The interpretive question Section 377 left to the courts

Section 377 posed a question that its own words could not answer, and the Navtej Singh Johar judgment is best understood as an attempt to answer that question with precision rather than with breadth. What exactly did the Supreme Court decide in 2018, what part of the provision did it deliberately leave standing, and why does that distinction between decision and remainder matter so much for an understanding of the offence? Those are the questions around which an account of the case must be organised, because the judgment did not simply erase a section from the Indian Penal Code. It read the section down, it preserved a residue that the Court treated as severable and defensible, and it left that residue in force as law. A later act of Parliament then altered the picture again, for the replacement code of 2023 repealed the whole section and carried none of the preserved residue forward. The interpretive problem, the judicial response to that problem, and the legislative sequel therefore have to be kept distinct. If they are collapsed into a single story of repeal, the care with which the Court drew its line is lost, and the quite different character of what Parliament did afterwards is obscured.

The starting point is the difference between a provision as written and a provision as applied. A criminal section speaks in general terms, through a form of words chosen long before the disputes that test it arise, and it leaves to courts the work of saying what those words cover in a particular prosecution and what they cannot lawfully cover at all. With Section 377 that gap between text and application was unusually wide. The central phrase in the section, carnal intercourse against the order of nature, was not defined anywhere in the Code, and no accompanying definition elsewhere in the Code supplied the missing content. The phrase carried an appearance of moral certainty while offering no legal test by which a citizen, a police officer, or a judge could determine its boundaries with confidence. To ask what Section 377 meant was therefore to ask a question that the enacted text directed away from itself and towards a body of judicial interpretation built up over time. The interpretive question left to the courts was not a peripheral difficulty in an otherwise clear offence. It was the centre of the offence, and any constitutional challenge to the section had to confront the way in which meaning had been supplied, extended, and contested through interpretation rather than through legislative definition.

That openness in the language explains why the litigation concerned acts, language, and reach rather than a named class of persons described in the statute book. The section never named homosexuality, and it named no identity of any kind. It described conduct, in abstract and condemnatory terms, and it attached a severe penalty to conduct falling within that description. A court asked to examine such a provision faces a choice of levels at which to conduct the examination. It may look at the words in isolation, as a definition of prohibited conduct. It may look at the words in their statutory setting, within the chapter in which the section stood and under the heading that framed it. It may look at the words in their practical operation, in the way the threat of their use shaped behaviour far beyond any courtroom. The Navtej proceedings required attention at each of those levels, because a challenge that addressed only the dictionary sense of the words would have missed the manner in which an undefined phrase acquires force in the world, and a challenge that addressed only practical abuse would have missed the prior question of what the law, properly interpreted, actually prohibited. The judgment had to hold text, context, and consequence together, and to give an answer that could be applied by other courts to the text that remained.

Reading down is the technique that allowed the Court to give such an answer without treating the section as indivisible. To read a provision down is to interpret it so that it remains valid in a narrowed form, excluding from its scope the applications that cannot stand, while retaining the applications that can. The technique depends on severability, on the proposition that the lawful and unlawful reaches of a single set of words can be separated by interpretation, and that the part which survives separation is coherent law rather than a fragment that the court has invented. That proposition always requires justification, because courts interpret legislation and Parliament enacts it, and the boundary between the two functions is part of what gives a judgment of this kind its discipline. The Court therefore had to show not only that a narrowed reading was available on the language, but also that the narrowed reading corresponded to a defensible legislative purpose that could be distinguished from the purpose that had to be rejected. In the case of Section 377, the preserved part concerned non-consensual acts, minors, and animals. Those subjects supplied the coherence that reading down requires. They described situations in which consent was absent or could not be given in law, or in which the conduct involved a being that the criminal law protects through prohibition rather than through an inquiry into agreement.

The choice to preserve that residue is one of the most instructive features of the judgment, and it is also the feature most easily misunderstood. Preservation was not hesitation, and it was not a qualification that weakened the central holding on the applications that were removed from the section. It was a statement about the proper limits of judicial power and about the continuing competence of the criminal law to address harm where consent is not present. By leaving an offence in place for non-consensual acts, for acts involving minors, and for acts involving animals, the Court signalled that its objection lay not in the existence of sexual offences as a category, but in the application of this offence to conduct that it held could not be criminalised. The line drawn was therefore analytical before it was moral. It turned on consent, on legal capacity, and on the distinction between conduct between persons able in law to choose and conduct in which choice is absent or legally irrelevant. A reader who treats the judgment as a simple declaration that Section 377 had gone misses that analytical line, and with it the reason why the surviving text took the form that it did. The remainder was not left by oversight. It was left by design, as a functioning offence with a defined, if narrowed, field of operation.

Against that background, the legislative sequel acquires its full significance. Parliament, in enacting the replacement code of 2023, repealed the whole of Section 377. Repeal by legislation is a different juridical act from reading down by a court, and the difference matters for more than reasons of form. A court reading down preserves what it holds can lawfully be preserved, because its authority is interpretive and its task is to save legislation where a valid construction is available. A legislature replacing a code is under no corresponding obligation to carry forward a judicially preserved residue, and in this instance it carried forward none of it. The offence that the Court had retained for non-consensual acts, for minors, and for animals was not re-enacted as part of the repeal. The section disappeared in its entirety, both in the applications that the Court had held impermissible and in the applications that the Court had expressly allowed to stand. That outcome creates the paradox that gives the article its point. A judgment celebrated or criticised for the precision of its distinction was followed by a legislative change that treated the section as a whole, without reproducing the distinction at all. To understand Section 377 after 2018 is therefore to hold two endings in view at once, a judicial ending that was partial and reasoned, and a legislative ending that was complete and silent on the residue.

The age of the provision adds a final layer to the interpretive question. Section 377 belonged to a code enacted in the nineteenth century, and longevity alone settles nothing about validity, since an old provision may still perform necessary work and a newer provision may still fail legal scrutiny. What longevity produced in this instance was accumulation. Interpretation gathered around the enacted words until the operative meaning of the central phrase and the interpretation supplied by courts could not be cleanly separated. Parliament had left that accumulation undisturbed, first by enacting an undefined phrase and thereafter by leaving the phrase unamended while a Law Commission recommendation for deletion remained unenacted. The question that reached the Court had therefore been left to the courts in a double sense, by drafting and by omission.

What Section 377 actually said

What exactly did Section 377 of the Indian Penal Code say?

Section 377 provided that whoever voluntarily had carnal intercourse against the order of nature with any man, woman or animal faced imprisonment for life or imprisonment up to ten years and a fine, with penetration sufficient to constitute the intercourse required for the offence.

Section 377 was a provision of the Indian Penal Code, the code whose wider history and structure are traced in the complete guide to the Indian Penal Code, and its placement and wording both repay close attention. It stood in Chapter XVI of that Code, under the chapter heading Unnatural offences, a heading that announced a moral classification before the operative words of the section were even reached. The operative words themselves were brief, severe, and remarkably open in their central concept. The section provided as follows: “Whoever voluntarily has carnal intercourse against the order of nature with any man, woman or animal, shall be punished with imprisonment for life, or with imprisonment of either description for a term which may extend to ten years, and shall also be liable to fine.” To that main text the section added an Explanation in the following terms: “Penetration is sufficient to constitute the carnal intercourse necessary to the offence described in this section.” The whole offence, with its life sentence at one end of the scale and its single explanatory sentence at the other, occupied only a few lines of the Code. The disproportion between the brevity of those lines and the volume of interpretation, argument, and anxiety that gathered around them is the first fact that any close reading has to acknowledge.

The origins of the text lie in the codification exercise from which the Indian Penal Code emerged. The Code was drafted under the First Law Commission chaired by Thomas Babington Macaulay, the draft was prepared in 1837, the Code was enacted in 1860, and it was brought into force on 1 January 1862. Those dates matter to an understanding of Section 377 for reasons that go beyond antiquarian interest. A code is a systematic instrument, intended to state the criminal law in ordered chapters and defined offences, so that liability depends on enacted language rather than on unwritten custom or discretionary moral judgment. The presence, inside such an instrument, of a phrase as undefined as against the order of nature sits uneasily with the codifying ambition of the whole project. The drafters defined many concepts with care elsewhere in the penal law tradition that the Code represented, yet the phrase on which liability under Section 377 depended was left without a definition in the Code itself. That omission cannot be treated as a minor drafting gap, because the phrase was not decorative. It was the element that distinguished criminal intercourse from intercourse that the section did not reach, and its content determined the boundary between liberty and a sentence that could extend to imprisonment for life.

Each element of the main text therefore requires separate attention. The opening word, whoever, gives the section a universal form, applying to any person who performs the described act, and the adverb voluntarily qualifies that act by directing attention to the will of the person concerned. The section did not in terms address orientation, attraction, or identity, and it did not divide the population into classes, some of whom were offenders by status and others of whom were not. Its unit of analysis was an act, performed voluntarily, described by reference to a standard that the act was said to offend against. The objects of the act were stated in the widest terms, with any man, woman or animal, a formulation that placed human and animal objects within a single sentence and a single offence, without distinguishing between situations in which consent could be present and situations in which it could not. That refusal to distinguish within the text is central to the later history of the provision, because the distinctions that the Supreme Court eventually drew in 2018, between consensual and non-consensual conduct, between adults and minors, and between human and animal objects, had to be drawn by interpretation. They were not drawn by the enacted words, which treated the field as one and imposed one penalty structure across it.

The penalty structure itself reveals the gravity with which the described conduct was treated by the Code. Imprisonment for life stood as the maximum, with imprisonment of either description for a term which may extend to ten years as the alternative formulation in the text, and liability to a fine in addition. A sentencing range of that breadth confers considerable discretion on a sentencing court, but it also sends a signal that operates well before sentencing is reached. The classification of an offence, and the punishment that the statute book attaches to it, shape the way police, prosecutors, complainants, and accused persons understand the seriousness of an allegation. Where the maximum is imprisonment for life, an allegation under the section carries a weight that a lesser maximum would not carry, even in a case that ends without conviction or without a prosecution being completed at all. That weight formed part of the practical operation of Section 377, and it cannot be separated from the text on the ground that courts in individual cases might impose far shorter sentences. The statute book speaks through its stated penalties, and the stated penalty here placed an undefined act, identified only by its supposed opposition to the order of nature, among the offences that the Code punished most severely.

The Explanation attached to the section performed a different and more technical function, and its effect was to widen rather than to narrow the reach of the offence. By providing that penetration is sufficient to constitute the carnal intercourse necessary to the offence, the Explanation settled a question of proof and definition in favour of a low threshold. It made clear that the prosecution did not have to establish some more complete or differently characterised act in order to bring conduct within the section, once penetration and the other elements were shown. Explanations in a code are often treated as subordinate material, clarifying the main text without altering its substance, yet in a provision whose main text was already undefined at its centre, a clarifying sentence that lowers the threshold of the central act has substantive importance. It told courts that the concept of carnal intercourse in this section was to be given at least the breadth that the Explanation required, and it removed one possible line of narrowing argument that might otherwise have been available to a defendant who sought to place conduct outside the section on definitional grounds. Read together, the main text and the Explanation combined a vague standard of wrongfulness with a precise and expansive statement of the physical element, an unusual pairing that left the moral boundary obscure while making the physical boundary comparatively clear.

