Category Archives: arbitrariness

The Supreme Court’s Triple Talaq Judgment

Today, a narrowly divided Supreme Court held that the practice of instantaneous triple talaq (talaq – ul – biddat) [hereinafter “triple talaq” for short] which authorised a Muslim man to divorce his wife by pronouncing the word “talaq” thrice, was legally invalid. On the outcome, the Court split three to two: Justices Nariman, Lalit and Joseph in the majority, with the Chief Justice and Justice Nazeer dissenting. However, Justice Nariman (writing for himself and Justice Lalit) and Justice Joseph used different – and partially contradictory – reasoning to arrive at the conclusion. With what is effectively a 2 – 1 -2 split, there will be considerable controversy over what, precisely, the Supreme Court held in this case. Before discussing the different opinions, therefore, it will be useful to provide a brief overview.

The constitutional status of triple talaq depended, in part, upon its legal status. In particular, there was a dispute over whether triple talaq had been codified into statutory law by the 1937 Muslim Personal Law (Shariat) Application Act. This was important, because all statutes are subject to fundamental rights. However, under existing jurisprudenceuncodified personal law is exempt from fundamental rights scrutiny. Therefore, if the 1937 Act did codify triple talaq, then the Court could examine whether it was consistent with the Constitution. If it did not, however, then the Court would have to ask whether triple talaq was part of Muslim personal law; and if so, whether to uphold its existing jurisprudence exempting personal law from fundamental rights scrutiny, or to reconsider it.

Within this framework, this is how the Court’s three judgments mapped out:

A. Does the 1937 Act codify triple talaq under statutory law?

Yes: Nariman and Lalit JJ

No: Kurien Joseph J., and Khehar and Nazeer JJ

A1. If the answer to A is yes, then does triple talaq (as codified by the 1937 Act) violate the Constitution?

Yes: Nariman and Lalit JJ (Article 14)

No: _____

N/A: Kurien Joseph J., and Khehar and Nazeer JJ

B. If the answer to A is no, then is triple talaq part of Muslim personal law – that is, is it uncodified Muslim personal law?

Yes: Khehar and Nazeer JJ

No: Kurien Joseph J

N/A: Nariman and Lalit JJ

B1: If the answer to B is yes, then can triple talaq be tested under the Constitution? 

Yes: ______

No: Khehar and Nazeer JJ

N/A: Nariman and Lalit JJ, Kurien Joseph J

C. In any event, is triple talaq protected under Article 25 as an “essential practice” of Islam?

Yes: Khehar and Nazeer JJ

No: Kurien Joseph J., Nariman and Lalit JJ.


A majority of three judges held that the 1937 Act did not codify triple talaq. Beyond that, however, there is no clear majority for any consequential legal proposition in this case (apart from a momentous change on the legal status of the doctrine of arbitrariness, which I shall deal with in a separate post). Justice Kurien Joseph – the “swing vote” in this case – agreed with the dissent that triple talaq had not been codified by the 1937 Act. This was at odds with the foundation of the judgment of Justices Nariman and Lalit, who held that the 1937 Act did codify triple talaq. However, Justice Joseph then disagreed with the next step in the dissent’s reasoning, which was the proposition that triple talaq was part of Muslim personal law (this, naturally, brought him into agreement with Justices Nariman and Lalit on the issue that triple talaq was not an essential or integral aspect of Islam, and therefore protected under Article 25 of the Constitution). What we therefore get, at the end of the day, is a majority in terms of outcome (3:2), a different majority on the interpretation of the 1937 (3:2) Act, but no majority for the reasoning leading up to the outcome.

The Judgment of Nariman J (joined by Lalit J)

Justice Nariman began by noting that talaq – ul – biddat was only one of the many permissible forms of divorce under Islamic law, and a strongly disapproved one at that (paragraph 9). With this brief background, he analysed the 1937 Act. Noting the Statement of objects and Reasons of the Act, which recognised a demand from the Muslim constituency that “Muslim Personal Law (shariat) should be made applicable to them.” Section 2 of the Act then stated that “Notwithstanding any custom or usage to the contrary… regarding… marriage, dissolution of marriage, including talaq, ila, zihar, lian, khula, and mubaraat… the rule of decision in cases where parties are Muslims shall be the Muslim Personal Law (Shariat).”

Justice Nariman held that the plain meaning of Section 2 was that, after 1937, the shariat was accorded statutory sanction in India. Or, to put it in simpler language, after the 1937 Act, what made the shariat legally enforceable in India (as applied to Muslims) was the 1937 Act. Before the 1937 Act, colonial judges were applying and enforcing the shariat (presumably) directly as religious sanctions, drawn from the Quran, the Hadith, and other authoritative texts. The 1937 Act, however, now mediated between Islamic scripture and its application in concrete cases.

It was argued by the Muslim Personal Law Board that the opening words of Section 2 – “notwithstanding any custom or usage to the contrary…” implied that the purpose of the 1937 Act was not to enforce Shariat, but to remove “custom and usage” as sources of Islamic personal law. Justice Nariman swiftly rejected this argument, holding that to allow a non-obstante clause to determine the interpretation of a Section that was otherwise unambiguous, would amount to “the tail wagging the dog” (paragraph 16).

Consequently, Justice Nariman was able to conclude that the 1937 Act (which included the statutory sanction of triple talaq) “would be hit by Article 13(1) if found to be inconsistent with the provisions of Part III of the Constitution, to the extent of such inconsistency.” (para 21). In other words, if the Court found that the practice instantaneous triple talaq violated any constitutional provision, then to the extent that Section 2 of the 1937 Act authorised it, it would be unconstitutional and void.

This would be true, of course, unless triple talaq was saved by any other constitutional provision. The Muslim Personal Law Board argued that it was saved by Article 25, which guaranteed the freedom of conscience and religion. Justice Nariman rejected this argument, pointing out that under Indian jurisprudence, Article 25 only protected “integral” or “essential” aspects of religion. In view of extensive and uncontroverted religious authority holding that triple talaq was an “irregular” way of conducting divorce, it could not, under any circumstances, be held to be an essential aspect of Islam (or under the Hanafi school of Islam, which practiced it) (paragraph 25).

