out these shaky premises, and in the process, defend the High Court’s eventual ruling from Naniwadekar’s principal line of attack, i.e., fraud on the constitution.

Naniwadekar’s analysis stands on a misreading of the text of Article 164 of the Constitution: most critically, he fails to properly distinguish between the terms “Legislature of the State” and “Legislative Assembly of the State”, both of which find mention in Article 164. This distinction is crucial, especially in bicameral states like Maharashtra, which have Legislatures consisting of a Legislative Assembly and a Legislative Council and where the Council’s membership is not entirely elected. This primary error leads to a secondary suspect assumption. He assumes, arguably incorrectly, that Article 164(4) mandates that a Minister must necessarily be “elected” as a member of the Legislature of the State. The text of Article 164(4) does not seem to prohibit a “nominated” member. But some discussion in the Constituent Assembly debates and previous decisions of the Supreme Court do lend some support to his assumption that Article 164(4) does require a Minister to be an “elected” member of the State Legislature. Finally, his analysis also proceeds on a potentially incorrect reading of the Representation of People Act, 1951 (the “ROPA, 1951”). Naniwadekar assumes that the Petitioner’s contention in the case is correct, i.e., that the ROPA, 1951 prohibits by-polls being conducted to fill in casual vacancies that arise in the fag-end of an Assembly’s term.The ROPA, 1951 however, contains is no such prohibition. At least, I was unable to find any.

The Facts Restated (with one significant correction)

Naniwadekar’s posts capture the facts of the case with precision, except for one significant error. I do not propose to reinvent the wheel, but simply extract his summary. I also indicate in bold the erroneous factual assumption which he makes, and explain its significance. Naniwadekar summarises the facts as follows:

Mr. Radhakrushna Vikhe Patil (“RVP”) was elected as a Member of the Maharashtra Legislative Assembly in the 2014 state elections, as a candidate of the Indian National Congress. He was Leader of the Opposition in the Assembly.

In early June 2019, he resigned from the Assembly; and ceased being Leader of the Opposition. Disqualification proceedings before the Speaker remain pending, where one of the issues would be whether the provisions of the Anti-Defection law can be avoided by resigning from the Assembly before being declared as a defector. Under the Representation of the People Act, 1951, it was not possible to hold a by-poll for RVP’s assembly seat (or any other by-poll). This was because the term of the Assembly itself was to get over in less than six months.

However, on 16th June 2019, RVP was appointed as a Cabinet Minister by the ruling alliance. There was no possibility that RVP would become a member of the Assembly for the remainder of the term of the Assembly, as there was no question of any by-election being held.

RVP’s appointment as a Minister (along with some other appointments) was challenged before the Bombay High Court in Civil Writ Petition 6996 of 2019 (Vijay Namdeorao Wadettiwar v State of Maharashtra).

Naniwadekar’s summation is correct except in stating that the Representation of the People Act, 1951 (the “ROPA”) precluded the possibility of holding a by-poll for RVP’s “Assembly Seat (or any other by-poll).” This is incorrect on two counts.

Firstly, the ROPA, 1951, on a plain reading, does not contain any bar on holding a by-poll in the last six months/one-year of a Legislative Assembly’s term. Secondly, the ROPA, 1951 also certainly does not enact any such bar on by-polls to casual vacancies that may arise in the State’s Legislative Council.

Subject to this significant correction, the Naniwadekar’s account of the facts is entirely apposite. I will back my correction up in my analysis below. I will also indicate how this correction is crucial to the proper analysis of the case at hand.

Summarizing Naniwadekar’s Argument for Constitutional Fraud

Naniwadekar’s line of argument, broadly, appears to be this:

(i) A person appointed as a Minister in a State Government’s Council of Ministers must become a member of the State’s Legislative Assembly within a period of 6-months. This is a facet of the underlying constitutional value of ministerial responsibility to the legislature contained in Article 164 of the Constitution.

