The Puttaswamy 2017 privacy judgment, delivered as Justice K. S. Puttaswamy (Retired) v. Union of India on 24 August 2017, is the verdict in which nine judges of the Supreme Court of India, speaking through six separate opinions and reaching one unanimous result, held that privacy is a fundamental right protected under Article 21 of the Constitution and as an intrinsic part of the freedoms guaranteed by Part III. The bench consisted of Chief Justice J. S. Khehar and Justices J. Chelameswar, S. A. Bobde, R. K. Agrawal, R. F. Nariman, A. M. Sapre, D. Y. Chandrachud, S. K. Kaul and S. Abdul Nazeer. The claim defended in the pages that follow is the claim the record supports most completely: the verdict did far more than recognise a right that earlier benches had left in doubt, because its lasting work was architectural, fixing where the right sits inside the constitutional text, which older authorities stand overruled and to what extent, which test a restriction on the right must pass, and which pattern later benches would follow when the right was carried into new fields.

The Aadhaar Petition That Forced the Privacy Question

The case began as a challenge to a programme of identification rather than as a seminar on constitutional theory. K. S. Puttaswamy, a retired judge of the Karnataka High Court, filed a writ petition under Article 32 in 2012 assailing the Aadhaar scheme of the Unique Identification Authority of India, under which residents were enrolled on a central database through fingerprints, iris scans and demographic particulars in exchange for a twelve-digit number. The petition contended that the scheme, which collected biometric material from the body itself and linked the resulting number to subsidies, bank accounts and public services, invaded a right to privacy that the Constitution protected. Other petitions joined the first, and the consolidated challenge pressed two distinct questions at once: an administrative question about the legality of the Aadhaar architecture, and a threshold constitutional question about whether privacy ranked as a fundamental right at all.

The threshold question reached centre stage because of the Union government’s answer to it. During the hearings of August 2015, Attorney General Mukul Rohatgi argued for the Union of India that the Constitution contains no fundamental right to privacy, that privacy is at highest a common law interest which a statute may qualify, and that the framers had deliberately declined to write a privacy guarantee into Part III. Rohatgi pressed the argument from the bench strength of the older authorities: an eight-judge bench in M. P. Sharma v. Satish Chandra in 1954 and a six-judge bench in Kharak Singh v. State of Uttar Pradesh in 1962 had each, in his submission, refused to recognise a general fundamental right to privacy, and smaller benches that had spoken of privacy in later decades had no power to displace what those larger benches had held. The submission converted the litigation. A three-judge bench of Justices J. Chelameswar, S. A. Bobde and C. Nagappan, by its order of 11 August 2015, held that the correctness of the Sharma and Kharak Singh line had to be settled by a larger bench before the Aadhaar challenge could proceed, and referred the privacy question accordingly.

The reference then travelled upward through the institution in stages. Interim orders in October 2015 confined Aadhaar enrolment to voluntary use for a set of schemes while the reference waited. On 18 July 2017 a five-judge Constitution Bench ordered that the reference be placed before a bench of nine judges, matching the scale of the question to the scale of the authorities that had to be revisited, since neither an eight-judge nor a six-judge precedent can be overruled by a bench of equal or smaller strength. The nine-judge bench heard argument across six days, from 19 July 2017 to 2 August 2017, and reserved judgment. On 24 August 2017 the result was announced: unanimous, in favour of a fundamental right. The speed between reference and verdict matters to the reading defended here, because the bench treated the privacy question as one of constitutional architecture that had been left unfinished since 1954, not as a dispute improvised by the Aadhaar controversy of a single decade.

Senior counsel Gopal Subramanium, leading the argument for the petitioners, framed privacy as a right that the Constitution presupposes rather than invents. In his submission, liberty under Article 21 and the freedoms under Article 19 lose their substance if the state may enter the home, read correspondence, track movement and compel disclosure of the body without constitutional restraint, and the absence of the word privacy from the constitutional text carried no decisive weight, since dignity, autonomy and liberty in that text are unintelligible without a protected zone around the person. Counsel for the Union, with Attorney General K. K. Venugopal appearing in the July 2017 hearings, retreated from the hardest edge of the 2015 position: the Union accepted that privacy might in some measure rank as a fundamental right, yet resisted a right of unlimited scope, warning that an unqualified privacy guarantee would obstruct welfare delivery, taxation, criminal investigation and the Aadhaar programme itself. The bench therefore had before it three positions in descending order of resistance: no fundamental right at all, a fundamental right of narrow and statute-dependent content, and a fundamental right anchored in the structure of Part III. The verdict adopted the third position and built the architecture to hold it.