The absence of a definition for against the order of nature is thus the textual fact from which the rest of the analysis follows. Because the Code did not define the phrase, its reach was settled by courts over time, and the category that emerged through that process covered forms of non-procreative sexual activity. That judicial construction is sometimes described as if it were a gloss added to a clear provision, but that description understates the role that construction played. Without construction, the phrase had no determinate legal content capable of consistent application across cases, and with construction, the law in operation became in large part the construction itself. Two consequences followed from that dependence. The first was uncertainty at the margin, since a standard drawn from decided cases develops incrementally and leaves new situations to be fitted, by analogy, into a pattern that was not designed with them in view. The second was a transfer of effective lawmaking influence towards courts, prosecutors, and police, who in their different roles had to decide what the order of nature required in circumstances that the Code had not addressed in defined terms. A citizen seeking to conform conduct to the law faced a text that condemned by reference to nature without specifying the content of the reference, and that failure of specification is a rule of law concern in itself, apart from any view taken on the morality of the conduct concerned.

That concern was recognised in official review long before the Supreme Court gave its judgment in 2018. The 172nd Report of the Law Commission of India, published in 2000 and devoted to a review of the rape laws, recommended that Section 377 be deleted. The location of that recommendation within a review of the rape laws is significant, because it placed the section in a discussion concerned with the rational organisation of sexual offences, with consent, and with the proper boundaries between offences that protect persons against violation and provisions that punish conduct on other grounds. A recommendation for deletion by the Law Commission does not amend the statute book, and Section 377 remained in force after 2000 with its text unchanged. The recommendation nevertheless forms part of the verified record of the provision, because it shows that the case for removal had been articulated through the ordinary institutional channel for law reform, following examination, and had not been acted upon by Parliament. The long interval between recommendation and judicial resolution returns the discussion to the theme of responsibility. When a legislature leaves a recommended deletion unenacted and leaves an undefined offence unamended, the questions that the offence generates do not disappear. They migrate, to prosecutors deciding whether to proceed, to police deciding how to act on a complaint or on information, and ultimately to courts deciding what the surviving text can lawfully mean.

The practical operation of the section extended well beyond completed prosecutions, and an account confined to convictions recorded under it would give a misleading picture of its place in the legal order. The provision’s practical weight lay as much in the threat of prosecution, and in the police harassment and extortion reported by the Naz Foundation litigation, as in prosecutions carried through to judgment. That point is not an embellishment added to the textual analysis. It is a point about how criminal law works when an offence is defined in vague terms, punishable severely, and associated in its chapter heading and language with moral condemnation. A threat is effective when the person threatened believes that a complaint may be entertained, that arrest or investigation may follow, and that the process itself, with its exposure and expense and stigma, will impose a burden regardless of the final outcome. Extortion depends on precisely that belief, since the sum demanded is paid to avert a process that the victim fears more than the payment. Harassment, similarly, draws power from the existence of an offence under which a person can plausibly be questioned, detained, or threatened, even where no charge follows. The Naz Foundation litigation placed that operation before the courts as part of the case against the section, and any assessment of what the section actually said has to be accompanied by an assessment of what its saying made possible in practice.

When the text, the setting, the history, and the practical operation are placed together, the shape of the answer given in 2018 becomes intelligible. The Court confronted a brief section that imposed imprisonment for life for voluntary carnal intercourse against an undefined order of nature, with any man, woman or animal, explained by a sentence that made penetration sufficient, housed under a heading that condemned the offences as unnatural, and construed over time to cover forms of non-procreative sexual activity. It chose a narrowed answer, reading the section down so that an offence remained for non-consensual acts, for minors, and for animals. The distinctions based on consent and capacity that the enacted sentence had not supplied were supplied through interpretation. Parliament, in the replacement code of 2023, preserved none of that remainder and repealed the section as a whole, so the provision ended twice, first through a judgment that rewrote its scope and secondly through a repeal that removed even the interpreted remainder from the statute book.

Naz Foundation: the Delhi High Court reads the provision down

The litigation that first displaced Section 377 did not begin as a claim framed only in the language of identity. It began in the work of prevention. The Naz Foundation (India) Trust, a non-governmental organisation working on HIV and AIDS prevention, filed a writ petition in the Delhi High Court in 2001 arguing that Section 377 impeded HIV prevention work among men who have sex with men and violated Articles 14, 15, 19 and 21 of the Constitution. The pairing in that petition deserves close attention, because it shaped everything that followed in the first courtroom round. On one side stood a practical claim about health work, a claim that a criminal provision interfered with the ability of an organisation to reach a population whose health needs required candour, trust and sustained contact. On the other side stood a constitutional claim of considerable breadth, invoking equality, non-discrimination, freedom and life and personal liberty together. The petition therefore refused from the outset to separate the provision as a piece of criminal law from the provision as a condition under which people lived, sought help and assessed risk.

The first response of the Delhi High Court was nevertheless to decline the question. In 2004 the court dismissed the petition as raising an academic question with no cause of action. That dismissal is instructive, because it exposed a recurring difficulty in litigation of this kind. Where a law is challenged for the climate it creates rather than for a discrete act of enforcement against the petitioner, a court inclined towards caution can characterise the dispute as abstract. The description of the question as academic performed a particular kind of work. It suggested that without a concrete grievance of the recognised kind, the constitutional objections, however seriously framed, lacked a proper footing for adjudication. It also left the provision entirely untouched while appearing to decide nothing about its merits. A dismissal on that ground is not an affirmation of constitutionality, yet its practical effect for those living and working under the provision is indistinguishable from one for as long as it stands. The law remains, the impediment alleged remains, and the only thing resolved is that the court will not resolve the substance.

The correction to that approach came from the Supreme Court. In 2006 the Supreme Court set the dismissal aside and remitted the petition for a decision on its merits. The significance of that order should be stated precisely. It was not a judgment that Section 377 was unconstitutional. It was not an endorsement of the reading the Delhi High Court would later adopt. It was a direction that the questions raised deserved an answer rather than a refusal. In procedural form the order was modest. In substantive consequence it was considerable, because it restored to the courts a dispute that the 2004 dismissal had placed beyond examination. It also established the posture in which the case would proceed to judgment. The merits were now unavoidable. The court that received the petition back would have to engage with the claim that a provision of the Indian Penal Code impeded health work and infringed four constitutional articles, and it would have to do so by reasoning rather than by threshold avoidance.

Three years separated the remittal from the judgment, and the judgment, when it arrived, altered the legal position in Delhi and supplied the reasoning against which every later round would be measured. It did so neither by inventing a new provision nor by erasing the old one from the statute book. It read the provision down, a technique whose restraint and whose radicalism are both easy to miss. To read a provision down is to preserve its words while confining their constitutional reach, to hold that a section survives only within limits that the Constitution permits. The result can look like continuity, because the text remains in place. In operation it is a profound change, because conduct that appeared to fall within the prohibition is placed outside it by judicial interpretation.

What did the Delhi High Court decide in the Naz Foundation case in 2009?

On 2 July 2009, the Delhi High Court held Section 377 unconstitutional insofar as it criminalised consensual sexual acts between adults in private, read sexual orientation into Article 15, and preserved the provision for non-consensual acts and acts involving minors.

The bench that delivered that holding was a Division Bench comprising Chief Justice A P Shah and Justice S Muralidhar. The date, 2 July 2009, marks the point at which the argument first advanced in 2001 received a full constitutional answer. The core of the answer lay in the qualifying phrase that ran through the holding, insofar as it criminalised consensual sexual acts between adults in private. Every word in that qualification performed work. Consensual separated acts to which the parties agreed from acts imposed on another. Adults separated those with the legal capacity that adulthood confers from minors, whose protection the court expressly left intact. Private separated the intimate sphere from conduct that the law might address on other grounds. Taken together, the three qualifications drew a boundary not around a class of persons but around a sphere of conduct, and held that within that sphere the criminal law, as embodied in Section 377, could not constitutionally enter.

The distinction between striking down and reading down shaped the reach of the judgment. The court held Section 377 unconstitutional insofar as it criminalised the specified acts. It did not declare the entire provision void. That choice preserved a functioning offence for the cases the court identified, namely non-consensual acts and acts involving minors. The preservation was not an incidental aside in the judgment. It was part of its logic. By expressly preserving the provision for those cases, the court answered in advance the objection that decriminalisation of consensual adult conduct in private would leave a gap in the protection of those who do not or cannot consent. The judgment thus presented itself not as an abolition of a safeguard but as a separation of two things the provision had bundled together, consensual intimacy between adults and coercive or exploitative acts, and it assigned constitutional protection to the first while leaving criminal liability in place for the second.

Equally consequential was the treatment of Article 15. The court read the word sex in Article 15 to include sexual orientation. The move was interpretive rather than textual, and its importance lies in what it recognised about the structure of discrimination. To read sexual orientation within sex is to say that a ground enumerated in the Constitution is not confined to the narrowest biological reading of its terms, and that disadvantage imposed because of orientation is disadvantage imposed on a ground the Constitution already names. That reading located the claim within the existing fabric of the equality code instead of placing it outside that fabric as a request for a new and unenumerated protection. It answered, within the logic of Article 15 itself, the argument that orientation stood apart from the concerns of the non-discrimination guarantee. After Naz Foundation, at least as a matter of the Delhi High Court reasoning, orientation was not an addition to Article 15. It was within the meaning of a word already there.

A further observation in the judgment has been overlooked beside its central holding, although it anticipated the terms of the later debate. The court observed that the provision could in any event be amended by Parliament. That observation performed two functions at once. It acknowledged the ordinary legislative power over a provision of the Indian Penal Code, making clear that judicial reading down did not occupy the field or deny the legislature its role. It also placed responsibility in perspective. If Parliament considered that the boundary drawn by the court was wrong, or that a different settlement was preferable, the route of amendment lay open. The judgment therefore did not present judicial intervention and legislative choice as opposites. It presented the judicial conclusion as the constitutional position unless and until the legislature chose, within constitutional limits, to legislate. That allocation of roles would become central in the second courtroom round, where the relationship between court and Parliament was not a background observation but the decisive ground of decision.

It is also worth holding in view what the judgment did not do, because the limits of Naz Foundation explain part of what followed. The judgment was that of a High Court. Its reading of the provision was authoritative within its sphere, yet it remained open to appeal to the Supreme Court, and appeals were in fact brought. The judgment did not amend the text of the Indian Penal Code. The words of Section 377 remained as Parliament had enacted them, confined by interpretation rather than altered by legislation. A settlement that depends on interpretation depends in turn on the interpretation surviving appellate scrutiny. The years between 2009 and 2013 were therefore not a period in which the question had been closed by the legislature or removed from contention. They were a period in which a judicial answer stood while remaining, by the ordinary structure of the legal system, provisional against the decision of a higher court.

Suresh Kumar Koushal: the Supreme Court restores the provision

Appeals by Suresh Kumar Koushal and others reached the Supreme Court, and on 11 December 2013 a two-judge bench of Justice G S Singhvi and Justice S J Mukhopadhaya reversed the High Court and upheld Section 377 in full. The reversal was complete in its operative effect. The reading down was undone. The provision that the Delhi High Court had confined to non-consensual acts and acts involving minors was restored to the full scope it had possessed before the judgment of 2009. Consensual sexual acts between adults in private, which for a period had stood outside the provision as a matter of constitutional interpretation, were brought back within its reach. To understand the reversal, it is necessary to attend not only to its conclusion but to the three strands of reasoning by which the bench supported it, because each strand answered a corresponding strand in the reasoning it displaced.