(ii) To appoint a person as Minister, where there exists no possibility at all that she could comply with the requirement in (i) above, would be constitutional fraud, since it would undermine ministerial responsibility.

(iii) There was “no possibility at all” of RVP being elected as a member of Maharashtra’s Legislative Assembly within 6 months of his induction as a Minister. This was because the Legislative Assembly was in the last six-months of its term and the ROPA, 1951 precluded by-polls to casual vacancies at this stage.

(iv) An implicit assumption in (i) – (iii) above, is that Article 164(4) requires that an inducted Minister must be “elected” as a member of the Legislature (if he is not already one at the time of his appointment as a Minister) within the prescribed six-month period. The mode of entry into the Legislature must be election, and not nomination.

A couple of extracts from Naniwadekar’s posts would be appropriate in culling out the above arguments. For instance, when distinguishing the Supreme Court’s judgment in Manoj Narula in his first post, Naniwadekar, presumably alluding to Article 164(4) of the Constitution, asserts that:

The point ultimately is that there is an express provision in the Constitution which provides that a minister must become a member of the Assembly within six months in order to continue. The question is whether one can be appointed as Minister when there is no possibility whatsoever of that condition being complied with… …But in the case of RVP, the Constitution does indicate that there is to be ministerial responsibility to the legislature and there is indeed a requirement that within six months, a minister must become a member of the Assembly. In that scenario, is it or is it not a fraud on the Constitution to appoint someone who has no chance whatsoever of complying with the mandate? That question is, with great respect, not concluded by Manoj Narula.” (emphasis supplied)

In his second post, Naniwadekar states:

In a situation of a vacancy shortly before the expiry of the term of the legislature, when no by-election at all is possible, then that vacancy should be filled up only by an existing member of the assembly: not because the Constitution requires this in express terms, but because not doing so will in every case have the effect of negating an underlying constitutional value. True, the Constitution permits a non-member to be a minister: but that permission is circumscribed by the underlying values which mandate that the person so appointed must necessarily be one who is capable of becoming a minister.(emphasis supplied)

As I understand it, each of these premises are erroneous. But before I set out why, it is necessary for me to extract the relevant legal provisions.

The Relevant Provision of Law

Constitutional Provisions

Article 164 (2) and (4) of the Constitution are relevant. They state as follows:

164. (2) The Council of Ministers shall be collectively responsible to the Legislative Assembly of the State.

(4) A Minister who for any period of six consecutive months is not a member of the Legislature of the State shall at the expiration of that period cease to be a Minister.

Additionally, Articles 168 and 171 are important. Article 168 provides that for Maharashtra, the Legislature shall consist of the Governor, and two Houses, i.e., Legislative Assembly and the Legislative Council. Article 171 prescribes the composition of the Legislative Council. Articles 171(3) (a)-(d) envisage that a total 5/6th of the Council’s membership shall be filled through elections by different electorates. Article 171(3)(e) envisages that the remaining 1/6th of the Council’s membership is to be “nominated” by the Governor. Lastly, Article 171(2) provides that: “the Legislative Council of a State shall not be subject to dissolution, but as nearly as possible one-third of the members thereof shall retire as soon as may be on the expiration of every second year in accordance with the provisions of law made by Parliament by law.

Provisions of the ROPA, 1951

Sections 15, 150- 151A of the ROPA, 1951 are relevant. Section 15 states as follows:

Notification for general election to a State Legislative Assembly. —(1) A general election shall be held for the purpose of constituting a new Legislative Assembly on the expiration of the duration of the existing Assembly or on its dissolution.

(2) For the said purpose, the Governor or Administrator, as the case may be shall by one or more notifications published in the Official Gazette of the State on such date or dates as may be recommended by the Election Commission, call upon all Assembly constituencies in the State to elect members in accordance with the provisions of this Act and of the rules and orders made thereunder:

Provided that where a general election is held otherwise than on the dissolution of the existing Legislative Assembly, no such notification shall be issued at any time earlier than six months prior to the date on which the duration of that Assembly would expire under the provisions of clause (1), of article 172 or under the provisions of section 5 of the Government of Union Territories Act, 1963 (20 of 1963), as the case may be.