Notice what the litigation posture explains about the verdict’s design. Because the Union had attacked the right at its root, the bench could not dispose of the reference with a narrow holding tied to biometric enrolment; it had to decide whether privacy exists as constitutional entitlement, where it resides in the text, and how a court should test a law said to invade it. Because the older authorities were large-bench decisions, the bench had to state precisely what Sharma decided, what Kharak Singh decided, and which parts of each survive. Because the Emergency-era judgment in Additional District Magistrate, Jabalpur v. Shivkant Shukla had held that Article 21 itself could be switched off by executive proclamation, and because that holding stood unreversed on the reports, the bench had to choose whether to leave the 1976 authority in place beside a freshly recognised privacy right or to strike it down in terms. The verdict did the striking, and the four components examined below are best read as answers to those four necessities: location, precedent, test and pattern.

Sharma, Kharak Singh and the Unfinished Lineage Before 2017

The first authority is M. P. Sharma v. Satish Chandra, decided on 15 March 1954 by a bench of eight judges speaking through Justice Jagannadhadas. The case concerned search and seizure: traders and companies under investigation for economic offences challenged the seizure of documents from their premises, arguing that compelled production of papers violated a privacy interest comparable to the protection the Fourth Amendment of the United States Constitution gives against unreasonable searches. The bench rejected the analogy. It held that the framers of the Indian Constitution had not imported a privacy right into Part III, that a power of search and seizure is in its nature a social necessity for the protection of the community against crime, and that the challenge failed. The reasoning was brief on the point that later mattered most: the opinion treated the absence of an express privacy clause in the Indian text, measured against the American text, as a considered exclusion.

Kharak Singh v. State of Uttar Pradesh, decided on 18 December 1962 by six judges, moved the question from papers to the person. Kharak Singh, a suspect in a dacoity case who had been discharged for want of evidence, challenged the surveillance regime of the Uttar Pradesh Police Regulations, which authorised domiciliary visits at night, secret picketing of his house, periodical enquiries into his habits and movements, and the tracking of his travel. The majority, speaking through Justice Ayyangar, struck down the domiciliary visits as an invasion of the liberty protected by Article 21, yet declined to recognise privacy as a guaranteed constitutional right, observing that the right of privacy is not a guaranteed right under the Constitution. Justice Subba Rao, with Justice Shah concurring in substance, dissented on the wider question: in the dissent’s reasoning, privacy is an essential ingredient of personal liberty, surveillance by the state chills the exercise of every freedom, and the whole surveillance chapter should have fallen. For fifty-five years the Indian law of privacy lived in the gap between the Kharak Singh majority and the Subba Rao dissent: surveillance was restrained in one of its forms, the general right was denied in terms, and the dissent supplied the vocabulary that later courts kept borrowing.

What did the surveillance majority in Kharak Singh actually decide?

The Kharak Singh majority of 18 December 1962 struck down night-time domiciliary visits under the Uttar Pradesh Police Regulations as a violation of Article 21, upheld the remaining surveillance measures of secret picketing, enquiry and movement tracking, and expressly declined to hold that privacy is a guaranteed right under the Constitution of India.

The decades after 1962 produced a series of smaller-bench decisions that built privacy doctrine floor by floor on a foundation the large benches had refused to certify. In Gobind v. State of Madhya Pradesh in 1975, Justice Mathew assumed a privacy right under Article 21, subjected the Madhya Pradesh surveillance regulations to a compelling state interest test borrowed from American doctrine, and upheld the regulations with a warning that privacy claims must be examined case by case. In R. Rajagopal v. State of Tamil Nadu in 1994, the Court articulated a right to be let alone in the law of publication, holding that a citizen has a right to safeguard the privacy of his own life, family, marriage and motherhood, subject to limits for public records and public officials. In People’s Union for Civil Liberties v. Union of India in 1997, the Court held telephone tapping to be a serious invasion of privacy permissible only under the procedure of the Indian Telegraph Act read with safeguards the judgment itself prescribed. In District Registrar and Collector, Hyderabad v. Canara Bank in 2005, the Court extended privacy reasoning to documents held by a bank, and in Selvi v. State of Karnataka in 2010 a three-judge bench protected mental privacy against involuntary narcoanalysis and polygraph examination, grounding the protection in Article 20(3) and in the substantive content of Article 21.

This middle line of cases is evidence for the architecture reading and against the simpler recognition story. By 2015 the Supreme Court had applied privacy to surveillance, publication, telephones, bank records and the mind, yet every one of those decisions was delivered by a bench smaller than the benches in Sharma and Kharak Singh, and none could formally displace the 1962 refusal. Indian privacy law was therefore a structure with occupied rooms and an unexamined foundation, and the Union’s 2015 argument aimed precisely at the foundation. Gautam Bhatia, in The Transformative Constitution published in 2019, describes the pre-2017 position as one in which privacy survived by incremental borrowing from Article 21’s expanding content rather than by settled entitlement. The nine-judge bench was convened to examine the foundation itself, and the verdict’s first architectural act was to state where the right lives.