The first strand concerned the object of the provision. The bench reasoned that the provision criminalised acts and not any identity or orientation. That distinction carried the weight of the answer to the discrimination claim. If Section 377 attached criminal liability to acts, and if any person performing those acts was equally liable, then on this reasoning the provision did not single out a class of persons for disadvantage. The Delhi High Court had approached the matter through the effect of the provision on a sphere of consensual adult conduct and through the reading of sexual orientation into Article 15. The Supreme Court approached it through the formal object of the offence. The contrast between those approaches is the analytical heart of Koushal. One court asked what the provision did to the lives and liberty of those whose consensual conduct it captured. The other asked what the provision, on its face, prohibited, and answered that it prohibited acts open to anyone to commit. Much in Indian equality jurisprudence turns on precisely that choice between an inquiry into effect and an inquiry into form, and Koushal is a stark illustration of how the choice of inquiry can determine the outcome before any further question is reached.

The second strand concerned misuse. The bench reasoned that the provision had rarely been shown to be misused by the standard the Court required. That formulation deserves careful parsing, because it set the terms on which evidence of harm could count. The Naz Foundation petition had been founded on an impediment to health work, a harm that operated through the existence and threat of the provision rather than through a catalogue of prosecutions. Koushal directed attention instead to demonstrated misuse established to the standard the Court required. On that standard, the rarity of shown misuse weakened the case against the provision. The reasoning illustrates a broader question about how constitutional courts should treat laws whose principal effect may be deterrent or stigmatising rather than prosecutorial. If proof of frequent misuse is demanded, a provision that chills conduct and drives it from view may generate little of the very evidence required to condemn it, precisely because its success in suppressing visibility suppresses the record of its operation. The bench did not accept that logic as a ground for intervention. It treated the paucity of shown misuse as a reason for leaving the provision in place, and in doing so it placed a heavy evidential burden on those who challenged a law by reference to its surrounding effects.

Within that reasoning the bench wrote a sentence that came to define the judgment in public and legal memory alike, that a minuscule fraction of the country’s population constitutes lesbians, gays, bisexuals or transgenders. The sentence must be considered for the work it did within the judgment and not only for its tone. It introduced the size of the affected population into a constitutional analysis of rights. That introduction sits uneasily with the premise of fundamental rights, which are ordinarily understood to protect the individual and the minority precisely against the weight of numbers. A right that depended for its enforcement on the population claiming it reaching a sufficient proportion of the whole would be a right of an unusual and diminished kind. The sentence therefore exposed, in a single line, the distance between Koushal and the tradition in which constitutional protection is at its most necessary when the group seeking it is small, unpopular or both. The bench nevertheless included the observation as part of its reasoning, and its inclusion signalled that the smallness of the population counted, in the Court’s assessment, against rather than for close judicial attention.

The third strand concerned institutional role. The bench held that whether the provision should be amended or deleted was for Parliament, not the courts. That holding answered the observation in Naz Foundation that Parliament could amend the provision in any event. Where the Delhi High Court had treated legislative power as a background fact compatible with judicial reading down, the Supreme Court treated it as a reason for judicial abstention on the central question. The provision was a creature of legislation. Its removal or alteration, on this view, belonged to the legislative process, with its capacity for debate, consultation and political accountability. The courts, by contrast, should not perform by interpretation what amounted in effect to a legislative revision. That argument has a recognised place in constitutional adjudication, and it would be a mistake to treat it as mere evasion. Questions about the proper boundary between judicial review and legislative choice are genuine questions. Yet in Koushal the deference operated in a particular direction. It preserved a provision that a High Court had found unconstitutional in a specified respect, and it did so while the affected population had been described in the same judgment as a minuscule fraction of the whole, a description that bears directly on the capacity of that population to secure legislative attention through ordinary political weight.

The sequence after judgment showed that the institutional question and the merits had not been separated as cleanly as the reasoning suggested. Review petitions against Koushal were dismissed in January 2014. A review is the ordinary means by which a judgment of the Supreme Court may be reconsidered by the Court itself, and its dismissal closed that means within weeks of the judgment under challenge. Curative petitions followed. On 2 February 2016 a three-judge bench, comprising Chief Justice T S Thakur with Justices Anil R Dave and J S Khehar, referred the curative petitions to a Constitution Bench. The referral is the significant fact in that sequence. A court that considered the matter conclusively settled by the dismissal of review need not have taken that step. By referring the curative petitions to a Constitution Bench, the Supreme Court placed the continuing challenge before a formation of the gravity reserved for substantial constitutional questions, and it thereby acknowledged, in institutional terms, that Koushal had not ended the judicial story. The provision stood in full force. The judgment of 2013 remained the operative law. Yet the door that Koushal had sought to close by directing the question to Parliament had been reopened, if only to the extent of a referral, by the Court itself.

The legislative door stays shut

The legislative route was tested directly, and the test came while the judicial route remained in the uncertain state described above. On 18 December 2015 Congress MP Shashi Tharoor moved to introduce a private member’s bill in the Lok Sabha that would have substituted Section 377 so that only non-consensual acts remained punishable. The content of that proposal repays attention, because it corresponded closely to the boundary the Delhi High Court had drawn in 2009. The High Court had preserved the provision for non-consensual acts and acts involving minors while removing consensual acts between adults in private from its reach. The bill would have achieved a cognate result by legislation rather than by interpretation, substituting a provision under which only non-consensual acts remained punishable. In the terms set by Koushal, this was exactly the kind of step the Supreme Court had said belonged to Parliament. A member of Parliament brought forward a text to take that step. The proposal therefore cannot be characterised as an attempt to circumvent the judgment of 2013. It was an attempt to act upon the judgment’s own allocation of responsibility.

The attempt failed at the earliest possible stage. The motion for leave to introduce the bill was defeated by 71 votes to 24, with one abstention, so the bill was never even introduced. The numbers should be read for what they establish and for what they do not. They establish that opposition at the threshold was decisive, nearly three votes against for every vote in favour among those who voted on the motion as recorded, with a single abstention standing apart from both. They do not establish the considered view of the Lok Sabha on the merits of a substituted Section 377 after debate, because no such debate on the bill took place. A defeat at the stage of leave to introduce prevents the text from entering the legislative process at all. The arguments for substitution were therefore not tested against amendment, scrutiny or a vote on the substance. The door was not closed after the proposal had been examined within the House. It was closed before the proposal could be placed formally before the House for examination.

A second attempt followed in March 2016 and was defeated by 58 votes to 14. The margin, expressed as a proportion, was wider than that of December 2015, and the number voting in favour was smaller. Repetition matters in assessing a legislative record. A single defeat at leave stage might be attributed to the circumstances of a day, to attendance, or to the ordering of business. A second defeat on a comparable attempt within a few months presents a pattern rather than an accident. Together the two votes formed the legislative record on which any assessment of the parliamentary route had to rest during this period. On that record, substitution of Section 377 to confine punishment to non-consensual acts could not command even the preliminary support required for introduction in the Lok Sabha. The provision was not retained after a substantive parliamentary judgment that it should be retained in its restored form. It was retained through the repeated refusal of the House to admit a proposal for its reform.

Against that record stood a notable public statement from within the executive of the same period. Finance Minister Arun Jaitley had publicly said in that period that the Delhi High Court’s view appeared more acceptable and that the Supreme Court had taken a conservative view. The statement is significant precisely because of its limits. It expressed a view about the relative acceptability of the two judicial positions, favouring the reasoning of 2009 over the reasoning of 2013, and it came from a senior minister. It did not introduce a bill. It did not commit the government to legislation. It did not alter a single vote recorded in the Lok Sabha on either attempt at introduction. Its importance lies in the contrast it reveals between expressed ministerial sympathy for the Delhi High Court position and the absence of any legislative movement capable of giving that sympathy effect. A legal system in which a Finance Minister can describe a High Court view as the more acceptable one while the provision that view confined remains in full force is a system in which the allocation of responsibility described in Koushal has produced a gap rather than a settlement.

That gap defined the close of this phase. The Delhi High Court had drawn a boundary founded on consent, adulthood and privacy, and had located sexual orientation within Article 15. The Supreme Court had removed that boundary, resting on the distinction between acts and identity, on the rarity of misuse shown to its required standard, on the size of the affected population, and on the assignment of amendment to Parliament. Parliament declined twice to admit the proposal that would have performed that assignment. The referral of the curative petitions to a Constitution Bench on 2 February 2016 therefore acquired weight beyond its procedural character, as the one movement towards further examination in a period of legislative refusal.

The ground moves: gender identity and privacy

The challenge to Section 377 that reached a Constitution Bench in 2018 did not arrive on ground that had stood still since Suresh Kumar Koushal. Between 2014 and 2018 two judgments of the Supreme Court altered the terms in which questions of identity, intimacy and constitutional protection could be argued. Neither judgment decided the validity of Section 377. Each judgment nevertheless changed what a court confronting Section 377 would have to answer, and each did so by locating protection in the person rather than in the tolerance of the majority. The first, National Legal Services Authority v Union of India, concerned gender identity and the constitutional position of transgender persons. The second, Justice K S Puttaswamy (Retired) v Union of India, concerned privacy and the place of that right within Part III of the Constitution. Read together, the two decisions supplied distinct but converging lines of reasoning. One spoke of who a person is and of the right to have that identity recognised on the terms the person affirms. The other spoke of the space in which a person forms and lives intimate choices, and of why that space is not held at the pleasure of numerical majorities. The petitions that followed were framed within that altered landscape.

National Legal Services Authority v Union of India, known as NALSA, was decided on 15 April 2014 by a two judge bench of Justice K S Radhakrishnan and Justice A K Sikri. The Court recognised transgender persons as a third gender and affirmed self identification of gender. The significance of the holding lay not only in the category that was recognised but also in the manner of recognition. Identity was not treated as a status to be conferred after external verification of the body. It was treated as a matter that the individual could identify for himself or herself, with the law required to give effect to that identification. A fuller account of the NALSA judgment and its recognition of transgender rights shows how that affirmation altered the starting point for later argument, because the person was placed before the classification. The State was not asked to create an identity out of administrative convenience. The State was required to recognise an identity that the person already held and expressed, and to organise its legal response around that recognition rather than around a refusal to see the person at all.

Central to the reasoning in NALSA was the separation of gender identity from biological sex. The judgment treated the two as distinct, and in doing so it refused the assumption that legal identity must follow anatomy in a fixed and unexamined way. That separation carried analytical weight well beyond its immediate context. If identity is not reducible to biological fact as recorded at birth, then a legal order that insists on that reduction must explain itself in constitutional terms rather than rely on the claim that it merely records nature. The explanation must address the person who experiences the mismatch between an assigned category and a lived identity, and it must do so with reasons that survive equality scrutiny. NALSA also affirmed that gender identity lies at the core of Articles 14, 15, 19 and 21, and it treated discrimination on grounds of gender identity as constitutionally suspect. The placement mattered. Gender identity was not presented as a peripheral social concern to be handled with sympathy when administrators chose to show it. It was placed inside the operative provisions through which equality before the law, prohibition of discrimination, freedoms of speech and expression and associated freedoms, and protection of life and personal liberty are secured.

Each of those articles contributed a different strand, and the judgment held the strands together rather than resting on any one alone. Article 14 directed attention to equal treatment, Article 15 to discrimination, Article 19 to expression, because gender is presented and lived in the world as well as held inwardly, and Article 21 to life and personal liberty in a sense reaching beyond survival to dignity and self definition. By holding that gender identity lies at the core of all four provisions, NALSA made it difficult thereafter to argue that identity was a matter of administrative grace. Protection was expressed in the language of rights rather than welfare, so exclusion could be framed as a rights violation demanding justification, rather than as a policy disappointment inviting deference.