Having strongly affirmed that it was the duty of the Court to strike down unconstitutional laws, and not leave the task up to Parliament (paras 26 – 30), Justice Nariman then came to the core of the case – the examination of the constitutionality of instantaneous triple talaq (paragraph 31 onwards). Focusing on Article 14 of the Constitution, he asked whether a law or a statute could be invalidated on the ground of “arbitrariness” (for a summary of the constitutional controversy on this point, see Mihir’s guest post here). After a detailed and technical discussion, Justice Nariman found that arbitrariness had always been a ground of legislative review under Article 14 (paragraphs 32 – 55), and judgments that held to the contrary were incorrectly decided.

The standard of arbitrariness required that if a law was “disproportionate, excessive… or otherwise manifestly unreasonable“, then it would be struck down under Article 14 (paragraph 45). Applying the standard to instantaneous triple talaq, Justice Nariman then held, in his concluding paragraph:

“Given the fact that Triple Talaq is instant and irrevocable, it is obvious that any attempt at reconciliation between the husband and wife by two arbiters from their families, which is essential to save the marital tie, cannot ever take place. Also, as understood by the Privy Council in Rashid Ahmad (supra), such Triple Talaq is valid even if it is not for any reasonable cause, which view of the law no longer holds good after Shamim Ara (supra). This being the case, it is clear that this form of Talaq is manifestly arbitrary in the sense that the marital tie can be broken capriciously and whimsically by a Muslim man without any attempt at reconciliation so as to save it. This form of Talaq must, therefore, be held to be violative of the fundamental right contained under Article 14 of the Constitution of India. In our opinion, therefore, the 1937 Act, insofar as it seeks to recognize and enforce Triple Talaq, is within the meaning of the expression “laws in force” in Article 13(1) and must be struck down as being void to the extent that it recognizes and enforces Triple Talaq. Since we have declared Section 2 of the 1937 Act to be void to the extent indicated above on the narrower ground of it being manifestly arbitrary, we do not find the need to go into the ground of discrimination in these cases, as was argued by the learned Attorney General and those supporting him.”

Three things stand out in Justice Nariman’s judgment. The first is his refusal to consider the question of whether personal laws are subject to the Constitution (although, in paragraph 22, he specifically casts doubt on the correctness of Narasu Appa Mali, and opines that it might need to be reviewed). In a guest post on this blog, Praharsh Johorey argued that the triple talaq case was an ideal opportunity to reconsider a judgment as clearly wrong as Narasu; elsewhere, I argued that a judgment invalidating triple talaq could either do it narrowly, through the 1937 Act and the essential religious practices test, or by taking a broad route, and reversing Narasu Appa Mali. Justice Nariman chose the narrow route, and in that sense, there is a feeling of a remarkable opportunity missed. To be fair, technically, it is difficult to fault him for this: once he had held that the 1937 Act codified Muslim personal law, there was no need for him to consider any other question. On this blog, I have often argued that judges should not go charging like wild horses over constitutional terrain, and ought to decide cases on the narrowest grounds available to them. I cannot, in good faith, criticise Justice Nariman for doing precisely that. Nonetheless, the sense of regret remains.

The second issue is Justice Nariman’s reliance upon the essential religious practices test to deny triple talaq the protection of Article 25. On this blog, I have tried to point out before that ERP is both constitutionally unprincipled and impractical, because it involves a secular Court making ecclesiastical judgments. I am not alone in this criticism: for the last four decades, ERP has been criticised by both scholars and practitioners; apart from a dissenting judgment by Justice Lakshmanan in 2004, however, it has never been seriously challenged within the judiciary. This case marked an ideal starting point for the Court to jettison this seriously flawed approach, and hold – along with Ambedkar in the Constituent Assembly Debates – that Article 25 simply wasn’t applicable to the laws of marriage, divorce, inheritance, which had a tangible impact upon the civil status of parties; in other words, one cannot, under the cover of religion, claim a vast domain of human life off-limits from constitutional values. As Ambedkar had said:

“The religious conceptions in this country are so vast that they cover every aspect of life, from birth to death. There is nothing which is not religion and if personal law is to be saved, I am sure about it that in social matters we will come to a standstill. I do not think it is possible to accept a position of that sort. There is nothing extraordinary in saying that we ought to strive hereafter to limit the definition of religion in such a manner that we shall not extend beyond beliefs and such rituals as may be connected with ceremonials which are essentially religious. It is not necessary that the sort of laws, for instance, laws relating to tenancy or laws relating to succession, should be governed by religion.”

In this second sense, the judgment represents a chance missed.

And thirdly, it appears to me that – notwithstanding his spirited revival of the doctrine of arbitrariness – Justice Nariman’s constitutionality analysis misplaces priorities. The core problem with instantaneous triple talaq was not its arbitrariness, but how, in giving men a unilateral power of instant divorce, it discriminated against Muslim women. It was more a question of unequal power and inequality (Article 15) than the rule of law (Article 14). Again, technically, one cannot fault the reasoning; in a broader sense, however, it seems to have achieved the right outcome, for the right reasons, but perhaps not… the best reasons.

The Judgment of Joseph J (for himself)

Justice Joseph wrote a brief judgment. He held that the Supreme Court, in Shamin Ara, had already held that “instantaneous triple talaq” was invalid under Islamic law (paragraph 1). It was necessary for him to carry out this analysis, because – in his view – the 1937 Act only made Islamic personal law applicable to Muslims, but was “not a law regulating talaq.” (paragraph 4) Noting that the primary authoritative source for Islamic personal law was the Quran, Justice Joseph then examined the Quranic suras that dealt with talaq, and found that:

“The Holy Quran has attributed sanctity and permanence to matrimony. However, in extremely unavoidable situations, talaq is permissible. But an attempt for reconciliation and if it succeeds, then revocation are the Quranic essential steps before talaq attains finality.51 In triple talaq, this door is closed, hence, triple talaq is against the basic tenets of the Holy Quran and consequently, it violates Shariat.” (paragraph 10)

Justice Joseph then cited multiple High Court judgments, leading up to the Supreme Court judgment in Shamin Ara, which had affirmed this proposition (paragraphs 11 – 23), and concluded that:

“Fortunately, this Court has done its part in Shamim Ara. I expressly endorse and re-iterate the law declared in Shamim Ara. What is held to be bad in the Holy Quran cannot be good in Shariat and, in that sense, what is bad in theology is bad in law as well.”