Privacy as Integral Content of Article 21 and Part III

The first component of the verdict is a holding about location. Justice Chandrachud, writing the plurality opinion for himself, Chief Justice Khehar, Justice Agrawal and Justice Nazeer, held that privacy is a constitutionally protected right which emerges primarily from the guarantee of life and personal liberty in Article 21 and additionally from the other freedoms guaranteed by Part III. The plurality’s formulation is deliberate in its layering. Privacy is not a free-standing clause pasted onto the Constitution from outside; it is intrinsic to life and liberty, because the right to life in Article 21 protects the entitlement of each person to lead a life of dignity, and dignity in turn has no content if the state may observe, record and direct every choice by which a life is composed. The same right draws reinforcement from Article 19’s freedoms of speech, association and movement, each of which presupposes a sphere in which thought and affiliation form unobserved, and from Article 14’s promise against arbitrary state action.

Which facets of private life did the plurality name at the core?

The plurality named the preservation of personal intimacies, the sanctity of family life, marriage, procreation, the home and sexual orientation as the core of the privacy right, and treated informational privacy, the control a person exercises over personal data in an information age, as a distinct facet requiring protection.

The plurality organised the protected interest into facets that later courts could apply. Bodily privacy covers integrity of the body and protection against compelled physical intrusion, the interest at stake in the biometric enrolment that generated the case. Decisional privacy covers the realm of intimate choice: whom to marry, whether to bear a child, which faith to profess or refuse, what to eat, what to read, the intimate relations of adult life. Spatial privacy covers the home and private space against entry and observation, the interest vindicated in part in Kharak Singh. Informational privacy covers control over the dissemination and use of data about a person, a facet the plurality treated as central in an age in which profiles are assembled from fragments surrendered for unrelated purposes. The taxonomy matters architecturally: a right stated as facets can be extended by analogy to new technologies and new state practices without fresh constitutional amendment, because each new intrusion can be assigned to a facet and tested.

The plurality also fixed the right’s moral anchor. Privacy, in the opinion’s reasoning, is an inalienable natural right that inheres in the person by virtue of being human, and the Constitution recognises rather than confers it. That anchor choice has a downstream consequence the opinion states plainly: a right that the Constitution recognises cannot be abolished by the amending power as though the text had created it, and the plurality connected the holding to the basic structure jurisprudence traced in the basic-structure-doctrine establishment and renewed in the most-consequential-constitutional-moment framework. Justice Bobde’s concurrence grounded privacy in the same natural-right register, treating it as a condition of the exercise of every enumerated freedom. Justice Chelameswar’s concurrence organised privacy as three freedoms of repose, sanctuary and intimate decision, and warned against definitions so wide that the right would dissolve into all of liberty. The six opinions diverge in emphasis and agree in location: the right sits inside Part III, with Article 21 as its principal seat.

The location holding also reversed a presumption that had governed since 1954. Sharma had treated the silence of the Indian text, against the noise of the American Fourth Amendment, as proof of deliberate exclusion. The plurality answered on the Constituent Assembly record and on principle: the framers drafted Part III in open language, the Assembly’s advisers had considered and set aside express surveillance clauses without rejecting the underlying interest, and constitutional rights in a document meant to endure are read by their substance rather than by a census of their words. The comparative survey in the plurality opinion runs from Warren and Brandeis’s 1890 Harvard Law Review essay The Right to Privacy, through Justice Brandeis’s dissent in Olmstead v. United States in 1928 with its defence of the right to be let alone, to Griswold v. Connecticut in 1965, in which the United States Supreme Court located marital privacy in the penumbras of enumerated guarantees. The plurality drew from that survey a method rather than a result: enumerated rights cast shadows in which unenumerated protections live, and a court reading a bill of rights must map the shadows as carefully as the text.

The Overrulings: Sharma, Kharak Singh and ADM Jabalpur

The second component is a set of overrulings stated with unusual precision, each calibrated to the exact extent of the error corrected. M. P. Sharma stands overruled to the extent that it held that the Constitution does not protect a right to privacy. Kharak Singh stands overruled to the extent that its majority held privacy not to be a guaranteed right, while its actual order against domiciliary visits, and Justice Subba Rao’s dissent on the wider question, are absorbed into the new doctrine. The surgical phrasing is itself architecture: by overruling propositions rather than judgments wholesale, the bench preserved the operative orders of the older cases, discarded only the denials that blocked the right, and signalled to later benches that overruling in this Court proceeds by proposition and extent, a discipline visible in the architectural-substantive-extension framework by which the Ninth Schedule jurisprudence was rebuilt in 2007.