The reach of NALSA for the Section 377 question was indirect, and that indirectness deserves precise statement. The judgment concerned transgender persons and gender identity. Section 377, as it was challenged in the later petitions, concerned consensual sexual acts between adults and the criminal law that attached to them. The two subjects are not the same, and careful argument did not pretend that they were. What NALSA supplied was a method and a premise. The method was to begin with the person and with self defined identity, and then to test state action against Articles 14, 15, 19 and 21 in combination. The premise was that a classification resting on a core aspect of identity attracts serious constitutional attention, and that discrimination tied to that aspect is suspect rather than routine. Once that method and premise stood as authority, an argument that treated sexual orientation as a trivial or marginal characteristic, deserving of protection only if large numbers claimed it, faced a body of reasoning that pointed in the opposite direction. NALSA had already insisted that dignity attaches to identity as lived and affirmed. A later court would have to explain why that insistence stopped at the boundary of orientation, or else carry the logic forward and confront the criminal provision on its merits.

If NALSA reordered the treatment of identity, Puttaswamy reordered the treatment of privacy, and with it the treatment of intimate life. Justice K S Puttaswamy (Retired) v Union of India was decided on 24 August 2017 by a unanimous nine judge bench. The Court held privacy to be a fundamental right protected under Article 21 and the freedoms in Part III. The unanimity and the size of the bench gave the holding a weight that no later bench dealing with Section 377 could set aside as a passing observation. Privacy was not left as an implied penumbra to be invoked cautiously and discarded when it proved inconvenient. It was affirmed as a fundamental right with a textual home in Article 21 and with support across the freedoms that Part III guarantees. That anchoring mattered for the structure of later argument. A claimant no longer had to present privacy as a foreign import or as a judicial invention without roots in the constitutional scheme. The claimant could point to a nine judge affirmation that privacy belongs within the scheme itself, and could then ask what follows when a criminal provision reaches into the intimate decisions that privacy protects.

Within that larger affirmation, the plurality opinion of Justice D Y Chandrachud, writing for four judges, addressed sexual orientation in direct terms. In that opinion the Court’s recognition that privacy, including sexual orientation, is constitutionally protected was expressed through the holding that sexual orientation is an essential aspect of privacy, dignity and identity. The formulation linked three concepts that are sometimes treated as separate. Privacy supplied the protected space. Dignity supplied the reason the space matters, because choices made within it are part of how a person lives with self respect. Identity supplied the connection between the choice and the person, because orientation was described not as a detachable act performed in isolation but as an aspect of who the person is. That linkage changed the question that Section 377 presented. The question could no longer be confined to whether the State disapproved of particular acts in the abstract. It had to include whether the State could use the criminal law against conduct that expresses an essential aspect of privacy, dignity and identity, and what justification would be required for such use.

The plurality also confronted the reasoning by numbers that had appeared in Koushal. It expressly rejected the reasoning that rights could be measured by numbers, writing that the Koushal observation that “a minuscule fraction of the country’s population constitutes lesbians, gays, bisexuals or transgenders” was not a sustainable basis to deny the right to privacy, because the purpose of fundamental rights is to insulate their exercise from the disdain of majorities. The sentence performed two operations at once. First, it removed population size as a threshold test for protection, so that a claimant would not have to prove numerical significance before the Constitution would listen. Second, it restated the function of fundamental rights as insulation against majoritarian disdain, so that hostility or contempt in the wider population could not serve as the measure of a right and could not supply its justification either. For Section 377, where disapproval of the persons affected had long shadowed discussion of the provision even when the formal argument was cast in other terms, that restatement had evident force. It required the provision to be defended, if it was to be defended at all, on grounds other than the smallness of the group or the distaste of the larger society.

How did the Puttaswamy privacy judgment change the Section 377 question?

Puttaswamy recognised privacy as a fundamental right and treated sexual orientation as an essential aspect of privacy, dignity and identity. It rejected the view that a small population could be denied protection for that reason, so Section 377 could no longer rest on majoritarian disapproval alone and had to answer a privacy claim grounded in Part III.

The deeper change lay in the order of reasoning that Puttaswamy made necessary. Before that judgment, a defence of Section 377 could proceed as if the burden lay on those affected to show why they deserved an exception from a general criminal prohibition, and as if the answer could turn on how many persons were affected or on how the majority regarded them. After Puttaswamy, that order was difficult to sustain. Privacy stood affirmed as a fundamental right. Sexual orientation stood described, in an opinion speaking for four judges of a nine judge bench, as an essential aspect of privacy, dignity and identity. The burden of justification therefore moved. The State would have to explain how a provision that criminalised consensual intimacy between adults could coexist with a fundamental right so described, rather than the individual having to explain why intimate life deserved constitutional notice at all. That shift did not decide the outcome in advance, because a fundamental right can still be subject to reasoned limits supported by legitimate aims and proportionate means. It did decide the seriousness of the inquiry. The inquiry would proceed as a fundamental rights inquiry, with the provision measured against the right, and not as a request for indulgence measured against majoritarian sentiment.

The treatment of majoritarian disdain requires separate attention. Courts are sometimes invited to treat social acceptance as a precondition for constitutional protection, as if a right becomes real only when its exercise no longer offends prevailing opinion. The plurality formulation rejected that sequence. If the purpose of fundamental rights is to insulate their exercise from the disdain of majorities, then disdain is the condition against which protection operates, not a reason to withhold protection until it fades. A person does not hold a fundamental right only in hiding. Insulation means that the exercise of the right is not put to a vote and is not suspended because a larger group regards the holder with contempt. Applied to Section 377, moral disapproval held by many, without more, could not complete the justification. Something else, grounded in a constitutionally permissible aim, would have to be shown.

At the same time, the limits of what Puttaswamy decided were stated with care inside the judgment itself, and those limits shaped the procedural road that remained. The plurality did not itself overrule Koushal, noting the matter stood referred in curative proceedings. That restraint mattered. Puttaswamy supplied reasoning that undermined a central strand of Koushal, namely the reliance on the smallness of the affected population as a reason to deny protection, and it affirmed a right under which the Section 377 challenge could be powerfully framed. It did not, however, pronounce the final word on Section 377 or strike the provision down. The formal authority of Koushal on the provision therefore remained to be dealt with by a bench competent to reconsider it. This distinction between reasoning that transforms the argument and a holding that disposes of the earlier decision explains why fresh proceedings and a reference were still necessary after August 2017. The ground had moved, but the earlier decision still stood on the record and still required reconsideration through the discipline of bench strength and precedent. The petitioners who moved the Court in that interval understood the distinction and framed their request accordingly, seeking reconsideration rather than pretending that Puttaswamy had already delivered the result.

Taken together, NALSA and Puttaswamy also changed how the Articles could be pleaded in combination. NALSA had placed gender identity at the core of Articles 14, 15, 19 and 21 together, resisting any suggestion that a claim must choose a single article and live or die within it. Puttaswamy had located privacy under Article 21 and the freedoms in Part III, again treating Part III as a connected scheme rather than as a set of sealed compartments. For the Section 377 challenge, that connected reading opened a fuller frame. Equality argument could address the classification and its effects. Discrimination argument could address the burden borne on grounds bound up with identity. Expression argument could address the freedom to live and communicate a central aspect of the self, a strand that the later reference order would single out through Article 19(1)(a). Liberty and privacy argument could address the intimate sphere into which the criminal law entered. None of these strands alone captures the whole complaint against a provision that brands consensual adult intimacy as crime, with the stigma, fear and exposure to coercion that criminal branding carries in ordinary life. Together, they allowed the complaint to be stated at the level at which it was experienced, as a burden on identity, equality, expression and intimate freedom at once.

Fresh petitions and the reference to a Constitution Bench

Fresh writ petitions were filed in 2016 by Navtej Singh Johar, a dancer, together with Sunil Mehra, Ritu Dalmia, Ayesha Kapur and Aman Nath. The petitions placed before the Court a direct challenge brought by named individuals who identified themselves and stated, in their own names, that the criminal law burdened their lives and their intimate choices. That form of challenge carried its own significance. The petitioners did not proceed through anonymity or through an organisation speaking at a distance from the persons affected. They presented themselves as citizens claiming fundamental rights that Puttaswamy would soon describe in terms of privacy, dignity and identity, and that NALSA had already described in terms of self identification and protection against discrimination grounded in identity. The lead petitioner’s occupation as a dancer was part of the public record of the case, and it underlined, without any need for embellishment, that the claim arose from lived lives in the open, in work, in art and in association, and not from an abstract seminar question about the limits of criminal law. The petitions asked the Court to confront Section 377 as it bore on real persons who were prepared to be named in the proceedings that challenged it.

Further petitions were filed by other petitioners, including the Humsafar Trust’s Ashok Row Kavi and the transgender activist Akkai Padmashali, and a petition was filed by alumni of the Indian Institutes of Technology. The breadth of that petitioner group mattered for how the challenge was understood, even though breadth, like numbers, cannot by itself establish a right after the reasoning in Puttaswamy. The petitions did not rest on a single voice or a single form of experience. An activist associated with the Humsafar Trust brought the perspective of organised work with affected communities over a long period. A transgender activist brought the perspective that NALSA had placed at the centre of constitutional attention, where gender identity and protection from discrimination were concerned. Alumni of the Indian Institutes of Technology brought the perspective of former students of public institutions of higher education, speaking as citizens who considered the provision incompatible with the constitutional order under which they had studied and worked. The plurality of petitioners did not convert the case into a campaign decided by accumulation. It showed instead that the burden of Section 377 was experienced across different lives and callings, and that the constitutional question could not credibly be dismissed as the concern of an isolated individual with an idiosyncratic grievance.

On 8 January 2018 a three judge bench headed by Chief Justice Dipak Misra referred the petitions to a Constitution Bench, observing that the decision in Koushal required reconsideration and noting that the challenge raised an aspect of Article 19(1)(a) which Koushal had not examined. Each element of that order performed distinct work. The observation that Koushal required reconsideration opened the formal path that Puttaswamy, for reasons of discipline, had left open, since the plurality there had noted the curative reference rather than overruling Koushal itself. The constitution of a larger bench supplied the institutional means by which an earlier decision of the Court could be revisited with the authority that the doctrine of precedent demands. The reference to Article 19(1)(a) widened the frame beyond the provisions that had dominated earlier discussion. Article 19(1)(a) protects freedom of speech and expression, and the referring bench noted that the challenge raised an aspect of that freedom which Koushal had not examined. That noting was significant because it identified unfinished business in the earlier judgment on its own terms. Reconsideration was not presented merely as disagreement with a prior outcome. It was presented also as the completion of an inquiry that had left a recognised freedom unaddressed, even though expression is bound up with identity, with the ability to speak of one’s life, and with the ability to live without concealing a central part of the self under threat of criminal penalty.

During the July 2018 hearings the Union government filed an affidavit leaving the question of the constitutionality of Section 377, insofar as it concerned consensual sexual acts between adults, to the wisdom of the Court, while asking that the hearing not expand into questions such as marriage. The affidavit defined the posture of the Union with a precision that shaped the hearing. On the central question before the Court, the constitutionality of the provision as it applied to consensual sexual acts between adults, the Union did not present a defence of the provision and instead left the question to the wisdom of the Court. That posture removed the adversarial support that a government defence would ordinarily supply for a provision under challenge, and it left any defence to be drawn from other participants or from the reasoning of Koushal itself, whose authority was under reconsideration. At the same time, the affidavit drew a boundary around the scope of the hearing. The Union asked that the hearing not expand into questions such as marriage. The request reflected a distinction between the validity of a criminal prohibition on consensual adult intimacy and the separate legal recognition of relationships through institutions such as marriage, with their own statutory schemes and their own questions. The Court was therefore invited to decide the criminal law question that the petitions presented, without converting the hearing into a general determination of every legal consequence that might follow for relationship recognition.