Section 150 of the ROPA provides for by-polls to casual vacancies in State Legislative Assemblies. Section 151 provides for by-polls to casual vacancies in State Legislative Councils. Section 151A, which allegedly contains the “bar” precluding by-polls in the fag-end of a Legislative Assembly’s term actually states as follows:

“151A. Time limit for filling vacancies referred to in sections 147, 149, 150 and 151.— Notwithstanding anything contained in section 147, section 149, section 150 and section 151, a bye-election for filling any vacancy referred to in any of the said sections shall be held within a period of six months from the date of the occurrence of the vacancy:

Provided that nothing contained in this section shall apply if— (a) the remainder of the term of a member in relation to a vacancy is less than one year; or (b) the Election Commission in consultation with the Central Government certifies that it is difficult to hold the bye-election within the said period.”


Section 151A is the only provision of the ROPA, 1951 cited by the Petitioner before the High Court to substantiate his contention regarding a bar.

Four points are crucial to note here.

Firstly, Article 164(2) of the Constitution provides for “collective responsibility” of the Council of Ministers to the “Legislative Assembly”. In contrast, Article 164(4) requires that a Minister shall be a member of the “Legislature” of the State. The two terms are not synonymous. In a bicameral state, the “Legislature” of the State will include the Legislative Council as well.

Secondly, membership of the State Legislature can be achieved in at least 4 ways. One can become: (i) a Governor; (ii) a member of the Legislative Assembly chosen by “direct elections from the territorial constituencies of the State (Art. 170(1)); (iii) an “elected” member of the Legislative Council (Arts 171 (3) (a)-(d)); (iv) a “nominated” Member of the Legislative Council (Arts 171 (3)(e)). Textually, Article 164(4) does not seem to preclude a Minister securing membership of the State Legislature through any of these modes, though membership by becoming a Governor can safely be ruled out on grounds of the absurdity of a person being both Governor and Minister.

Thirdly, a Legislative Assembly has a fixed term subject to dissolution. In contrast, the Legislative Council of a State is a permanent body that is not subject to dissolution, and usually subject to biennial elections. To speak of the last 6-months of the term of the Legislative Council is erroneous.

Fourthly, neither Section 15 nor Section 151A bar the conduct of by-polls to casual vacancies arising in the fag-end of the Legislative Assembly’s term. Section 15(2) proviso merely prohibits the Election Commission from notifying the general elections too early, i.e, even before six month prior to the expiry of the Assembly’s term. The main portion of Section 151A mandates a time limit of six-months to hold by-polls in relation to casual vacancies that may arise either in the Legislative Assembly or the Legislative Council. The proviso to the section only carves out an exception to the six-month time limit in the event that the remainder of the “term of a member” in relation to a vacancy is less than one year. It does not preclude a by-poll at that stage, it merely exempts the Election Commission from adhering to the six-month limit.

The Errors in Naniwadekar’s Analysis

From my analysis of the provisions of law above, the following points emerge in relation to the Naniwadekar’s argument.

Firstly, proposition (i) of his argument as stated above is incorrect. There is no constitutional requirement that a minister must become a “member of the Assembly” within six-months in order to continue as a Minister. Article 164(4) says nothing of the kind. It states that a minister must become a “member of the Legislature”. Naniwadekar fails to distinguish between the terms “Legislature of the State” and “Legislative Assembly of the State”. Both terms appear in Article 164 of the Constitution, the former in Article 164(4) and the latter in Article 164(2). He has not considered that Maharashtra has a Legislative Council that is a part of the State’s Legislature. It is a permanent House, with biennial elections, one-third of whose members retire approximately every two years. There always existed the possibility of by-polls to fill up casual vacancies due to resignation or death of other Council members that may arise in the Legislative Council within 6-months of RVP’s appointment as a Cabinet Minister.