The largest correction was reserved for the Emergency. In ADM Jabalpur, decided on 28 April 1976, a five-judge bench had held by four to one that a Presidential order under Article 359 suspending enforcement of Article 21 left a detainee without any remedy against a detention that was mala fide, unauthorised or mistaken, because the right to life and liberty was, in Chief Justice Ray’s formulation, the sole repository of the protections Article 21 contains and the repository stood suspended. Justice Khanna’s dissent, which insisted that life and liberty do not derive from the Constitution alone and cannot be extinguished by proclamation, cost him the Chief Justiceship when he was superseded in January 1977. The Puttaswamy plurality addressed that holding by name and without euphemism: Justice Chandrachud wrote that the judgments rendered by the majority in ADM Jabalpur were seriously flawed, that life and personal liberty are inalienable rights, and that the Khanna dissent states the correct position in law. A son of one of the ADM Jabalpur majority judges thus wrote the opinion that overruled his father’s most consequential holding, a circumstance the plurality neither advertised nor avoided, and one that gave the correction a weight inside the institution that a bare citation could not carry.

The overruling of ADM Jabalpur is treated in this series in the institutional-architectural-failure-pattern, where the 1976 verdict is read as the Court’s deepest institutional failure; the Puttaswamy correction is the pattern’s closing term. Its legal effect reaches beyond symbolism. The Forty-fourth Amendment of 1978 had already amended Article 359 to place Articles 20 and 21 beyond suspension even during an Emergency, yet a statute-book amendment and a judicial overruling do different work: the amendment barred future governments from suspending the article’s enforcement, while the overruling withdrew the judicial reasoning that had once blessed the suspension, namely the theory that Article 21 is the sole source of the right to life. Under the Puttaswamy plurality, life and liberty are recognised by the Constitution rather than granted by it, from which it follows that no organ of the state, by proclamation or by amendment dressed as interpretation, can treat the right as a concession revocable at pleasure. The correction therefore locks the first component in place: location inside Article 21 would mean little if the article remained, in doctrine, a tap the executive could turn off.

The Test: Legality, Legitimate Aim and Proportionality

A right without a test for its limitation is a declaration; the third component of the verdict supplies the test. The plurality held that an invasion of privacy must satisfy a threefold requirement drawn from the structure of Part III itself: first, the existence of a law authorising the invasion, since executive action without legal basis fails at the threshold; second, a legitimate state aim or need that the law pursues, corresponding to the interests for which the Constitution permits freedoms to be restricted; third, proportionality, meaning that the extent of the invasion bears a rational and measured relation to the aim pursued, so that the means chosen do not burden the right beyond what the aim can justify. Justice Kaul’s concurrence stated the discipline in the form later benches would quote most often, adding to legality, legitimate aim and proportionality a requirement of procedural guarantees against abuse, the safeguards that convert a paper test into an operating restraint on surveillance and data collection.

Scholarship on the judgment, Gautam Bhatia’s account among it, restates the discipline in the four-stage vocabulary of structured proportionality used by constitutional courts in Germany, Canada and South Africa: a legitimate aim, a suitable means rationally connected to that aim, a necessary means that is the least restrictive adequate alternative, and a balancing stage at which the gain to the aim is weighed against the loss to the right. The four-stage restatement is analytical elaboration rather than contradiction of the plurality’s threefold requirement, since suitability and necessity are contained in the plurality’s demand that the invasion be proportionate to the need, and balancing is the operation by which proportionality is judged in a hard case. Suhrith Parthasarathy, in essays published in Economic and Political Weekly after the verdict, treated the test as the verdict’s most consequential export, the element that would decide whether the recognition of the right changed outcomes in litigation over surveillance, data retention and welfare architecture, or merely changed the vocabulary in which the state won.

The test’s design repays attention, because each stage answers a failure the lineage had exhibited. Legality answers the Kharak Singh defect, where surveillance rested on police regulations of doubtful statutory parentage: after Puttaswamy, a privacy invasion resting on executive instruction alone fails without further enquiry. Legitimate aim answers the Gobind defect, where a compelling interest was gestured at rather than demonstrated: after Puttaswamy, the state must name the aim and the court must accept the aim as one the Constitution counts legitimate. Proportionality answers the ADM Jabalpur defect at the level of method, since a court equipped with a structured test for rights limitation has no need to choose between abdication and veto; it can measure. The bench also marked the test’s domain: privacy is not absolute, no judge on the bench held it absolute, and the architecture is therefore a machinery for adjudicating collisions between the right and competing public purposes, not a wall against governance. That concession, often mistaken for weakness in the verdict, is part of its strength in the reading defended here, because a test that governments can sometimes pass is a test that courts will actually apply.