That combination clarified the responsibility resting on the Constitution Bench. The bench could not treat the Union posture as an answer, because validity under Part III must be determined by constitutional reasoning and not by the presence or absence of a defence. What the affidavit established was the ground of contest. The provision, as applied to consensual sexual acts between adults, stood to be judged against articles that NALSA and Puttaswamy had illuminated, with the numbers based reasoning of Koushal undermined and the expression aspect under Article 19(1)(a) flagged as unexamined. What remained was the decision itself, on whether a criminal provision directed at consensual adult intimacy could survive the privacy, dignity, identity, equality and expression commitments the Court had restated at the highest level of authority.

Navtej Singh Johar: four opinions, one result

Navtej Singh Johar v Union of India was decided on 6 September 2018 by a Constitution Bench of five judges, and its citation, (2018) 10 SCC 1, records a decision whose formal structure is as instructive as its result. The Bench consisted of Chief Justice Dipak Misra and Justices R F Nariman, A M Khanwilkar, D Y Chandrachud and Indu Malhotra. The decision was unanimous in outcome, yet it was delivered through four opinions rather than through a single judgment speaking for the whole Court. Chief Justice Misra wrote for himself and for Justice Khanwilkar. Justices Nariman, Chandrachud and Malhotra each wrote a separate concurring opinion. That arrangement matters for the way the judgment should be read. Unanimity settled the operative question beyond argument, while plurality left future readers with four distinct routes to the same destination, each route emphasising a different strand of constitutional reasoning and each route available to be drawn upon when the principles are tested in later disputes.

The unanimity also carries a particular significance against the background that the judgment displaced. The operative conclusion of the whole Bench overruled Suresh Kumar Koushal, the earlier decision that had left Section 377 intact in its full breadth. A divided Bench overruling an earlier authority invites argument about which strand commanded support and which remained contested. A unanimous Bench speaking through four concurring opinions leaves no such opening on the central point. None of the judges was prepared to sustain the criminalisation of consensual sexual acts between adults in private, and none treated the surviving applications of the provision as a reason to preserve the part that failed constitutional scrutiny.

Chief Justice Misra, writing also for Justice Khanwilkar, supplied the opinion that most directly framed the moral contest at the heart of the case. His opinion placed constitutional morality above social morality. Social morality, in the sense used in the judgment, is the morality of prevailing sentiment, the attitudes that a majority in society happens to hold about how others should live. Constitutional morality is a different discipline. It is the morality written into the Constitution itself, expressed through its commitments to equality, liberty and dignity, and binding upon courts precisely when prevailing sentiment pulls in the opposite direction. If constitutional adjudication merely ratified social morality, fundamental rights would become permissions granted by popular tolerance rather than entitlements held against the State. The Chief Justice treated identity and dignity as central to Article 21, and his opinion used the line, attributed to the German thinker Johann Wolfgang von Goethe, “I am what I am, so take me as I am”. The sentence states the claim of the individual as a demand to be recognised as one already is, rather than as a plea to be remade into a form that others find acceptable.

That framing shaped the practical directions with which the Chief Justice closed his opinion. He directed that the judgment receive wide publicity and that public officials, including police personnel, be sensitised. The direction acknowledges that a provision such as Section 377 operates not only through prosecutions that reach a courtroom. It operates through the knowledge, held by officials and by members of the public alike, that the provision exists and can be invoked. Wide publicity made the change of legal position a matter of open knowledge rather than specialist information. Sensitisation of public officials, expressly including police personnel, addressed the institutional side, recognising that dignity proclaimed in a judgment but denied at a police station has been recognised in form and withheld in substance.

Justice Nariman brought a different temperament to the same conclusion, and his opinion is the most sharply analytical of the four. He applied the manifest arbitrariness test under Article 14, a test developed in Shayara Bano, and the choice of that test determined the character of his reasoning. Manifest arbitrariness asks whether a law is capricious, irrational or without an adequate determining principle, and it permits a court to test the substance of legislation against Article 14 directly. Applied to Section 377, the test directed attention to the words of the provision itself, and Justice Nariman attacked the undefined phrase “order of nature” at its centre. A criminal prohibition must tell those whom it governs what conduct is forbidden, with sufficient clarity that a person can order his or her conduct accordingly and that an official cannot expand or contract the prohibition at will. The phrase “order of nature” performed no such service. It carried no definition in the provision, it drew its apparent meaning from assumption and convention rather than from any stated legal standard, and yet upon that phrase depended the boundary between lawful intimacy and serious criminal liability. A prohibition so constructed fails at the first task of criminal law, which is to mark the forbidden with precision, and it fails under Article 14 because a standard that cannot be stated cannot be applied evenly.

Justice Nariman also recorded that homosexuality is not a mental disorder, pointing to the Mental Healthcare Act, 2017 and to the international classification of diseases. The point performs precise legal work. Criminalisation of a class of persons frequently rests, whether openly or by implication, upon a premise that the defining characteristic of the class is a defect requiring correction, containment or cure. If that premise is false, a justification built upon it falls with it. By recording the position taken in the Mental Healthcare Act, 2017, a statute enacted by Parliament itself, Justice Nariman showed that the legal order had already abandoned the premise in one branch of its own law while retaining a criminal provision that presupposed it in another. By pointing also to the international classification of diseases, he placed that statutory position within a wider professional recognition to the same effect. Where the Chief Justice reasoned from the moral commitments of the Constitution, Justice Nariman reasoned from coherence, definition and the internal consistency of the legal order, and he found the provision wanting on each of those measures.

Justice Chandrachud wrote the opinion that most fully joined the doctrinal analysis to the lived operation of the provision over time. He treated the choice of a partner and intimate association as protected choices under Articles 14, 19 and 21, reading the three provisions together rather than assigning the case to one of them in isolation. That joint reading reflects the structure of the harm itself. A law that criminalises consensual adult intimacy in private discriminates in the distribution of legal freedom, and so engages Article 14. It burdens the freedom to express identity and association, and so engages Article 19. It invades the sphere of personal liberty, privacy and dignity within which decisions about intimacy and companionship are made, and so engages Article 21. Justice Chandrachud also traced the history of the provision and its use in harassing the community. The provision could not be evaluated honestly as an abstract text alone. Its meaning in practice included the leverage it gave to those who wished to threaten, blackmail, expose or intimidate, and a court assessing its validity was required to weigh the law as it operated and not only the law as it was drafted. He held that members of the community are entitled to the full protection of the Constitution as equal citizens. The formulation is deliberately unqualified. It does not speak of tolerance, of indulgence or of a limited exemption carved out of a general prohibition. It speaks of full protection held by equal citizens by right, and it locates the community inside the ordinary guarantees of the Constitution rather than at their margin.

Justice Indu Malhotra, the fourth author, wrote the strand of the judgment remembered above all for a single sentence, which should be given exactly as she wrote it: “History owes an apology to the members of this community and their families, for the delay in providing redressal for the ignominy and ostracism that they have suffered through the centuries.” The sentence is not a doctrinal proposition and it does not pretend to be one. Its function is acknowledgment. A judgment can correct the law, but correction alone leaves unspoken the fact that the law stood for a long period and that real persons bore the consequences of its standing. Justice Malhotra refused to leave that fact unspoken. The apology is owed by history rather than demanded from any individual litigant before the Court, and its subjects include families as well as members of the community themselves, a recognition that ostracism radiates outward from the person directly targeted to those bound to that person by affection and household. The reference to delay is equally deliberate. Redressal, when it arrives after a long interval, does not arrive as though the interval had not occurred. Read alongside the other three opinions, her concurrence completes the register of the judgment. The Bench reasoned, defined, traced and declared, and through her opinion it also apologised.

Turning from the individual opinions to the doctrines that run across them, Article 14 supplies the most formal ground of invalidation, and the manifest arbitrariness strand deserves to be understood on its own terms. Article 14 guarantees equality before the law and the equal protection of the laws. The manifest arbitrariness test applied by Justice Nariman, developed in Shayara Bano, asks whether a law is capricious, irrational or without an adequate determining principle, and it permits the substance of legislation to be tested against Article 14 directly. Section 377, in its application to consensual acts between adults in private, was vulnerable under that inquiry at several points at once. Its central phrase was undefined, as Justice Nariman showed. Its prohibition made no distinction, within the same textual sweep, between acts involving consent and acts involving coercion, treating the feature on which criminal responsibility ordinarily turns as though it were legally irrelevant. And its burden fell upon a class defined by an innate characteristic, so that the prohibition operated in practice as a penalty upon identity expressed through intimacy rather than as a regulation of harmful conduct. A law that cannot state its own standard and that selects its subjects by who they are is a law without an adequate determining principle, and that is the core of the Article 14 holding.

Article 15 carries a distinct and in some respects a bolder proposition. The article forbids discrimination by the State on grounds only of religion, race, caste, sex, place of birth or any of them. The judgment read the ground of sex to include sexual orientation. A narrow reading would confine each listed ground to the most obvious instances known when the text was adopted, and would leave every other form of discrimination to be pursued, if at all, under the general language of Article 14 alone. The reading adopted in Navtej Singh Johar treats the listed grounds as expressions of a principle rather than as a closed inventory of historical examples. Sex, understood in that way, encompasses the attributes and orientations through which sexuality organises a person’s intimate life, and discrimination visited upon a person because of sexual orientation is discrimination that operates through sex in a real and not a merely verbal sense. The consequence is practical as well as symbolic. Article 15 is a specific and emphatic prohibition, and placing sexual orientation within it denies the State the argument that such discrimination is only an ordinary inequality to be justified by ordinary reasons. It also aligns the discrimination analysis with the dignity analysis elsewhere in the judgment, since both proceed from the same premise, that sexual orientation is an intrinsic aspect of the person rather than a disqualifying deviation from a norm that the law is entitled to enforce.

Article 19 contributes the strand concerned with expression, and its central concept is the chilling effect. The freedom of speech and expression under Article 19(1)(a) is burdened not only when a person is prosecuted for something said. It is burdened when the existence of a criminal prohibition discourages expression that the prohibition does not formally target, because the person who would speak cannot be certain where expression ends and liability begins. Section 377 produced that burden in an acute form. To speak openly about one’s sexual orientation, to associate visibly with a community organised around it, to write, to gather and to seek companionship without concealment, each of these expressive and associative acts risked drawing attention that could be converted, by a hostile official or a hostile acquaintance, into a threat of prosecution under a provision whose central phrase nobody could define with confidence. The rational response to such a legal environment is silence and concealment, and that response is precisely the injury that the doctrine of the chilling effect exists to identify. A prohibition that hangs over a community, seldom carried to judgment but always available to be invoked, achieves through apprehension what it might never achieve through trial. Justice Chandrachud’s treatment of intimate association as a protected choice under Article 19, alongside Articles 14 and 21, captures this dimension. Association and expression are the outward forms of an identity that privacy protects inwardly, and a constitution that secured the inner sphere while abandoning the outer would protect the right to be in private while denying the right to appear in public as the person one is.