It is important to note that Joseph J. expressed no opinion on the question of whether uncodified personal laws are subject to the Constitution, and therefore, there is no majority in this judgment that supports that point of view. In paragraph 5, he made the limited observation that “I wholly agree with the learned Chief Justice that the 1937 Act is not a legislation regulating talaq. Consequently, I respectfully disagree with the stand taken by Nariman, J. that the 1937 Act is a legislation regulating triple talaq and hence, the same can be tested on the anvil of Article 14.” That is, his disagreement with Nariman J. was limited to the question of whether triple talaq, through the 1937 Act, could be tested under Article 14; however, since Nariman J. himself expressed no opinion on whether, if triple talaq remained uncodified, it could be tested under Article 14 (by overruling Narasu), Justice Joseph could not possibly have disagreed with him on this point, because there was nothing to disagree with.

That said, Justice Joseph’s analysis of Section 2 of the 1937 Act does not seem correct. The distinction between the 1937 Act enforcing the shariat, and the Act “regulating” triple talaq, is irrelevant to the constitutional analysis. What matters is not that the procedure of triple talaq is contained in a statute, but that the source of authority of triple talaq is a statute. The moment that is conceded, the statute in question – and along with everything that it authorises – becomes subject to Part III and the Constitution. On this issue, Justice Nariman’s view appears to be the correct one.

The Judgment of the Chief Justice (for himself and Justice Nazeer)

The Chief Justice’s judgment has the merit that, after page 176, when the recording of submissions ends, and the analysis begins, it is clear and easy to follow. That, however, is its only merit. The judgment advances novel constitutional propositions unsupported by the constitutional text, history, or precedent, and it severely undermines the constitutional balance between individual rights and religious precepts.

The Chief Justice began by noting that the sources of Islamic personal law are not limited to the Quran (paragraph 121), and that, in fact, all parties have agreed that talaq – ul – biddat is “bad in theology but good in (Islamic personal) law” (paragraph 127). Declining to go into an interpretation of rival hadiths provided by both parties, he noted that:

“The fact, that about 90% of the Sunnis in India, belong to the Hanafi school, and that, they have been adopting ‘talaq-e-biddat’ as a valid form of divorce, is also not a matter of dispute. The very fact, that the issue is being forcefully canvassed, before the highest Court of the land, and at that – before a Constitution Bench, is proof enough. The fact that the judgment of the Privy Council in the Rashid Ahmad case1 as far back as in 1932, upheld the severance of the matrimonial tie, based on the fact that ‘talaq’ had been uttered thrice by the husband, demonstrates not only its reality, but its enforcement, for the determination of the civil rights of the parties. It is therefore clear, that amongst Sunni Muslims belonging to the Hanafi school, the practice of ‘talaq-e-biddat’, has been very much prevalent, since time immemorial.” (paragraph 144)


“We are satisfied, that the practice of ‘talaq-e-biddat’ has to be considered integral to the religious denomination in question – Sunnis belonging to the Hanafi school. There is not the slightest reason for us to record otherwise. We are of the view, that the practice of ‘talaq-e-biddat’, has had the sanction and approval of the religious denomination which practiced it, and as such, there can be no doubt that the practice, is a part of their ‘personal law’.” (paragraph 145)

The problem with this argument is that paragraph 145 does not follow from paragraph 144. Under the essential religious practices test, as applied by the Supreme Court over time, not everything sanctioned by religion is integral to it. The Chief Justice slid seamlessly between noting that instantaneous triple talaq is practiced by Indian Muslims as a part of their religion, to holding that is an essential part of it, without showing independently that the threshold of ERP has been met. Recall that the Supreme Court has held, in the past, that neither worshipping at a mosque nor cow-slaugher on Id, are integral parts of Islam, on the basis that Islam does not mandate either practice. Under this standard, in this case, it would under the ERP, it would have to be shown that Islam mandated instantaneous triple talaq. This, the Chief Justice did not show; and while I disagree with the ERP test, given that the Chief Justice had chosen to apply it, I think it important to point out that he applied it incorrectly.

The Chief Justice then advanced a proposition that is utterly bizarre. In paragraph 146, he said:

“‘Personal law’ has a constitutional protection. This protection is extended to ‘personal law’ through Article 25 of the Constitution. It needs to be kept in mind, that the stature of ‘personal law’ is that of a fundamental right. The elevation of ‘personal law’ to this stature came about when the Constitution came into force. This was because Article 25 was included in Part III of the Constitution. Stated differently, ‘personal law’ of every religious denomination, is protected from invasion and breach, except as provided by and under Article 25.”

Notably, no authority is advanced to support this proposition. That is because there is none. No Court has held that “personal law” is a fundamental right. In fact, that sentence is incoherent – how can “personal law” have the “stature” of a “fundamental right”? Rights under Article 25 belong to individuals, not to “laws”. More importantly, Article 25 does not confer constitutional protection upon personal laws. It guarantees that all persons are equally entitled to freedom of conscience and the right freely to profess, practise and propagate religion.”

To go from “all persons are equally entitled to… freely… practice… religion” to “Article 25 protects personal laws” is to put language into a rack and torture it into a shapeless mass. What might have the Chief Justice been thinking? Perhaps he was thinking this: personal law falls within religion. Article 25(1) protects religion. Therefore, Article 25(1) protects personal laws. That train of thought, however, misses the fact that Article 25(1) does not protect religion per se, but protects an individual’s freedom to practice her religion; in other words, it does not protect religious norms, rules, or institutions, but individual rights. Now, it might be argued that, potentially, a Muslim man could approach the Court and argue that by denying him the option of triple talaq, his Article 25(1) right was being violated; such a case, however (apart form being decided on separate grounds altogether), is conceptually different from conferring the “stature” of fundamental rights upon an entire system of (personal law) rules, and the distinction is crucial.

Most of all, what is entirely unacceptable about this proposition is that, as the Chief Justice himself observed (in the extract quoted above), marriage affects an individual’s civil status and civil rights. The effect of holding that “personal laws” are protected under the Constitution’s religious freedom guarantee is to grant to religious bodies the power of determining individuals’ civil status (and their civil rights), without constitutional recourse. This seems to be a negation of the very basic meaning of secularism.