The Pattern and the Four-Component Right-Architecture Map

The fourth component is the verdict’s establishment of a pattern for rights adjudication beyond the privacy field itself. A right located in the structure of Part III, stated as facets, limited by a structured test and freed from the Emergency-era theory that Article 21 is a revocable grant: that combination is a template. A later bench confronting a claim under Article 21, or a claim that an enumerated freedom carries an unenumerated penumbra, inherits from Puttaswamy a sequence of operations. Identify the facet of the protected interest that the state action touches. Locate the interest within the text by structure and by the dignity anchor. Test the invasion for legality, legitimate aim and proportionality, with procedural safeguards where the invasion is covert or continuous. The sequence is the verdict’s most portable product, and it explains why the judgment is cited in fields as distant from biometric enrolment as sexual orientation, religious practice, reproductive choice and internet access.

The four components and their first applications are set out in the Four-Component Right-Architecture Map, the citable summary of the reading defended in this article.

Component	Holding of 24 August 2017	First post-2017 application
Integral location in Article 21	Privacy intrinsic to life and liberty and to the freedoms of Part III	Navtej Johar on 6 September 2018, reading down Section 377
Calibrated overruling	ADM Jabalpur overruled; Sharma and Kharak Singh overruled to the extent of their denials	Article 21 treated as non-suspendable in doctrine as in the amended text
Structured restriction test	Legality, legitimate aim, proportionality and procedural guarantees	The Aadhaar verdict of 26 September 2018
Facet-based pattern	Bodily, decisional, spatial and informational facets extended by analogy	Sabarimala on 28 September 2018 and Joseph Shine on 27 September 2018

Read as a map, the verdict’s unity becomes visible. Each row answers a defect the earlier rows of Indian privacy law had left open: location answers Sharma’s exclusion by silence, overruling answers Kharak Singh’s denial and the Emergency’s suspension theory, the test answers the ad hoc balancing of the Gobind line, and the pattern answers the fragmentation of a doctrine that had grown case by case without a stated method. The scholarly disagreement the article must adjudicate, between the recognition reading, the Aadhaar-context reading and the architecture reading defended here, is taken up after the trajectory evidence is in, since the map’s fourth row can only be assessed by watching the pattern operate in the hands of later benches.

Six Opinions, One Result: The Bench Assembled

Which judge wrote the plurality opinion, and who joined it?

Justice D. Y. Chandrachud wrote the plurality opinion, joined by Chief Justice J. S. Khehar and Justices R. K. Agrawal and S. Abdul Nazeer, so that four of the nine judges spoke in its terms, while the remaining five judges wrote or joined concurrences reaching the same result.

The unanimity of 24 August 2017 is a fact about result, not about reasoning, and the six opinions repay separate attention because later courts cite them separately. Justice Chelameswar, whose presence on the referring bench of 2015 had kept the question alive, wrote of privacy as an element of liberty that precedes the Constitution, organised into repose, sanctuary and intimate decision, and grounded in the freedoms of Part III read together. Justice Bobde treated privacy as a natural right inhering in personhood, the condition on which the enumerated freedoms rest. Justice Nariman built a taxonomy of privacy interests, moving from bodily integrity through spatial and decisional claims to informational control, and confronted the argument from welfare directly: a state that conditions food and subsidy on surrender of all privacy, in his reasoning, misunderstands the beneficiary as a subject of largesse rather than a holder of rights. Justice Sapre located privacy within Article 21 and the expressive freedoms, emphasising its character as a right the Constitution declares rather than creates. Justice Kaul, in the opinion most occupied with technology, warned that data assembled for one purpose can be mined for another, that profiling sorts persons into categories they never chose, and that the state and private parties alike must be brought under discipline if the right is to survive the digital environment in which it will be tested.

Two features of this distribution deserve record. The first is that no opinion defended the 2015 position of the Union in its full form; the argument that privacy enjoys no constitutional protection found no taker on a bench spanning the Court’s ideological range, and the Attorney General’s retreat in the July 2017 hearings foreshadowed the shutout. The second is that the plurality’s reasoning, not any single concurrence, became the working text for later benches, because a four-judge plurality within a unanimous nine supplies the closest available approximation to a majority rationale. Gautam Bhatia’s reading of the judgment in The Transformative Constitution treats the plurality as the Court’s most ambitious statement of a Constitution that remakes social relations rather than merely policing state excess, and locates Puttaswamy beside Kesavananda Bharati as a judgment whose importance lies in the method it hands to successors. Pratap Bhanu Mehta, in essays written in the weeks after the verdict, praised the result and the unanimity while cautioning that a court which discovers rights by structure must also build the institutional machinery, in data law and surveillance law, without which discovered rights remain proclamations. Both assessments, the celebratory and the cautionary, are assessments of architecture: each asks what the verdict built, not merely what it recognised.