Article 21 is the centre of gravity of the judgment as a whole, and the opinions converge upon it from their different directions. The article provides that no person shall be deprived of life or personal liberty except according to procedure established by law, and the judgment drew from it a connected set of protections. Dignity stands first among them, not as an ornament of the reasoning but as its foundation, since the opinions treated the equal citizenship of the community as a matter of dignity owed rather than favour granted. Autonomy follows from dignity, because a person recognised as a bearer of dignity must be recognised as the author of the decisions that shape an intimate life. Privacy gives autonomy its protected space. The consensual acts with which the case was concerned took place in private, and the operative conclusion of the Bench was expressly confined to consensual sexual acts between adults in private, a confinement that reflects the constitutional weight carried by the private setting. Choice of partner completes the set. Intimacy is a relation between persons, and the freedom recognised by the judgment extends to the choice of the person with whom an adult life of companionship and intimacy is shared, a choice that Justice Chandrachud placed under Articles 14, 19 and 21 together. Each element disciplines the others. Dignity without autonomy would be status without freedom. Privacy without dignity would be mere secrecy, valuable only as concealment. Together they state a single proposition in several forms, that the intimate life of an adult is that adult’s own domain, into which criminal law may not enter merely because a majority disapproves of what occurs there with consent.

That final qualification brings the analysis to the opposition between constitutional morality and social morality, which the Chief Justice placed at the front of his opinion and which functions as the judgment’s answer to the strongest argument on the other side. That argument holds that the conduct in question is regarded as immoral by a large part of society, that criminal law exists in part to give effect to the moral convictions of the community, and that a court which removes the prohibition substitutes its own morality for that of the people. The judgment rejected the premise on which the argument depends. In a constitutional order, the morality that binds a court is the morality of the Constitution, and the Constitution commits the Republic to equality, to liberty and to dignity for every person, including persons whose lives the majority disapproves. Social morality is a fact about sentiment, and sentiment, however widely held, is not a source of constitutional authority. If it were, the character of fundamental rights would be pointless, because rights are needed exactly at the moment when their exercise is unpopular. Constitutional morality is a discipline of restraint imposed upon majorities and upon the State alike. It requires that disapproval, to be translated into criminal prohibition, pass through the tests the Constitution prescribes, of equality, of freedom, of dignity and of privacy. Disapproval that cannot survive those tests remains a private or social attitude that citizens are free to hold and to express, but not to enact into the criminal law against consenting adults. Social morality was therefore not refuted as a social phenomenon. It was denied jurisdiction over the question the Court had to decide.

What the judgment left standing

A judgment that reads a provision down rather than striking it down must be read twice, once for what it removes and once for what it deliberately retains, and the second reading is not a footnote to the first. The operative conclusion of the whole Bench was that Section 377 is unconstitutional insofar as it criminalises consensual sexual acts between adults in private. Every part of that formulation does work. The declaration is qualified by the words insofar as, which confine it to a specified application. That application is defined by three cumulative features, consent, adulthood and privacy, and the removal of the provision extends exactly as far as those features and no further. What lies outside them was left standing by decision rather than by oversight, because the reasons that destroyed the provision in one application did not touch it in the others.

What did Navtej Singh Johar leave untouched in Section 377?

Section 377 remained in force for non-consensual sexual acts against adults, for all sexual acts with minors, and for acts of bestiality. The Court read the provision down rather than striking it down in full, preserving those applications while removing consensual sexual acts between adults in private from its reach altogether.

The first carve out concerns non-consensual sexual acts against adults, and it follows directly from the logic of the judgment itself. The constitutional case against the provision rested upon consent at every stage. Autonomy is the freedom to choose, and an act imposed upon a person without consent is not an exercise of that person’s autonomy but a violation of it. Privacy protects the sphere within which free choices are made, and it confers no immunity upon a person who enters another’s sphere by force or coercion. Dignity, which the opinions treated as the foundation of the whole analysis, is injured by non-consensual acts rather than expressed through them. A judgment built on those premises could not coherently disable the criminal law from addressing conduct that negates each of them, and it did not attempt to do so. Section 377 therefore continues to govern non-consensual sexual acts against adults, standing alongside the rest of the criminal law as one of the provisions through which the State discharges its obligation to protect the bodily integrity and sexual autonomy of every person. The provision was not condemned as an instrument. Its use against consenting adults in private was condemned, and its use against coercion was preserved by the same reasoning.

The second carve out concerns all sexual acts with minors, and here the operative words are all and minors. The judgment drew its boundary at adulthood, without qualification as to the nature of the act or the presence of apparent agreement. The reasons are structural. The protections elaborated in the four opinions, autonomy, privacy, choice of partner and freedom of intimate association, are protections of the considered choices of adults, and the law treats minors as persons whose capacity to make choices of that gravity is not yet the capacity the Constitution presupposes when it shields intimate decisions from State interference. Consent, which performed decisive work in the adult context, cannot perform the same work where one party is a minor, because the law regards a minor’s agreement to sexual activity as incapable of carrying the legal significance that an adult’s consent carries. The retention of Section 377 for all sexual acts with minors therefore coheres with the protective architecture of the criminal law as a whole, in which the vulnerability of the child is the ground of prohibition rather than an exception to a general liberty. Nothing in the reasoning of the judgment disturbs that architecture.

The third carve out concerns acts of bestiality, which Section 377 continues to govern. Its logic completes the pattern of the other two. The constitutional analysis throughout the case proceeded from the rights of persons, from their dignity, their autonomy and their capacity to consent to a shared intimate life. An act of bestiality involves no second person whose consent could be given, no partner whose choice the Constitution could protect, and no relation of intimate association between citizens that the guarantees of Part III exist to shelter. It therefore falls outside every premise on which the declaration of unconstitutionality was built, as non-consensual acts fall outside the premise of autonomy and acts with minors fall outside the premise of adult capacity. The three carve outs, taken together, map the boundaries of the holding with unusual neatness. Where consent, adulthood and privacy are all present, the provision fell. Where any of those conditions is absent, it stands, and it stands not as a residue awaiting future correction but as law whose constitutional foundations the judgment gave no reason to doubt.

The technique by which this result was achieved, the reading down of the provision, deserves examination in its own right, because it explains how a court can declare a section unconstitutional in part while leaving the section itself upon the statute book. Reading down means that the Court preserved the provision by narrowing its application to the part that could survive constitutional scrutiny, rather than voiding the whole section. The words of Section 377 remained in the Indian Penal Code after 6 September 2018 exactly as they had stood before it. What changed was the authoritative statement of what those words could lawfully reach. A court that strikes a provision down removes it, and a legislature that wishes to legislate again on the subject must begin afresh. A court that reads a provision down retains it in a confined form, holding that the text, properly construed in conformity with the Constitution, never lawfully extended to the application now excluded. The distinction determines what police, prosecutors and trial courts are to do with the provision after judgment, and in this case it determined that the provision remained available for the classes of conduct described above, without any gap during which that conduct stood outside its reach and without any need for Parliament to enact a replacement in order to keep it criminal.

The limits of the judgment are the counterpart of its precision. Navtej Singh Johar decided the constitutionality of a criminal prohibition as applied to consensual acts between adults in private, and it decided that question completely. It did not strike Section 377 from the statute book, and the operative order is the measure of what it did. By that measure it removed a criminal prohibition from a defined sphere of adult private life while leaving the remainder of the provision exactly where it found it. The four opinions supply principles of considerable reach, on dignity, on autonomy, on the inclusion of sexual orientation within Article 15, on the chilling effect under Article 19 and on the subordination of social morality to constitutional morality, and those principles plainly carry significance beyond the operative order. But significance is not the same as decision. That combination, a decisive holding on the question presented, an apology for the delay in reaching it, and a deliberate preservation of everything the reasoning did not condemn, is the final character of the judgment.

The preserved residue at work, and then the repeal

From 6 September 2018 the provision that remained in the Indian Penal Code was narrower than the provision that had entered the Code in 1860, and the difference between the two forms explains the whole later history. The Constitution Bench in Navtej Singh Johar did not strike Section 377 down. It read the section down, so that consensual sexual acts between adults in private fell outside its reach, while a defined residue stayed in force as a criminal offence. Decriminalisation described one part of the work and preservation described the other.

The first element of the residue concerned non-consensual acts against adults. A sexual act imposed on an adult without consent had never depended for its wrongfulness on the identity of the persons involved, and the read down section kept that class of conduct within the criminal law. The point matters because the case for reading the section down had been built on consent, privacy and adulthood, and each of those ideas marked a limit as well as a liberty. Consent distinguished a private choice from a violation. Privacy distinguished the intimate life of adults from conduct that injured another person. Adulthood distinguished persons capable of that choice from children, for whom a different protective regime applies. The preserved offence was therefore not an accidental survival or an unfinished piece of reform. It was the part of the old text that corresponded to harms which the reasoning of the judgment gave no ground to excuse, and its continued operation gave that reasoning practical effect.

The second element concerned children. Section 377, in its read down form, continued to cover all acts with minors, and that coverage sat alongside a separate protective statute. Offences against children are covered by the Protection of Children from Sexual Offences Act, 2012, which is gender neutral in its protection and which covers every child below eighteen. The existence of that Act did not make the residue redundant in form, since the Penal Code provision and the special statute formed part of the same protective landscape, but it did mean that the position of children rested on more than one legislative foundation. For children an answer grounded in a dedicated statute was available. For the other classes within the residue the answer was far less complete, and the incompleteness was recognised before the repeal took effect.

The third element was bestiality. The original text had spoken of carnal intercourse against the order of nature with any man, woman or animal, and the reading down in 2009 and again in 2018 left sexual acts with animals within the offence. The reasoning is different in kind from the reasoning that protects consenting adults. An animal cannot consent in the sense that the criminal law recognises for a person, and the act was therefore not brought within the sphere of private adult choice that the judgment protected. By keeping bestiality inside Section 377, the courts kept it framed as a sexual offence under the general penal code, with the gravity that such a framing carries. The later question would be one of legal architecture. If the sexual offence disappeared, what remained was a law against cruelty, a different conception of the wrong, expressed in a different statute and carrying a different order of penalty.

Between September 2018 and June 2024 a judicially limited offence performed the work that the limitation itself assigned to it. That arrangement depended, however, on the continued existence of the Penal Code. A reading down is an interpretation of a statute that stays on the book. It settles what the words mean while the words remain law. It does not create a free standing offence that survives the repeal of the statute in which the words appear. The vulnerability of the arrangement was therefore structural and plain. If Parliament replaced the whole Code and chose not to carry the residue forward in some new form, the protection built by interpretation would fall with the text to which it was attached, however deliberate the distinction in the judgment had been.

The replacement came through the general revision of the criminal codes. The Bharatiya Nyaya Sanhita was first introduced in the Lok Sabha on 11 August 2023 and was referred to the Parliamentary Standing Committee on Home Affairs, chaired by Brij Lal. The Committee submitted its report on 10 November 2023. The issue before Parliament was the narrower and separate one of succession. A code that repealed the Penal Code as a whole had to decide, offence by offence, what would stand in the place of provisions that were still doing work. For the residue of Section 377 that was a decision about existence, because a provision that was not re-enacted in some equivalent form would not pass into the new code by implication.

The government withdrew the first bills and introduced revised bills on 12 December 2023. The revised Bharatiya Nyaya Sanhita contained no equivalent of Section 377, a position that can be traced through the Bharatiya Nyaya Sanhita 2023 explained account and through the IPC vs BNS comparison of the two codes. The Sanhita, Act 45 of 2023, received Presidential assent on 25 December 2023 and commenced on 1 July 2024. On commencement the Indian Penal Code stood repealed, subject to the savings in Section 358 of the Sanhita for acts done under the old Code. The savings preserved consequences for the past. They did not create an offence for the future. From that date the preserved residue ceased to exist as law, not because a court had reconsidered Navtej Singh Johar and not because Parliament had enacted a replacement directed to the same conduct, but because the statute that contained it had been replaced by a statute that contained nothing of the kind.