The Chief Justice then held that the 1937 Act did not codify triple talaq, but only negated the use of “customs and usages” in adjudicating cases between Muslims (paragraph 156). I have already argued above that this is a flawed reading of the 1937 Act. Consequently, he held that the only limitations upon personal law can be those found in the opening phrase of Article 25(1): “public order, health, and morality.” On this, he noted:

“… it is impossible to conclude, that the practice impinges on ‘public order’, or for that matter on ‘health’. We are also satisfied, that it has no nexus to ‘morality’, as well.

But why? He provided no reasoning for this. If “morality” under Article 25(1) refers to the concept of constitutional morality, then surely gender equality and non-discrimination art part of that definition of morality? And if not, what else does morality mean? What does the Chief Justice think it means, and why is instantaneous triple talaq “moral”? There are no answers.

The other preliminary phrase in Article 25(1) is “subject to… the other provisions of this Part” (that is, Part III). The Chief Justice held that this is also inapplicable, because Articles 14, 15 and 21 – which triple talaq potentially violates – are only applicable to State action against individuals, and not to private violations of rights (paragraph 165). However, not only does this argument go against the Supreme Court’s recent liquor ban judgment, which the Chief Justice himself signed on to, and which held that Article 21 places an affirmative obligation upon the State to protect fundamental rights – but it also ignores the fact that triple talaq is only legally effective because it is sanctioned by the Courts. Triple talaq does not operate in some parallel, extra-legal domain; rather, it is not only recognised (as an aspect of personal law) by the State, but it can also be enforced through the courts. Therefore, the State involvement is inextricable.

Lastly, the Chief Justice addressed an argument that instantaneous triple talaq violates principles of constitutional morality, which he rejected by reiterating the proposition that personal laws themselves are a part of fundamental rights, and ending with this paragraph:

“Religion is a matter of faith, and not of logic. It is not open to a court to accept an egalitarian approach, over a practice which constitutes an integral part of religion. The Constitution allows the followers of every religion, to follow their beliefs and religious traditions. The Constitution assures believers of all faiths, that their way of life, is guaranteed, and would not be subjected to any challenge, even though they may seem to others (-and even rationalists, practicing the same faith) unacceptable, in today’s world and age. The Constitution extends this guarantee, because faith constitutes the religious consciousness, of the followers. It is this religious consciousness, which binds believers into separate entities. The Constitution endevours to protect and preserve, the beliefs of each of the separate entities, under Article 25.” (paragraph 193)

It is a particularly stark irony that Chief Justice needed to replace the word “persons” (which is what Article 25(1) says) with the word “entities”, in order to sustain this unsustainable conclusion.

I have engaged with the dissent at some length, because a 3 – 2 split is a judgment by a hair’s breadth. Had one judge flipped, the dissent would have become the majority. While I feel that the majority opinions could have been stronger on some points, I feel – even more strongly – that the dissent, which elevates personal law to the status of the Constitution, and in fact, elevates it above all other fundamental rights in Part III, would – had it carried the day – done profound damage to the constitutional fabric. It would have fatally undermined the framers’ attempts to frame a secular Constitution, where religion could not become the arbiter of an individual’s civil status and her civil rights, and would, in a single stroke, have set back a long struggle for the rights of basic equality and democracy against the claims of religion.

What this divided judgment means for future jurisprudence dealing with the relationship between personal law and the Constitution, remains to be seen. The question is perhaps more open now than it ever was.




Filed under arbitrariness, Article 14, Article 21 and the Right to Life, Equality, Freedom of Religion, Personal Law, Secularism

The Constitutional Challenge to S. 139AA of the IT Act (Aadhaar/PAN) – II: The Union’s Arguments

In the previous post, I detailed the petitioners’ arguments in the constitutional challenge to S. 139AA of the Income Tax Act, which effectively makes enrolling for an Aadhaar number compulsory for taxpayers. After Petitioners completed their arguments, the Union of India – through the Attorney-General, Mr Arghya Sengupta, and Mr Zoheb Hossain – responded in defence of S. 139AA. Mr Shyam Divan and Mr Arvind Datar then replied for the Petitioners. In this post, I shall provide a brief account of the arguments, isolate the issues that the two-judge bench of the Supreme Court must address, and outline the possible outcomes.

The Attorney-General’s Arguments: A Limited Right to Bodily Integrity

The Attorney-General argued that the Petitioners’ Article 21 challenge to S. 139AA – focusing on bodily integrity – was nothing but a camouflaged privacy challenge, which the Court could not examine at this stage (see previous post on the dropping of privacy arguments from the present proceedings). However, assuming that there was an independent right to bodily integrity under Article 21, the thrust of the Attorney-General’s argument was that this right could be limited by the State – and in fact, was limited by the State in a number of other domains, in ways equally, or more, intrusive than Aadhaar. For instance, Section 32 of the Registration Act, 1908, required all ten fingerprints as a pre-requisite for registering property. People routinely subjected themselves to biometric collection while traveling abroad. More broadly, the Attorney-General argued that Aadhaar only did something that was already normalised and routinised in society: in an era of ubiquitous photography, what was so unacceptably intrusive about an iris scan? Given the ubiquity of online transactions through smartphones, what was so intrusive about having to part with one’s data for the purpose paying taxes? To accede to the Petitioners’ arguments would be to set a legal standard that could only be fulfilled by hermits living alone in the mountains, and not citizens part of the modern, digital economy.

Although the Attorney-General did not specifically use the term, what he appeared to be doing was borrowing the American doctrine of “reasonable expectation of privacy“: that is, the scope of one’s right to privacy – or, in the language in which this case was argued – “bodily integrity” – is determined by what is socially sanctioned and understood as reasonable at any given time. Briefly put, the Attorney-General’s case was that Aadhaar calls upon citizens to give up only that which they voluntarily and regularly give up as part of their daily lives; consequently, there was no constitutional violation to start with.

The Attorney-General then argued that in any event, citizens had no absolute rights over their bodies. In this case, there was a larger public interest that justified the level of infringement. For instance, the Medical Termination of Pregnancy Act regulated the conditions under which a woman could abort her foetus. Random breath checks for drunken driving were required in the interests of road safety. At this point, Justice Sikri interjected, and wondered whether those examples were analogous, because in those cases, the restrictions were clearly reasonable. To this, the Attorney-General replied that in the present case as well, there was a larger public interest: the effective and efficient collection of taxes, which was an integral part of life in an ordered society.