The Trajectory: Aadhaar, Navtej Johar, Sabarimala and After

The pattern’s first full application came in the litigation from which the reference had sprung. On 26 September 2018 a five-judge bench delivered the Aadhaar verdict, upholding the Aadhaar Act in large part while striking down Section 57, which had permitted private companies to demand Aadhaar authentication, restricting the retention of authentication records, and reading down provisions on disclosure in the interest of national security. Justice Sikri’s majority opinion ran the Aadhaar architecture through the proportionality discipline Puttaswamy had installed, asking whether biometric enrolment pursued a legitimate aim in welfare delivery, whether the means were suited and measured, and where the balance failed. Justice Chandrachud, dissenting, applied the same discipline and reached the opposite destination, holding the Act unconstitutional in its entirety and describing Aadhaar as an architecture of surveillance prone to exclusion. That both the majority and the dissent argued inside the Puttaswamy test, disputing its application rather than its authority, is the strongest early evidence for the architecture reading: within thirteen months the test had become the grammar of the field. The Aadhaar verdict receives full treatment later in this series, as an unlinked forward reference here, because its own record runs to a separate judgment of over fourteen hundred pages.

On 6 September 2018, three weeks before the Aadhaar verdict, Navtej Singh Johar v. Union of India read down Section 377 of the Indian Penal Code so that consensual sexual relations between adults ceased to be a crime. The five judges in Navtej Johar cited Puttaswamy as the premise of the analysis: sexual orientation falls within the decisional facet of privacy, identity is protected by the dignity anchor, and a law criminalising intimacy must pass a test the colonial provision could not survive. Justice Chandrachud’s Navtej opinion extended his plurality’s logic from protection against the state to recognition by the state, and Justice Indu Malhotra’s opinion treated the history of the provision as an apology owed by the legal order. The sequence from Puttaswamy to Navtej Johar is the pattern operating at speed: facet identified, location fixed, test applied, precedent distinguished, and a line of authority running back through the Delhi High Court’s 2009 Naz Foundation judgment absorbed into doctrine.

On 28 September 2018 the Sabarimala verdict, in Indian Young Lawyers Association v. State of Kerala, struck down the exclusion of women of menstruating age from the Sabarimala temple by four votes to one, with the majority opinions invoking dignity, equality and the privacy-adjacent claim that faith practices cannot be policed by biological markers imposed by others. The tension between the individual devotee’s Article 25 freedom and the denomination’s Article 26 autonomy, framed in this series through the religious-secular architectural-tension framework and through the contested-settlement framework, is the pattern under strain, since privacy doctrine built for claims against the state had to be fitted to a claim inside a religious community. A companion application of the same week, Joseph Shine v. Union of India on 27 September 2018, struck down the adultery offence in Section 497, with the opinions treating the provision as a violation of decisional privacy and dignity within marriage. One day earlier in the same crowded fortnight the Aadhaar verdict had shown the test disciplining a welfare architecture; taken together, the September 2018 verdicts applied the 2017 pattern to criminal law, religious practice and identification infrastructure within a single month.

The statutory sequel arrived more slowly, and its pace is part of the record. The committee chaired by Justice B. N. Srikrishna, appointed in July 2017 while the nine-judge bench sat, submitted its report and a draft Personal Data Protection Bill on 27 July 2018, proposing a data protection authority, consent architecture and penalties for misuse of personal data. A revised Personal Data Protection Bill was introduced in the Lok Sabha in December 2019 and referred to a joint committee of Parliament, where examination continued into 2020. In Anuradha Bhasin v. Union of India, decided on 10 January 2020, the Court applied proportionality reasoning to indefinite internet restrictions in Jammu and Kashmir, ordering review of suspension orders and treating access as bound up with the freedoms of speech and trade. The trajectory through 2020 therefore runs in two channels: courts applying the pattern with increasing confidence, and a legislature still drafting the statute that Justice Kaul’s concurrence had treated as the right’s necessary complement.

Two further applications mark the pattern’s reach into forums the 2017 bench never addressed. On 19 September 2019 the Kerala High Court, in Faheema Shirin v. State of Kerala, held that access to the internet forms part of the right to privacy and the right to education under Article 21, striking down a women’s hostel rule that confiscated mobile phones and barred laptop use in study hours; a single judge of a High Court, applying the plurality’s dignity reasoning to a hostel warden’s regulation, demonstrated how completely the 2017 architecture had passed into the ordinary equipment of Indian adjudication. In February 2020 a nine-judge bench of the Supreme Court, constituted on the Sabarimala review reference to settle questions on the interplay of Articles 25 and 26 with other fundamental rights, took up a docket in which Puttaswamy’s location of privacy inside Part III is one of the premises in contention, since the freedom of religion questions cannot be answered without deciding how individual rights operate inside collective practices. The verdict’s third year thus closed as its first had, with the pattern being carried into new subject matter by courts reasoning from its paragraphs rather than from its result alone.