If I am correct, proposition (i) of Naniwadekar’s argument must be corrected to state: “A person appointed as a minister in a State Government’s cabinet must become a member of the State’s Legislature within a period of 6-months. This is a facet of the underlying constitutional value of ministerial responsibility to the legislature contained in Article 164 of the Constitution.”

If proposition (i) is restated as above, proposition (ii) would still be valid. If this restatement is not made, however, proposition (ii) would manifestly become incorrect. Put simply, it would be wrong to assert that appointing a person as Minister amounts to constitutional fraud merely because he is incapable of being elected to the Legislative Assembly within six-months of his appointment. It may, however, be correct to argue that it would amount to constitutional fraud to appoint a person as Minister if he in incapable of becoming a member of the State Legislature.

Proposition (iii) is incorrect because, as explained above, the ROPA contains no bar on conduct of by-polls in the last six-months of a Legislative Assembly’s term.

Proposition (iv) is unsubstantiated by the text of Article 164(4). The plain text does not require “election” to the State Legislature to continue holding the post of a Minister. Even “nomination” to the Legislative Council would suffice as per the plain text. However, the Constituent Assembly’s debates and many decisions of the Supreme Court do seem to proceed on the assumption that a Minister would be an elected member of the Legislature.

The upshot is this. The Bombay High Court was correct in rejecting an argument that RVP’s appointment was a fraud on the constitution. To that limited extent, Naniwadekar’s analysis is suspect. Otherwise, his analysis is brilliant. If you haven’t read his posts, please do so immediately.


Guest Post: “For Your Eyes Only”: On the Supreme Court’s Sealed Cover Jurisprudence


[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

[This is a guest post authored by Digvijay S. Chaudhary and Afzal Mohammad.]

There has been a lot of talk about the Supreme Court’s sealed cover exercise. The issue has been flagged only recently but the exercise has existed since the inception of the court. The following post follows from research that the authors conducted where they analysed this exercise from the inception of the Supreme Court until early 2019. The focus of the research was only the Supreme Court, but there exist thousands of cases in various High Courts regarding the same. The following post will trace the history and purpose of the exercise and then detail how it has evolved over the years and finally examine its legal backing.


It is important, first, to distinguish the various ways in which the Supreme Court has resorted to the sealed cover exercise. It is not always the case that whenever a sealed cover is used, the court does not reveal the contents of the sealed cover to the other party. The court has also (barring a couple of instances which shall be dealt later) never justified its stance of using the sealed cover. In a majority of the cases, it has been mentioned on record that a document was submitted to the court in a sealed cover but after this mere mention, there isn’t any record of such a document; whether it was relied on by the court or the impact those documents had on the judgment—no such information exists. It has also been observed that many times that the counsel themselves urge submitting documents in a sealed cover.

A distinction must also be drawn between the act of claiming privilege over documents and the sealed cover exercise we are dealing with. Under section 123 of the Indian Evidence Act, certain documents are not permitted to be presented in the court; the content of such documents is privileged and therefore, they are barred from being put on the record. Sometimes such documents lie in the hands of investigative agencies and if their content is disputed, they are called for by the court in a sealed cover. Only when the claim of privilege is not allowed, are these documents are made public. For the purposes of our discussion, such sealed cover documents shall not be included. It must also be borne in mind that all the exchanges that happen inside a courtroom are not part of the official record and hence details of the document in question cannot always be gathered by reading the judgment. As stated above, certain documents are ordered by the court in a sealed cover and their whereabouts and usage are not recorded by the judges in their judgment.

It can be inferred that over the years, the court’s approach has been four-fold when asking for documents in a sealed cover: Protection (sealing the evidence so as to avoid hampering), confidential (where the court acknowledges the confidentiality of the document and provides the document only to the counsel of both the parties and doesn’t dictate the contents in an open court), secret/arbitrary (where only the presiding judge opens and sees the sealed cover) and unknown (where it’s only mentioned that a certain document was submitted or asked to be submitted in a sealed cover, but nothing more is mentioned of such a document).