To substantiate this argument, the Attorney-General took the Court through the history of PAN Cards, and the perceived need to replace them with Aadhaar. He argued that on a random verification of 0.2% of all PAN Cards, a number of duplicate PANs were thrown up. Besides, over the previous twenty years, the existence of shell companies, the presence of directors in multiple companies, and multiple PAN holders, had all come to light. India was a highly tax non-compliant society. Furthermore, in addition to tax dodging, there were also problems of black money, which was used to finance terrorism. For all these reasons, it was important to develop a system in which identification could not be faked. The Attorney-General submitted that the only way to accomplish this was to digitise fingerprints and iris scans, and keep them for posterity. At present, fingerprints and iris scans could not be duplicated, and consequently, the shift to Aadhaar was necessary. Furthermore, this data would be encrypted and stored in a centralised server, and shared only with the police in case it was needed for resolving a crime.

Aadhaar itself, the Attorney-General submitted, was originally conceputalised to prevent leakages in the public distribution system, in payment of wages under the NREGA, in the payment of pensions, and so on. He argued that with Aadhaar, more than Rs 50 crores had been saved by plugging leakages. Consequently, even if there was some infringement of individual rights, it was balanced by the larger public goal, as mandated – according to the Attorney-General – by Rousseau’s conceptualisation of the social contract. Individuals were in a contractual relationship with the State from their birth, reliant upon it for a host of benefits; how then could they refuse to pay their taxes in the manner mandated by the State?

Indeed – according to the Attorney-General – the Supreme Court had itself sanctioned the use of Aadhaar in PDS schemes in 2011, and in SIM card verification just a couple of months before. Furthermore, an accurate identification system such as Aadhaar was needed to ensure that India was in compliance with its obligations under international agreements such as the FATCA. For all these reasons, the Attorney-General submitted that Aadhaar was entirely within the parametres of Article 21 of the Constitution.

He concluded his arguments by submitting that within the contours of the social contract, nobody had a right to make themselves invisible: “you may want to be forgotten. But the State does not want to forget you.

Returning to this argument at a later point in his submissions, the Attorney-General also argued that a number of Supreme Court cases – such as Kathi Kalu Oghad – had held that compelling persons accused of a crime to provide their fingerprints had been held not to violate the constitutional right against self-incrimination. So why couldn’t the State put in preventive measures to check tax dodging in advance? Justice Sikri interjected to say that the Kathi Kalu line of cases might not be apposite, because they involved accused in criminal cases; surely it was not right to treat the entire country as presumptively accused of tax dodging. The Attorney-General replied that he was only arguing that the right to bodily integrity was not absolute – it could be taken away in certain cases. “Which cases, is the question“, Justice Sikri responded. “Even your life can be taken away under Article 21“, the Attorney-General continued. “But only with due process,” Justice Sikri replied. “We must balance individual dignity with State interests.” The Attorney-General responded: “at the end of the day, if you can give your fingerprints for registering property, why can’t you give your fingerprints for this?

Still later in his submissions, the Attorney-General cited precedents from the American Supreme Court about urine testing for school athletes, DNA testing of a rape accused, and – perhaps paradoxically – Roe vs Wade – to reiterate that the right to bodily integrity was not absolute.

Article 19(1)(g)

The Attorney-General’s rebuttal to Mr Datar’s argument on Article 19(1)(g) was brief. He stated that there was no violation of Article 19(1)(g), and expressed surprise that in this day and age, someone was making a constitutional argument based on Article 19(1)(g).

S. 139AA and the Aadhaar Act

The Attorney-General submitted that there was no conflict between Section 139AA and the Aadhaar Act. Responding to Mr Shyam Divan’s submission that the coercive character of 139AA could not stand alongside the voluntary nature of Aadhaar, the Attorney-General argued that Section 7 of the Aadhaar Act was at least partially mandatory: the State could tell citizens “either you should have an Aadhaar Act, or you jolly well apply for it.” Furthermore, under S. 57 of the Aadhaar Act, Aadhaar could be used for purposes other than those stipulated in the Act itself. And in any event, he argued, Parliament’s power to prescribe uses for Aadhaar was plenary, and subject only to the Constitution. The Attorney-General added that the Aadhaar Act came built in with safeguards: Section 29, for instance, prohibited the sharing of information. True, there had been some leaks of late; but those leaks, he argued, had not come from the UIDAI, or the central government, but the Jharkhand state government; and in any event, biometric details had not been compromised – only bank account information had.

Parliamentary Legislation and Court Orders

The Attorney-General’s final argument was that whatever the status of the pre-2016 Supreme Court orders stipulating that Aadhaar could only remain voluntary, all these were overridden by subsequent legislation. There was no such thing as “legislative estoppel“. No Court could injunct Parliament from passing laws as it deemed fit. By passing the Aadhaar Act – and then s. 139AA – the Parliament had simply exercised its plenary powers, and passed validating legislation taking away the basis of the prior court orders.

The Attorney-General concluded his arguments by citing a World Bank Report praising the Aadhaar system. Everyone needed an identity, he argued. Many people in India had no identity. Aadhaar was a method to bring them into the mainstream, prevent exclusion, and guarantee them their dignity.

The Arguments of Mr. Arghya Sengupta: Article 14 and Proportionality

Continuing the case for the Union, Mr Arghya Sengupta argued that Mr Datar’s claim that 139AA violated Article 14 of the Constitution was incorrect, because Article 14 did not require the Court to undertake a proportionality analysis. He cited K.T. Plantations for the proposition that a proportionality test effectively amounted to judges substituting their wisdom for that of Parliament. Taking the Court through comparative law, Mr Sengupta submitted that traditional judicial review claims in the United Kingdom had never included a proportionality test. While the European Court of Human Rights did incorporate a proportionality analysis into its rights-analysis, this only caused greater confusion than resolution. Relying upon Lord Pannick and Lord Hoffman, Mr Sengupta submitted that “the reasons for not treating people equally often involve considerations of social policy.” Justice Sikri interjected to observe that equality claims in the United Kingdom – which didn’t have a Constitution – might be treated differently from how they were in India. Mr Sengupta responded that the broader point was that proportionality only entered the picture when some balancing of rights was involved. Article 14 only required the Courts to ask whether there existed a valid reason for treating people differently from one another. There was no question of balancing. In fact, Article 14 was not about “rights” at all; fundamentally, it was about “wrongs”. Mr Sengupta concluded this argument by citing Professor Rebecca Dixon for the proposition that even the proportionality test had begun to collapse into the traditional test, and argued for the retention of the traditional Indian test of intelligible differentia and rational nexus.