Adjudicating the Three Readings of the Verdict

Three readings of the verdict compete in the literature, and the evidence assembled in the preceding sections allows a judgement among them. The recognition reading, carried by the first wave of commentary and by the standard encyclopaedic treatment, presents 24 August 2017 as the day India acquired a right to privacy: a gap in Part III, visible since Sharma, was at last filled. The Aadhaar-context reading, carried by commentary written in the shadow of the identification programme, presents the verdict as a preliminary skirmish in the Aadhaar war, a nine-judge throat-clearing before the real contest of September 2018. The architecture reading defended here presents the verdict as the installation of a right’s operating system: a location, a set of calibrated overrulings, a structured test and a transferable pattern, with recognition as one component among four.

The recognition reading is true and insufficient. It is true because the holding of 24 August 2017 did settle a question that Sharma and Kharak Singh had left answered in the negative, and any account that omits the settlement misdescribes the event. It is insufficient because recognition alone predicts nothing about the verdict’s behaviour in later hands. A bench that had merely recognised the right, in a short opinion confined to the proposition that privacy is fundamental, would have left the September 2018 courts to invent their own methods; the Aadhaar majority, the Navtej Johar bench and the Joseph Shine bench would have shared a slogan rather than a discipline. What those benches in fact shared, as the trajectory record shows, was the discipline: the facet analysis, the location argument and the proportionality sequence, each traceable to identified paragraphs of the plurality and the Kaul concurrence. Recognition explains the headline of the verdict; architecture explains its citations.

The Aadhaar-context reading is also true and also insufficient, in the mirror direction. The reference did arise from Aadhaar, the petitioners’ target was an identification architecture, and the verdict’s most demanding early application was the Aadhaar judgment itself. Yet the context reading cannot explain the bench’s own choices of scope. A bench concerned only with clearing the ground for Aadhaar could have answered the reference in the narrowest available terms, holding that some privacy interest exists under Article 21 and remitting everything else to the five-judge bench that would hear the Aadhaar challenge. Instead the nine judges wrote five hundred and forty-seven pages, overruled a judgment of the Emergency that had nothing to do with biometrics, surveyed comparative doctrine from 1890 forward, and composed a test stated for privacy invasions generally. Opinions are sized to their ambitions, and the ambition on display was the reconstruction of the right’s foundations, with Aadhaar as the occasion rather than the subject. Read in the frame of the integrated three-tensions framework, the verdict belongs to the long contest over the terms on which the Indian state meets the person, a contest that runs across governments and cannot be assigned to a single programme of a single administration.

The strongest counter to the architecture reading deserves statement in its strongest form. The post-2017 application of the proportionality test has been uneven. Critics of the Aadhaar majority, Suhrith Parthasarathy prominent among them, argued that the majority recited the test and then deferred at each stage where deference mattered, accepting the welfare aim without demanding evidence of less invasive alternatives and treating exclusion errors as administrative mishaps rather than rights violations. The internet restrictions litigation of 2019 and 2020 produced orders that announced proportionality and then accepted executive assurances with limited scrutiny. On this evidence, the counter runs, the pattern is a vocabulary that legitimises outcomes reached on other grounds, and the architecture reading overstates the coherence of a practice that remains, in truth, ordinary judicial discretion dressed in a four-stage suit. The counter is partly sound, and the concession should be made without flinching: the test has not disciplined every bench that has recited it, and the distance between the Puttaswamy standard and the Aadhaar majority’s application of that standard is the largest open wound in the doctrine.

The concession does not overturn the reading, for two reasons grounded in the record rather than in loyalty to it. First, uneven application is evidence about benches, not about the existence of the standard: a test that a dissent can wield against a majority, as Justice Chandrachud wielded proportionality against the Aadhaar Act, is a test with bite, since the disagreement inside the 2018 bench was conducted as a disagreement about what the shared standard required. Second, the architecture reading is a claim about what the verdict installed, not a prediction that installation guarantees performance. Kesavananda Bharati installed the basic structure doctrine in 1973 and its application remained contested for decades without the installation claim failing; doctrines in this Court operate as resources for argument whose force varies with the bench, the case and the quality of advocacy. The pattern’s analytical value survives uneven application because the pattern supplies the terms in which unevenness itself is identified and criticised, as Parthasarathy’s critique demonstrates by its own method: the charge that the Aadhaar majority under-applied proportionality is intelligible only because Puttaswamy fixed what full application requires.

What the Verdict Left Open

What warning did Justice Kaul issue about the digital age?

Justice Kaul warned that the digital age dissolves the boundary between public and private life, because data surrendered for one purpose can be combined, mined and profiled for others, and he insisted that informational privacy requires statutory safeguards, security practices and restraint by private parties as well as by the state.