In the early years of the Supreme Court, the court used to resort to the sealed cover for protection of evidence as, “the absence of sealing naturally gives rise to the argument that the recording medium might have been tampered with.” Other purposes that we found the court resorting to the sealed cover were in cases concerning bids and tenders. That changed in 1972, when in the case of Unichem Laboratories v. Workmen, the statements of a private company were presented to the court in a sealed cover; no objections are present on record regarding their presentation or if the court itself asked for them in a sealed cover, but what is important is the court’s furnishing of reasons as to why the documents could not be made public: the court stated that these statements could have been extracted in the judgment but due to the fact that the company was a private company and the documents were presented in a sealed cover, therefore, they could not be made public. This deserves attention because the court reasons that the documents could not be made public as they were presented in a sealed cover—the sealed cover, here, acts as an authority which the court is supposed to rely on. Such a casual stance by the court, in this case, indicates that there might have been previous instances too where sealed cover was resorted to by the counsel/judges but written record of the same does not exist, as such instances have not found place in the judgments.

The pre–2000 era witnessed fewer variations in the documents submitted in a sealed cover. In this phase, the contents of the sealed cover were often revealed in an open court. In TADA cases too, evidence was frequently presented to the court in a sealed cover.

Wings of Fire

Post–2000, the sealed cover widened its ambit to investigation reports, reports in a court-ordered enquiry, report of the governor, status reports and steps taken in an investigation. Submission of status reports rose exponentially in the last four years (since 2016), as this term doesn’t even find a mention prior to this period. CVC reports, SIT reports, case diary, report of the inquiry conducted by the CBI, accounts, and list of assets of promoters, all came within the ambit of the sealed cover during this phase.

This exercise originates not only from the court’s end but also from the counsel, who themselves urge submitting documents in a sealed cover and which the court usually accepts. But in one such instance, in the case of Subramaniam Swamy v. Arun Shourie, the respondent prayed that, in view of the sensitive nature of the facts, he would choose to refrain from setting out those facts in the affidavit but would prefer to put them in the form of a signed statement in a sealed cover for the perusal of the Court, and which may be treated as an integral part of the counter affidavit. The Court, however, rejected his prayer and observed that the procedure suggested by the respondent was not acceptable, and was inconsistent with the recognized form of the pleadings. The respondent was granted liberty to withdraw the sealed cover from the Court. He was given an opportunity to file an additional affidavit. There are no reasons recorded by the learned judge as to why the procedure was not an acceptable procedure or any other details regarding such submission by the counsel. Nonetheless, this case is the only one the researchers could find in which the court has rejected the submissions in a sealed cover.

In another case, the High Court of Kerala directed the members of the CBI’s investigating team stating that, should they feel any interference with their freedom either from CBI or from elsewhere, such member shall be free to address this Court through the Registrar-General in a sealed cover. When this matter was brought to the Supreme Court, the court recognised the faith that the High Court had put in the members and allowed them to address the court through the Registrar in a sealed cover, but it did not comment on the validation of this exercise by the court.

The most extensive use of the sealed cover, however, has been observed in the case of Ratan N. Tata v. Union of India. There was a long list comprising a copy of a complaint, mentioned in the counter affidavit in a sealed cover. This sealed cover was directed to be kept in lock and key and be produced before the court on the next hearing. On the next hearing, there is no mention of this sealed cover. However, the ASG submitted another sealed cover which contained investigation reports of the inter-ministerial committee. This sealed cover was opened in court, resealed by the court master and returned to the counsel. Then, sealed envelopes containing the transcripts of all the tapes were presented to the court. In another hearing, the outcome of the scrutiny of alleged tapes was handed to the court in a sealed cover. The court opened the same and perused them. Finally, the report filed by the CBI was presented in a sealed cover to which the court ordered the registry to keep it in the same fashion and not to open it without the leave of the court. There are many similar examples.