The gap where the preserved offence stood

Does the Bharatiya Nyaya Sanhita punish a non-consensual sexual act against an adult man?

No. The Bharatiya Nyaya Sanhita contains no offence that replaces the preserved residue of Section 377. Non-consensual sexual acts against adult men and transgender persons, and bestiality, are not made offences as such in the new code, and no gender neutral sexual offence covers them.

The gap is best understood as a problem of correspondence. The read down Section 377, for all the defects of its origin and its language, performed a specific function after 2018. It attached criminal liability to non-consensual sexual acts against adults, to sexual acts with minors and to bestiality, within a single provision of the general penal code. A successor provision needed only to cover the same ground. The debate over the Sanhita was therefore not a debate about restoring the offence against consenting adults, and it should not be described in those terms. The debate concerned the residue alone, and the question was whether a code presented as a comprehensive replacement for the Penal Code should leave that residue without any corresponding offence.

The Parliamentary Standing Committee on Home Affairs addressed the question directly. In its report of 10 November 2023 the Committee recorded that the Sanhita contained no provision for non-consensual sexual offences against male, female or transgender persons, or for bestiality. It then stated its conclusion in terms that leave little room for doubt. The Committee said: “The Committee feels that to align with the objectives stated in the BNS’s Statement of Objects and Reasons, which inter alia highlights the move towards gender-neutral offences, it is mandatory to reintroduce and retain Section 377 of the IPC.” The Committee did not argue from attachment to the old code or from disapproval of the judgment in Navtej Singh Johar. It argued from the stated objects of the new code itself. If the Sanhita presented its revision as a move towards gender neutral offences, then the removal of the one provision that, after 2018, reached non-consensual sexual acts without regard to the gender of the victim sat uneasily with that object.

The recommendation was not accepted. The government withdrew the first bills, a course that opened the text to revision, and introduced revised bills on 12 December 2023. The revised Sanhita still contained no equivalent of Section 377. That sequence gives the omission its particular character. Here the defect, if defect it was, had been identified in a formal report, with reasons tied to the objects of the bill, and the revised text was introduced after that report had been submitted. Parliament then passed the code in that form. It was a considered legislative choice, and its consequences fall to be assessed as a choice rather than as an oversight. This article states the general legal position for information and is not advice on any individual case.

The Section 377 timeline and what survives

Stage	What happened	What conduct remained criminal
1860, Penal Code enacted	The provision criminalised carnal intercourse against the order of nature with any man, woman or animal	All such acts, consensual or not
2009, Naz Foundation	The Delhi High Court read the provision down	Non-consensual acts, acts with minors and bestiality
2013, Koushal	The Supreme Court restored the provision in full	All such acts again
2018, Navtej Singh Johar	A Constitution Bench read the provision down again	Non-consensual acts, acts with minors and bestiality
November 2023, Standing Committee	The Committee on Home Affairs recommended retaining an equivalent offence	No change in the law at that stage
December 2023, revised Sanhita	The revised code passed without any equivalent	No change until commencement
1 July 2024, commencement	The Sanhita commenced and the Penal Code stood repealed	Consensual adult acts lawful, with no offence covering non-consensual acts against adult men or transgender persons, or bestiality, as such

The gendered definition of rape in the Sanhita explains part of the gap and also defines its limits. Section 63 defines rape as acts committed by a man against a woman, so the victim of that offence is defined in gendered terms. Section 64 prescribes rigorous imprisonment of not less than ten years, extending to imprisonment for life, for rape. They do not replace the residue of Section 377. A male victim of a non-consensual sexual act is outside the definition, and a transgender victim is outside the definition. No provision of the Sanhita creates a gender neutral sexual offence for adult victims to stand beside Section 63. The gravest sexual offence in the new code is framed around one class of victim, while the provision that had reached other classes in general terms was allowed to fall, and nothing of equivalent generality or gravity was put in its place.

The argument made for the omission, stated in its strongest form, is that the conduct in question has not been placed beyond the reach of the criminal law altogether. On that view, a non-consensual sexual act will ordinarily involve physical injury or force, and it can be prosecuted under the general offences of hurt, grievous hurt, assault and criminal force. Supporters of that position can add that a prosecutor is not left without any charge to bring. The argument is a real one, and it answers the exaggerated claim that such conduct became lawful in every sense on 1 July 2024. It does not, however, answer the narrower claim that a specific sexual offence disappeared without replacement, and it is the narrower claim that the critics have pressed.

The difficulty, as stated by critics including the petitioners in the Delhi proceedings, is threefold. First, the general offences describe physical injury or force in general terms. They were framed for violence as such, not for sexual violation as such. Second, those offences carry markedly lower penalties than the sexual offence they are said to replace. Third, they do not name or capture the sexual character of the violation. A conviction for hurt records that injury was caused. It does not record that the injury was inflicted through a sexual act imposed without consent. The contest is therefore not between punishment and impunity. It is about whether the criminal law should describe a sexual wrong as a sexual wrong, with a penalty that reflects that character, or whether a general law of violence is a sufficient answer. The committee took the first view. The revised code, in substance, took the second.

For transgender persons a further route is sometimes identified in the Transgender Persons (Protection of Rights) Act, 2019, a statute whose scheme is set out in the Transgender Persons Act 2019 explained account. Section 18 of that Act deals with offences against transgender persons, and clause (d) of Section 18 punishes sexual abuse of a transgender person with imprisonment of not less than six months and extending to two years, together with fine. Its limits, however, are plain on the face of the penalty. Critics, including transgender rights advocates, argue that a maximum of two years for sexual abuse is a trivial penalty beside the ten year minimum that Section 64 of the Sanhita prescribes for rape. The 2019 Act was also framed as a protective measure for a recognised class, not as the general sexual offences law of the country, and to treat its single clause as a full replacement for the residue of Section 377 is to load it with a function its penalty structure shows it was not built to bear.

For bestiality the remaining route is of a similar character. After the repeal, the only route identified for such conduct is prosecution as cruelty under the Prevention of Cruelty to Animals Act, 1960, with penalties far below those that Section 377 carried. That Act is directed to the welfare of animals and to the prevention of cruelty in general terms. A prosecution under it frames the wrong as cruelty to an animal. It does not frame the conduct as a sexual offence, and its penalties were set for the broad run of cruelty cases rather than for the specific conduct that the Penal Code had singled out. The point is that a specific offence, carrying a penalty that marked the gravity with which the conduct had been treated in the general code, was replaced by a general welfare offence of a markedly lower order. The pattern is the same in each area. No law can be found that does the work the preserved residue had done.

The gap has since been placed before the Delhi High Court. In August 2024 advocate Gantavya Gulati filed a public interest petition in that Court, Gantavya Gulati v Union of India, directed to the absence of any provision covering the conduct that the residue had covered. The High Court directed the Union to treat the petition as a representation and to decide it as expeditiously as possible, preferably within six months. The proceedings have since continued, with the Court pressing the Union over the delay and calling for an account of the steps taken. At the same time the Court has observed that creating a new offence is a legislative function, not one that a court can perform by interpretation. A court can record the gap and can require the executive to consider a representation. It cannot write a new sexual offence into the Sanhita, define its elements or fix its penalty, for those are acts of legislation. No final legislative decision on the representation has been recorded, and the position therefore rests where the revised code left it.

What the judgment did not decide, and what remains open

A judgment is in part defined by the questions it leaves to other institutions and other cases, and Navtej Singh Johar is no exception. In public discussion the decision is sometimes treated as if it settled the whole legal position of lesbian, gay, bisexual and transgender persons in India, as if decriminalisation carried with it further entitlements that followed by logical necessity. That treatment conflates distinct questions. The Constitution Bench was asked whether a criminal provision, in its application to consenting adults, violated the Constitution. It answered that question and, in answering it, it was careful to identify the conduct that remained punishable. It did not decide marriage. It did not confer a right to marry, it did not alter the law governing marriage, and it did not direct any authority to open any marriage statute to any class of couples. To read it as a marriage judgment is to misstate its holding and to obscure both its real strength and its real limits. Its strength lay in the ground it did decide, the removal of criminal liability from private consensual adult life. Its limit lay in everything beyond that ground, which awaited separate litigation and, where legislation was required, separate legislative action.

Marriage equality was litigated separately in Supriyo v Union of India, decided on 17 October 2023. A five judge Constitution Bench held there, by a majority of three to two, that there is no fundamental right to marry, and it declined to read the Special Marriage Act to permit marriages of same sex couples. The closeness of the division shows that the question was a substantial one argued on its own terms, not a formality that followed from the earlier decision. The sequence of the two cases is instructive. The first removed a criminal prohibition that the Constitution was held to forbid. The second declined to construct, by interpretation of a marriage statute, an institution that the majority held the Constitution did not compel. A reader who merges the two into a single story loses the analytical point of each. Decriminalisation and marriage recognition raise separate questions about the reach of constitutional rights, the proper role of a court and the proper role of Parliament, and the law reports keep them separate for good reason.

For teachers and researchers assembling the case sequence for a class or a brief, the discipline of keeping those questions separate is the practical lesson of this part of the record, and it is helped by setting the cases out in their order, with the committee report and the revised code placed after the judgments rather than folded into them, a task made easier by viewing the judgment sequence presentation in the ReportMedic presentation viewer. Presented in that order, the materials show three distinct institutional acts that are too often run together. The Court in 2018 limited an offence and preserved a residue. The Court in 2023 declined to extend a marriage statute. Parliament, across 2023 and 2024, replaced the penal code and omitted the residue altogether. Each act has its own reasons and its own history, and each must be judged against the question that was actually before the institution that performed it.

What remains open follows from that separation. The marriage question, after Supriyo, rests with Parliament as a matter of legislation rather than with the courts as a matter of the interpretation attempted in that case. The offence question, after the Delhi proceedings, rests with Parliament for the same institutional reason, since a court has observed that creating a new offence is a legislative function and no final legislative decision on the pending representation has been recorded. In neither area does the existing case law supply the missing provision. In neither area can the gap be closed by a more generous reading of a judgment that was framed, deliberately, to decide less than the whole subject. The openness is therefore not a temporary confusion in the reports. It is the settled consequence of limited judgments followed, in the penal field, by a general repeal, and it will persist in that form until the legislature chooses to address the residue that its own committee advised it to retain.

The verdict on Navtej Singh Johar

The defended position of this article is that a carefully limited judgment was followed by an unlimited repeal, and that only reading the two records together shows the resulting gap. Navtej Singh Johar limited itself with precision. It freed consensual acts between adults in private from the criminal law and it preserved, in the same movement, an offence for non-consensual acts, for acts with minors and for bestiality. The repeal that arrived with the Bharatiya Nyaya Sanhita observed no corresponding limit. It removed the whole of Section 377 and enacted nothing in its place, against the stated advice of the committee that had examined the bill and by means of a revised text introduced after that advice had been given. Neither record, read alone, discloses the full position. The judgment alone suggests a balanced settlement in which liberty and protection were adjusted together. The code alone suggests a general modernisation in which an old provision simply disappeared. Read together, they show something more exact and more troubling. The balance struck in 2018 depended on a text that the legislature later withdrew in full, and the class of victims whom the preserved residue had protected was left to general offences, a protective statute and a cruelty statute that none of the participants in the debate had designed for that purpose. The fault, on that reading, does not lie in the judgment, which did the limited work it set out to do and did it with care. It lies in a succession decision that treated a provision with a living protective function as if it had none, and the measure of Navtej Singh Johar is that its own careful distinctions are what make that later failure visible.