On the merits of Article 14 itself, Mr Sengupta argued that Mr Datar was incorrect in arguing that the disproportionate penalty for not complying with Article 139AA rendered it violative of Article 14. Relying upon the McDowell Casehe repeated his submission that proportionality could not be invoked to strike down a statute under Article 14. Nor could a statute be struck down on grounds of arbitrariness. Justice Sikri interjected that, by virtue of Mardia Chemicals, it might be possible to invalidate a statute on grounds of arbitrariness. In response, Mr Sengupta cited Rajbala vs State of Haryana, which had rejected the arbitrariness doctrine (for a previous discussion of this debate on this blog, see here; for an analysis of Rajbala, see here).

Coming to the traditional classification test under Article 14, Mr Sengupta opposed Mr Datar’s argument that by making Aadhaar compulsory only for individual assessees, S. 139AA violated the rational nexus test. He argued that, by definition, only individuals could have Aadhaar numbers (as opposed to companies, or HUFs). Consequently, Parliament had chosen to first focus on the problem of black money and tax evasion committed by individuals, and had brought in Aadhaar to check that. No enactment, Mr Sengupta argued, could completely solve a social problem. Parliament had decided to make a start with individuals, and at a future date, would devise ways for dealing with the other categories of assessees as well.

Justice Sikri said that he understood that there was no discrimination if companies were incapable of even having Aadhaar numbers. However, the question was why discriminate between two people, both of whom were willing to pay tax, if one of them was willing to enrol for an Aadhaar number, and the other was not. Mr Sengupta replied that the purpose of 139AA was not to discriminate, but to prevent duplication of PAN cards. So the discriminatory object test under Article 14 – as Mr Shyam Divan had argued – was inapplicable. In fact, much like in the US, when TIN was replaced by SSN, in future, the State might choose to replace PAN with Aadhaar entirely.

Conscientious Objection

Mr Sengupta submitted that Mr Divan calling his clients “conscientious objectors” who were being discriminated against was entirely misplaced. Citing texts on civil disobedience and conscientious objection, he argued that what Mr Divan was essentially arguing for was a license to break the law. You may not want to stand up for the national anthem, he pointed out, but that did not mean you could sit down. Justice Sikri observed that that might not be an entirely accurate framing; the petitioners had, after all, approached the Court to have the law struck down. Mr Sengupta replied that there could be conscientious objection to all kinds of laws, but that in itself did not make them discriminatory.

Informational Self-Determination

Mr Sengupta’s final argument was on informational self-determination. He submitted that there was no absolute right to informational self-determination. The State could – and did – collect a wide range of information from individuals: births, deaths, marriages. The information that the State required from its citizens was extensive, and nobody challenged it. In any event, Mr Sengupta argued, whatever right to informational self-determination – in the apparent guise of privacy – did exist, it would have to be conditioned and defined by cultural factors. India was very different from Germany, from where Mr Divan had drawn his doctrine. There would have to be devised an Indian doctrine of informational self-determination, drawn from Indian conditions. Citing Mark Tushnet on the dangers of comparative law, Mr Sengupta argued against “importing” the conception of privacy into India.

At this point, Mr Divan interjected and said that his argument was not a privacy argument, but an argument about his right to his body. Justice Sikri observed that there might be overlaps between the two concepts. Mr Divan responded that there might indeed be overlaps, but that his right to bodily integrity was not subsumed within his right to privacy. Mr Sengupta argued that it was not open to this Court to draw a distinction between privacy and informational self-determination; given that the very question of privacy was pending before the Constitution Bench, it was for the Constitution Bench to decide what the scope of privacy was, and whether or not it included informational self-determination. Returning to his argument about importing foreign law into India, Mr Sengupta cited Justice Antonin Scalia of the American Supreme Court who, in a death penalty case, had resisted the use of comparative precedent, arguing that in judging whether the death penalty in a particular case was “cruel and unusual punishment”, only “American standards of decency” ought to be looked at.

In any event, Mr Sengupta continued, even the German Constitutional Court required a balancing between individual and community interests. He repeated his submission that the right to informational self-determination was fundamentally about privacy, since the right to control information about oneself was a facet of privacy. If the Court was going to go into that, then there was a compelling State interest in the present case: that of preventing duplicate PANs, and ensuring efficient collection of taxes. Justice Sikri stated that the Petitioners would have to show why they had a right to pay taxes in the manner that they desired. Mr Sengupta continued by saying that biometric collection was the most sophisticated system presently known. PAN was the technology of 1975, he concluded; but Aadhaar was the technology of 2016.

The Arguments of Mr Zoheb Hossain

Mr Zoheb Hossain observed that India is a progressive tax regime. Progressive taxation was itself a facet of Article 14 of the Constitution. 139AA served this progressive goal by eliminating the inequality between taxpayers and tax evaders, by making duplication of PANs impossible. There was no discrimination against individuals; in fact, there were other provisions of the Income Tax Act – such as dividend distribution tax – applicable only to companies.

Mr Hossain then argued that Mr Divan was incorrect to argue that compelled taking of biometric details and iris scans amounted to compelled speech. Citing United States vs O’Brien, he argued that not every act or conduct amounted to “speech”.

Mr Hossain concluded by arguing that the standard for injuncting a parliamentary legislation was extremely high. There could be no injunction unless the statute was manifestly unconstitutional. Consequently, if the Court was inclined to refer the case to the larger bench, it ought not to grant a stay on the operation of S. 139AA.

(Disclaimer: The writer assisted the Petitioners in the constitutional challenge before the Court.)