There also have been instances where the court has issued reasons for resorting to the sealed cover. In Board of Control for Cricket in India v. Cricket Assn. Of Bihar, the Supreme Court accepted the statement of Justice Mudgal committee as to why the report was being submitted in a sealed cover: namely, that they were only allegations and were not established and, therefore, the report was being submitted in sealed cover so as not to sully the reputation of any innocent person. The court went through the report and said that the allegations in the report submitted in sealed cover required verification and investigation, and that the allegations were relevant to the subject-matter of the PIL. In SEBI v. Sahara India Real Estate Corpn., the Supreme Court stated that the press clippings be taken on record but should be kept in a sealed cover to maintain confidentiality in regard to the proposed transactions. In Sunil Bharti Mittal v. CBI, the court stated that since the matter was being monitored by this Court, progress reports of investigation were filed from time to time in sealed envelopes. And after perusing certain documents in a sealed cover, the court directed CBI to take action in accordance with the views expressed by it.

Sometimes, the document in the sealed cover is opened by the court, resealed and given to the registry to be kept in a sealed cover until further notice. Any future references/information about these reports/documentary are missing. Despite such a large extent of this exercise, the court has cautioned itself of this exercise only once. In the Alok Verma case the court stated that “we also make it clear that the present order requiring furnishing the report of the CVC in sealed cover to the learned counsel for the petitioner is being made in the peculiar facts of the case and as a one time measure. And, the above course of action has been considered necessary by the Court keeping in mind the need to preserve and maintain the sanctity of the institution of the CBI and public confidence in the said institution.

Supreme Court’s Backing For The Sealed Cover

The Supreme Court derives power for the sealed cover exercise from the Supreme Court rules. Rule 14 Chapter XX of the Supreme Court Rules provides:

“Notwithstanding anything contained in this order, no party or person shall be entitled as of right to receive copies of or extracts from any minutes, letter or document of any confidential nature or any paper sent, filed or produced, which the Chief Justice or the Court directs to keep in sealed cover or considers to be of a confidential nature or the publication of which is considered to be not in the interest of the public, except under and in accordance with an order specially made by the Chief Justice or by the Court.”

As per Article 145, the Supreme Court rules do not exist independently, and are subject to other laws in force. Section 327 of CrPC also provides for withholding publication of certain documents. However, section 327 lists the cases in which publication can be withheld. First, it bars access to the court in circumstances the presiding judge thinks fit. Second, it provides for in-camera trial for offences under section 376. Third, it prohibits publication of any matter in relation to in-camera proceedings only. This is logical as the concept of in-camera trials is created as an exception to the open justice principle and therefore, publication is restricted only in such circumstances.

The above rule, on the other hand, restricts publication whenever the court deems fit. This position stands in direct contrast to section 327. The said rule also stands in direct contrast to section 123 of the Indian Evidence Act. Section 123 has to be read with section 162 which provides for the act of claiming privilege over a confidential document by the state. Whenever privilege is claimed on any document, such documents are, until the question of privilege is decided, kept in a sealed cover. However, if privilege is not claimed on evidence under section 123, then there exists no other backing in law to provide for documents in a sealed cover. In addition to section 327 of CrPC, the rule now acts as an exception to this section too.


The court’s approach over the years has been patronizing; from protection, confidentiality, and secrecy, the sealed cover has touched all these spheres. Infantilisation through this exercise has been stated previously, but we think that this has much to do with convenience as well. The exponential rise (around 50% of all cases involving a sealed cover have been in the past 10 years) in the sealed cover exercise by the court in recent years is indicative of the fact that the apex court is slowly developing a constitutionalism of convenience.

In a history spanning more than 68 years and more than 402 cases, the Supreme Court has furnished reasoning in only the above cases. This speaks a lot of the neglected side of this exercise and how liberally it is used without giving any regard to reasons. It is important to reiterate that the Supreme Court does not publish the transcripts of hearing, it does not live-stream proceeding (not yet), does not publish written materials and requires an advocate-on-record’s signature on a form to physically access its premises; with such conditions on open justice, how does the Supreme Court plan on ensuring public confidence?