Frequently Asked Questions

Q: What did Navtej Singh Johar actually decide?

A five-judge Constitution Bench of the Supreme Court held on 6 September 2018 that Section 377 of the Indian Penal Code was unconstitutional insofar as it criminalised consensual sexual acts between adults in private. The Court did not strike the section down. It read the provision down, so that consensual adult conduct fell outside it, while the section continued to operate for non-consensual acts, acts with minors and bestiality. The bench was unanimous in that result, speaking through four opinions, and it overruled the 2013 decision in Suresh Kumar Koushal which had restored the provision in full.

Q: Why did the Supreme Court reverse itself on Section 377?

The reversal was not a change of mood but a change in the constitutional record. In 2014 the Court recognised transgender persons and gender identity in NALSA, and in 2017 a nine-judge bench held in Puttaswamy that privacy is a fundamental right and expressly rejected the reasoning in Koushal that the small size of the affected minority could justify denying a right. Fresh writ petitions placed the question before a Constitution Bench, which is the bench strength required to reconsider a question of that order, and the bench in Navtej Singh Johar then applied the intervening doctrine to the provision and overruled Koushal.

Q: Was Section 377 struck down completely?

No. The distinction matters because everything that followed depends on it. The Court used the technique of reading down: it kept the section on the statute book and confined its operation to the conduct that could survive constitutional scrutiny, namely non-consensual sexual acts against adults, all sexual acts with minors, and bestiality. Consensual sexual acts between adults in private were held to fall outside the provision because their criminalisation violated Articles 14, 15, 19 and 21. The section therefore continued in force in a narrowed form from September 2018 until the Indian Penal Code itself was repealed in 2024.

Q: What did Naz Foundation and Suresh Kumar Koushal hold?

In 2009 the Delhi High Court in Naz Foundation held Section 377 unconstitutional insofar as it criminalised consensual sexual acts between adults in private, reading the word sex in Article 15 to include sexual orientation, while preserving the provision for non-consensual acts and acts with minors. In 2013 a two-judge bench of the Supreme Court in Suresh Kumar Koushal reversed that decision and upheld the section in full. It reasoned that the provision criminalised acts rather than any identity, described the community as a minuscule fraction of the population, and said that amendment or deletion was for Parliament rather than the courts.

Q: Which constitutional articles did the Navtej Johar bench rely on?

The bench relied on four provisions working together. Article 14 supplied the test of manifest arbitrariness, which the undefined phrase order of nature and the provision’s indiscriminate sweep could not survive. Article 15 supplied the prohibition of discrimination on the ground of sex, read to include sexual orientation. Article 19 protected the expression of identity and intimacy, which the threat of prosecution chilled. Article 21 supplied the core: dignity, autonomy, privacy and the freedom to choose a partner, built on the foundation laid by the privacy decision in Puttaswamy a year earlier.

Q: Did the Navtej Johar judges apologise to the community?

One of the five judges did, in terms that have been quoted ever since. Justice Indu Malhotra, in her separate concurring opinion, wrote: “History owes an apology to the members of this community and their families, for the delay in providing redressal for the ignominy and ostracism that they have suffered through the centuries.” The sentence was not part of the operative order and bound nobody, but it was a formal acknowledgment, from the bench that finally decided the question, that the legal system itself had been the instrument of the harm for the period the provision had stood.

Q: What survives of Section 377 after the new codes?

Nothing of the section itself. The Bharatiya Nyaya Sanhita, which replaced the Indian Penal Code on 1 July 2024, contains no equivalent provision at all. Consensual adult conduct remains lawful, as Navtej Singh Johar required. But the residue the Court had deliberately preserved in 2018, covering non-consensual acts against adults and bestiality, was not carried forward either, because the new code defines rape as an act by a man against a woman and creates no gender-neutral sexual offence for adult victims. Offences against children remain covered by the Protection of Children from Sexual Offences Act, 2012, which is gender-neutral.

Q: Does Navtej Johar cover marriage rights?

No, and the judgment said so by its own limits. The petitions challenged a criminal provision, and the Court decided the criminal question: whether the state could punish consensual adult intimacy. It recognised dignity, identity and the choice of a partner as constitutional values, but it granted no right to marry and no recognition of any union. Marriage equality was litigated separately in Supriyo v Union of India, where a Constitution Bench held in October 2023, by a majority of three to two, that there is no fundamental right to marry and declined to extend the marriage law to same-sex couples.

Q: Who were the five judges on the Navtej Singh Johar bench?

The Constitution Bench comprised Chief Justice Dipak Misra and Justices R F Nariman, A M Khanwilkar, D Y Chandrachud and Indu Malhotra. All five agreed on the result, so the decision was unanimous, but they wrote four opinions rather than one: Chief Justice Misra wrote for himself and Justice Khanwilkar, and Justices Nariman, Chandrachud and Malhotra each wrote separate concurring opinions. Reading the case through only one opinion misses part of the doctrine, because the manifest arbitrariness analysis is fullest in Justice Nariman’s opinion, the privacy and partner-choice analysis in Justice Chandrachud’s, and the apology in Justice Malhotra’s.

Q: What happened to the curative petition filed against the Koushal judgment?

After review petitions against Koushal were dismissed in January 2014, curative petitions were filed, the curative jurisdiction being the final corrective remedy the Court recognises against its own judgments. On 2 February 2016 a three-judge bench referred those curative petitions to a Constitution Bench, so the provision’s fate formally remained open. In the event, the question was answered first through the fresh writ petitions filed in 2016, which were themselves referred to a Constitution Bench in January 2018 and decided that September in Navtej Singh Johar. The curative route thus set the stage, but the fresh petitions delivered the decision.

Q: What role did the Puttaswamy privacy judgment play in the Section 377 litigation?

Puttaswamy removed the intellectual foundation of Koushal before Navtej Singh Johar was ever heard. The nine-judge bench held in August 2017 that privacy is a fundamental right, and the plurality opinion of Justice Chandrachud held that sexual orientation is an essential aspect of privacy and identity. It also addressed Koushal directly, holding that the small size of the affected community, described in Koushal as a minuscule fraction of the population, was not a sustainable basis to deny the right to privacy, because fundamental rights exist precisely to be insulated from majority disapproval. The plurality stopped short of overruling Koushal only because the curative proceedings were pending.

Q: Why was the Section 377 challenge referred to a Constitution Bench?

Two reasons of law converged. A substantial question about the interpretation of the Constitution, and any reconsideration of an earlier precedent in the manner proposed, calls for a bench of at least five judges, and the petitions squarely asked the Court to disown the reasoning of Koushal, which had itself been delivered by a smaller bench resting on premises that NALSA and Puttaswamy had since undermined. When the three-judge bench took up the fresh writ petitions in January 2018, it recorded that the decision in Koushal required reconsideration and referred the matter to a larger bench, which is why the 2018 case was heard and decided by five judges.

Q: What is the difference between reading down a provision and striking it down?

Striking down removes the provision, or the offending part of it, from the law altogether, as if it had been repealed to that extent. Reading down keeps the provision in force but declares that it must be understood and applied in a narrower sense, so that the unconstitutional applications fall away while the rest continues to operate. In Navtej Singh Johar the Court chose the second course: Section 377 stayed in the Penal Code, but after September 2018 it could no longer be applied to consensual sexual acts between adults in private, while continuing to punish non-consensual acts, acts with minors and bestiality.

Q: Did the Union government oppose the Navtej Singh Johar petitions?

Not on the question the bench was deciding. When the Constitution Bench heard the petitions in July 2018, the Union filed an affidavit stating that it left the question of the constitutionality of Section 377, insofar as it concerned consensual sexual acts between adults, to the wisdom of the Court. That was a marked departure from the defence of the provision that had characterised the earlier rounds of litigation. The Union did ask that the hearing be confined to that question and not travel into matters such as marriage, and the bench’s decision respected that boundary, deciding the criminal provision and nothing beyond it.

Q: Why did Shashi Tharoor’s private member’s bill on Section 377 fail?

It failed at the threshold, twice, without ever being debated on its merits. In December 2015 Shashi Tharoor sought leave of the Lok Sabha to introduce a private member’s bill that would have substituted Section 377 so that only non-consensual acts remained punishable; members objected to the introduction itself, and the motion was put to a vote and defeated by 71 votes to 24, with one abstention. A second attempt in March 2016 met the same fate, defeated by 58 votes to 14. Because leave was refused, the text was never taken up, and the amendment Parliament was told in Koushal to consider was never considered.

Q: Did the Parliamentary Standing Committee recommend restoring Section 377 in the BNS?

Yes, expressly. The Parliamentary Standing Committee on Home Affairs, which examined the Bharatiya Nyaya Sanhita bill under the chairmanship of Brij Lal and reported on 10 November 2023, recorded that the new code contained no provision for non-consensual sexual offences against male, female or transgender persons, or for bestiality. It said that to align with the stated move towards gender-neutral offences, it was mandatory to reintroduce and retain Section 377 of the Penal Code, and it recommended that the government include the provision in the proposed law. The revised bill introduced the following month did not adopt that recommendation, and the code was enacted without any equivalent.

Q: What did the Delhi High Court direct in the petition over the missing Section 377 offence?

In proceedings begun by advocate Gantavya Gulati in August 2024, the Delhi High Court directed the Union to treat the petition as a representation and to take a decision on it as expeditiously as possible, preferably within six months. The petition asked, in substance, what offence now protects an adult man or a transgender person against a non-consensual sexual act after the repeal of Section 377. The High Court did not itself create any offence, and in the continuing proceedings it has treated the creation of an offence as a legislative function, while pressing the Union over the delay in deciding the representation and calling for an account of the steps taken.

Q: Which law now covers sexual offences against children that Section 377 once reached?

The Protection of Children from Sexual Offences Act, 2012, commonly called POCSO. It is a special statute that defines a child as any person below the age of eighteen and criminalises penetrative and non-penetrative sexual assault on a child in gender-neutral terms, so it protects boys and girls alike and always did so independently of Section 377. That is why the deletion of Section 377 did not open a gap for minors: the part of the old provision that concerned children overlapped with a dedicated statute that remains in force, unlike the parts concerning non-consensual acts against adults and bestiality, for which no equivalent dedicated offence was enacted.

Q: How does the Transgender Persons Act punish sexual abuse, and why is it criticised?

Section 18 of the Transgender Persons (Protection of Rights) Act, 2019 lists offences against a transgender person, and clause (d) covers acts that harm or endanger the life, safety, health or well-being of a transgender person, expressly including sexual abuse. The punishment for the whole clause is imprisonment of not less than six months and up to two years, together with a fine. The criticism, made by transgender rights advocates and repeated in the litigation over the new codes, is one of proportion: rape under the Bharatiya Nyaya Sanhita carries a minimum of ten years extending to life, so a sexual violation of a transgender person is graded by statute as a fraction of the same violation of a woman.

Q: Can a man who is sexually assaulted file a rape complaint under the Bharatiya Nyaya Sanhita?

No, because the definition does not reach his case. Section 63 of the Bharatiya Nyaya Sanhita defines rape as specified acts committed by a man against a woman, so the victim contemplated by the offence is a woman and the perpetrator a man. A male survivor of a non-consensual sexual act cannot be the complainant in a rape case under that section, and the code contains no substitute gender-neutral sexual offence for adults. What remains available are the general offences, such as those concerning hurt, grievous hurt and assault or criminal force, which describe physical injury in general terms and carry markedly lower penalties.