Filed under aadhaar, arbitrariness, Article 14, Article 21 and the Right to Life, Bodily Integrity, Bodily Privacy/Integrity, Equality, Free Speech, Privacy

The Bombay High Court’s beef ban decision – II: On the Unconstitutionality of the Reverse Onus Clause (Guest Post)

(In this guest post, criminal law specialist Abhinav Sekhri examines the concurring opinion of Gupte J., which struck down Section 9B of the Maharashtra Animal Preservation Act (the reverse burden clause)

In a post yesterday, the Resident Author of this Blog engaged in a thorough analysis of various aspects of the decision delivered by the Bombay High Court in Sheikh Zahid Mukhtar v. State of Maharashtra. Consideration of the issues discussed by the separate opinion of Justice Gupte was left, which is what I seek to address. Justice Gupte tackles Section 9-B of the Maharashtra Animal Preservation (Amendment) Act, 2015 [hereafter, the Act], which reads:

In any trial for an offence punishable under sections 9 or 9A for contravention of the provisions of this Act, the burden of proving that the slaughter, transport, export outside the State, sale, purchase or possession of flesh of cow, bull or bullock was not in contravention of the provisions of this Act, shall be on the accused.

The separate opinion struck down Section 9-B of the Act as unconstitutional. In this comment, I first give a basic introduction to reverse burden (or reverse onus) clauses, moving on to discuss the specific reasoning adopted in the Separate Opinion for holding the provision unconstitutional.

An Introduction on Reverse Burdens


Ordinarily, in criminal cases the burden to establish the guilt of an Accused rests on the state. This dovetails with the essential principle of every Accused having a presumption of innocence in her favour. Over time, though, many statutory inroads have been made into this principle. Beginning with requiring an Accused to establish certain facts (evidential burdens), today there are several instances of Accused persons requiring to establish innocence itself (legal burdens). The Supreme Court held these clauses to be constitutional even where they impose legal burdens, in Noor Aga v. State of Punjab [(2008) 16 SCC 417].

Consider an example from the Narcotic Drugs & Psychotropic Substances Act, 1985 [hereafter, NDPS Act]. This employs two reverse-burden clauses. Section 54 of that Act creates a presumption that the Accused is guilty of an offence, if she fails to ‘satisfactorily account’ for possession of contraband. Section 35 states that in a prosecution under the NDPS Act, it would be presumed that the Accused has the ‘culpable mental state’ necessary for the offence.

How these operate was clarified by the Supreme Court in Inder Sain v. State of Punjab [(1973) 2 SCC 372, a case under the old Opium Act, 1878]; Noor Aga v. State of Punjab [(2008) 16 SCC 417]; Dharampal Singh v. State of Punjab [(2010) 9 SCC 608]; Bhola Singh v. State of Punjab [(2011) 11 SCC 653]. According to these decisions, the prosecution must prove initial facts – there was contraband, and it was in the conscious possession of the Accused – which creates a presumption of guilt, shifting the burden on the Accused to rebut that presumption. Note, though, that the prosecution is not required to prove that the Accused was knowingly in possession of the contraband itself. For instance, in Dharampal, 65 kg of Opium was recovered from the boot of the car owned and driven by the Accused. The prosecution was not required to prove that the Accused knew that the boot had the contraband. The presumption was cast as soon as the prosecution proved the material was contraband, and that it was in the boot of the car owned and driven by the Accused. It was now on the Accused to rebut this presumption.

The Separate Opinion’s Analysis of Reverse Burdens

At the outset, I must applaud the opinion for having extensively discussed the issue of reverse burdens. The vociferous opposition initially faced by these clauses across the common-law world has certainly shifted to a resigned acceptance in light of the perceived needs of law enforcement. However, we are now in a time when legislatures resort to such egregious provisions at the drop of a hat, exhibiting a numbness to the severe curtailment of liberties they entail. Parts of the opinion specifically address this problem, as Justice Gupte attacks the very need of having such a clause while dealing with a substance such as beef, which, as admitted by the state, carries no intrinsic harm or threat to society as opposed to say drugs [Paragraph 215, Pages 235-236].

Moving on to considering the opinion more substantively. Justice Gupte bases his attack upon Articles 14 and 21 of the Constitution of India. After citing various decisions, both Indian and foreign [Paragraphs 202-212, Pages 210-231], Justice Gupte arrives at the following four-fold test for considering the validity of any reverse burden clause [Paragraph 213, Pages 231-232]:

  1. Is the State required to prove enough basic or essential facts constituting a crime so as to raise a presumption of balance facts (considering the probative connection between these basic facts and the presumed facts) to bring home the guilt of the accused, and to disprove which the burden is cast on the accused?
  2. Does the proof of these balance facts involve a burden to prove a negative fact?
  3. Are these balance facts within the special knowledge of the accused?
  4. Does this burden, considering the aspect of relative ease for the accused to discharge it or the State to prove otherwise, subject the accused to any hardship or oppression?

The First Condition restates that the provision must require basic facts to be established by the prosecution before talk of any presumptions and reversing burdens. This must bear enough of a ‘probative connection’ with the presumption sought to be drawn. Going back to the example of Dharampal above, we can usefully juxtapose it with the decision in Bhola Singh. In the latter, the Supreme Court set aside the conviction for a co-owner of a truck which was caught transporting contraband based on the presumption. The Supreme Court held the presumption was not attracted, as the prosecution didn’t prove any basic facts connecting the Accused with the contraband. The prosecution didn’t show the Accused knew how his truck was being used by the contractor, let alone him knowing that the truck was being used to smuggle contraband.

From this, the Second and Third Conditions address the kind of facts that are to be presumed. The idea behind this is an understanding that reverse burden clauses only make sense where the issue is lying within the ‘special knowledge’ of the Accused or to prove ‘negative facts’. ‘Negative facts’ are the existence of permissions such as licenses or tickets, which play a role where offences rely on the inexistence of these permissions. ‘Special knowledge’ is a reference to mental states of an Accused. This is trickier. Any culpable mental state will always be a matter for the ‘special knowledge’ of an Accused, inviting the argument that in every trial the burden must be on the Accused to disprove intention after the prosecution establishes the physical act. Enter, the fourth condition, which places the handbrake on extending such logic too far. It reminds us of the fundamental David v. Goliath nature of the contest though, and how difficult it is for an Accused to find and present evidence to establish his innocence beyond his own word for his deeds.

Overall, then, there is little to disagree with the framework Justice Gupte establishes for evaluating reverse burden clauses in his separate opinion. The first and fourth conditions are the actual ‘tests’ upon which the reverse burden clause will be tested. If you clear the first test, the framework requires determining what kind of presumption is placed on the Accused, before considering whether placing such a presumption is unduly burdensome and oppressive. Considering there is no such clear test available in judicial decisions at present, it is a welcome contribution which should prove helpful in providing a systematic consideration of the various reverse burden clauses we have at present, and are bound to have in the future.

Applying the Tests to Section 9-B of the Act

Section 9-B is meant to apply to every offence under the Act, i.e. offences under Sections 5, 5-A, 5-B, 5-C, 5-C and 6:

  1. Section 5 makes it an offence to slaughter cows, bulls, or bullocks in any part of Maharashtra;
  2. Section 5-A makes it an offence to transport and/or export cows, bulls or bullocks (or cause to, or offer to do so) within or outside Maharashtra in contravention of the Act or with knowledge that the animal will, or is likely to be, slaughtered;
  3. Section 5-B makes it an offence to purchase/sell/otherwise dispose of cows, bulls or bullocks (or offer to do so), knowing or having reason to believe the animal will, or is likely to be, slaughtered;
  4. Section 5-C criminalized possession of the flesh of a cow, bull or bullock slaughtered in contravention of the Act;
  5. Section 5-D criminalized possession of the flesh of a cow, bull or bullock slaughtered outside Maharashtra;
  6. Section 6 regulated the slaughtering of cows, bulls or bullocks and made it an offence to slaughter such animals without obtaining a certificate.

The obvious problem with Section 9-B is its broad and indiscriminate phrasing, placing the burden upon the Accused for proving every ingredient of the offence. In order to make the section apply to all offences in the Act, it burdened the Accused with proving “that the slaughter, transport, export outside the State, sale, purchase or possession of flesh of cow, bull or bullock was not in contravention of the provisions of this Act.” The basic fact concept would require the State to at least prove that the cow or flesh in question was slaughtered, transported, exported, sold or purchased. It would also require that a connection be drawn between that cow or flesh in question and the Accused. Section 9-B allowed the prosecution to get away without doing anything, violating the very basic First Condition.

Justice Gupte, however, begins by specifically looking at the possession offences under Section 5-C and 5-D of the Act through the lens of the Second Condition [Paragraph 214, Pages 233-235. After immediately concluding that the Accused could never establish the ‘negative fact’ of the flesh not being that of a cow or bull, Justice Gupte immediately also concludes that the Accused could never establish that the flesh was obtained in contravention of the Act to begin with, and immediately thereafter concludes this violates his Fourth Condition of oppression. At the end of these assertions, he states that it would be easier for the State to prove these foundational facts, and absence of any such requirement was itself a problem. While having created a clear four-fold test, Justice Gupte doesn’t maintain these neat distinctions in his scrutiny of the provision on its anvil. The confusion is apparent in the Paragraph, and blunts what would have otherwise been a forceful criticism of the provision.

These objections were in some measure expected by the State. As recorded in the Separate Opinion, the Advocate General for Maharashtra conceded Section 9-B would apply only after the prosecution established certain initial facts. At one point, it is recorded how the Advocate General interpreted Section 9-B to apply only to create a presumption after every physical act had been established. This interpretation would have meant Section 9-B operated to presume that the concerned physical act – transport/export/possession etc. – was done knowing that it was contrary to the Act.

This would have made Section 9-B akin to those clauses which presume the existence of the mental state, considering it to be within the ‘special knowledge’ of an Accused. However, as Justice Gupte rightly noted, “Constitutionality cannot be a matter of concession by the State at the hearing” [Paragraph 215, Page 235-236]. Justice Gupte doesn’t stop there, though, and then proceeds to attack the ‘special knowledge’ justification offered by the Advocate General [Paragraphs 216- 217, Pages 237-240]. As I mentioned above, it is here that Justice Gupte criticizes the usage of reverse burden clauses in this context of possessing cow meat, citing the inherently harmful nature of substances such as narcotics which can warrant possession-based presumptions. This is critical, and exposes future legislation employing reverse burdens with possession based offences to closer scrutiny under Article 14 of the Constitution.

The Opinion then moves to Section 5-A and 5-B. Section 9-B would have placed the entire burden on the Accused, which was naturally unsustainable. A version that kicked in only to create a presumption that the Accused had knowledge that the offending acts of transport/export/sale/purchase would lead to slaughter or were otherwise in contravention of the Act was also found unsustainable. The basis for this conclusion was an insufficient connection existed between the prosecution proving the physical acts and the ultimate presumption of knowledge (violating the First Condition). The offence was purely based on knowledge, and presuming that very ingredient would subject the Accused to oppression (violating the Fourth Condition). Section 9-B was unconstitutional when applied together with Sections 5-A and 5-B as well.

Lastly, Justice Gupte arrives at Sections 5 and 6, which were present before the Amendment. The analysis is crisp as it is brief – Sections 5 and 6 read with Section 9-B would involve a presumption of foundational facts, which would render its use unconstitutional for a violation of his First Condition.

What to Expect Now?

Newspapers have already reported that the State of Maharashtra plans to challenge this decision in the Supreme Court. A part of me hopes this happens, as it would enable the Supreme Court to, hopefully, endorse the exposition of the Separate Opinion on reverse burden clauses. The test created is clear, and potentially allows for greater clarity in examining these provisions which currently abound our statute book. In its current form, Section 9-B is far too broad to be considered sustainable. One would assume that the chances of the Supreme Court overturning the verdict are unlikely.

The higher probability is of a modified Section 9-B emerging from the Maharashtra legislature, incorporating the position espoused by the Advocate General during the hearing before the Bombay High Court. A version of Section 9-B which explicitly places an initial burden on the prosecution would then place the focus squarely on the other condition created by Justice Gupte, of such a provision placing an Accused under oppression and undue hardship. That consideration, whenever it does happen, will prove to have a lasting impression on how reverse burden clauses are viewed in our criminal justice system.

Leave a comment

Filed under arbitrariness, Article 14, Reverse Onus Clauses