Three open questions mark the boundary of the verdict’s achievement. The first is surveillance law. The plurality and the concurrences assumed a legal framework for interception resting on the Indian Telegraph Act of 1885 and the safeguards of the PUCL judgment of 1997, a framework built for telephone taps and strained by bulk collection, facial recognition and communications metadata. Puttaswamy supplies the test any reformed surveillance statute must pass, and the procedural-guarantee limb of the Kaul formulation supplies its design brief, yet the statute itself remains unwritten within the period this article covers, and interception continues under executive rules of colonial vintage. The second open question is horizontal reach. Privacy invasions arrive from employers, platforms, data brokers and neighbours as readily as from ministries, and Justice Kaul’s concurrence flagged the claims of the right against non-state actors as a question the Constitution’s largely vertical structure answers awkwardly; the data protection bill’s long passage through committee is, in substance, the legislature’s attempt to build the horizontal machinery the verdict could describe but could not enact. The third open question is remedial. A right tested by proportionality invites the state to justify rather than to desist, and the remedial habit of the post-2017 courts, reading down statutes and issuing guidelines in place of invalidation, leaves the right’s beneficiaries with victories of principle whose administrative translation lags, a lag visible in the years between the PUCL safeguards of 1997 and any comprehensive interception reform.

A practical illustration of the verdict’s reach into record-keeping itself can be stated plainly. The informational facet protects the material a person surrenders to institutions, and the discipline of purpose limitation, that data collected for one end may not be silently redirected to another, governs archives as much as databases; guidance addressed to practitioners on how journalists use VaultBook to protect sources belongs to the same family of practice as the safeguards the Kaul concurrence demanded, since source protection is informational privacy asserted against compelled disclosure. The verdict’s architecture, in other words, reaches from the Constitution Bench to the newsroom file, and its test will be the measure of any Indian statute that finally governs the data economy.

The Reference Heard: Argument Before the Nine Judges

The quality of a constitutional verdict often reflects the quality of the contest that produced it, and the contest of July and August 2017 was argued at a level the eventual opinions acknowledge in their texture. Gopal Subramanium opened for the petitioners with a submission built on structure. The Constitution of India, he argued, is a document that distributes power against a background of retained liberty; Part III enumerates freedoms because experience taught the framers which freedoms governments invade first, not because the enumerated list exhausts the liberty a person holds. Privacy in that submission is the atmosphere in which the enumerated freedoms operate: speech requires a mind formed in private, association requires meetings the state has not catalogued, movement requires journeys the state has not logged, and conscience requires a sanctuary no official may enter. Subramanium pressed the comparative record as well, pointing to the constitutions drafted after 1945 across Europe, Africa and Asia that protect dignity and private life in terms, and to the international covenants India had signed, as evidence that privacy belongs to the standard equipment of a rights-bearing constitution rather than to the luxuries of any single legal culture.

Shyam Divan, appearing for petitioners in the connected Aadhaar challenges, carried the argument from structure to the body. Biometric enrolment, in his submission, treats the citizen’s body as state infrastructure: the fingerprint and the iris become keys the person cannot change, surrendered once and usable everywhere, so that a compromise of the database is a compromise of the person for life. Divan warned of function creep, the documented tendency of identification systems built for welfare delivery to migrate into policing, banking, telecommunications and employment screening, each migration enlarging the dossier while the consent given at enrolment stays frozen in its first, narrow form. He warned equally of exclusion, the authentication failure that denies a pensioner grain because worn fingerprints no longer match a template, and framed exclusion not as an administrative defect but as the predictable cost of making rights conditional on a machine’s recognition of the body. The plurality’s insistence that privacy protects the poor with the same force as the prosperous, and its treatment of the welfare argument on the merits rather than as a trump, answers Divan’s submissions in substance.

Counsel addressing the informational dimension, among them advocates appearing for intervenors from the digital rights field, directed the bench to the economics of data. Aadhaar was one database; the environment into which the verdict would speak contained thousands, held by telecommunications companies, hospitals, employers, platforms and ministries, each collecting under consent forms no reader could realistically audit. The submission urged the bench to state the right at a level of generality that would govern the environment rather than the single programme before the Court, since a holding confined to biometrics would be obsolete against the next collection technology within a decade. Justice Kaul’s concurrence, with its attention to profiling, big data and the convergence of state and corporate databases, is the opinion that most visibly absorbed this strand of argument, and its insistence on procedural guarantees and on discipline for private holders of data reads as a direct reply to the intervenors’ brief.

The Union’s argument in the July 2017 hearings, led by Attorney General K. K. Venugopal, differed in register from the 2015 submission and the difference is itself evidence. The Union no longer denied that privacy could rank as a fundamental right in some of its dimensions; instead it argued for a right of limited and enumerated content, strong for the home and for intimate life, weaker or absent for data surrendered in transactions with the state, and always subordinate to legislation enacted for public welfare. Venugopal cautioned the bench against a holding so wide that every tax enquiry, census operation and criminal investigation would begin its life under constitutional suspicion. Several states, appearing through their counsel, supported the petitioners on the existence of the right while reserving their positions on its application, an alignment that deprived the Union’s resistance of the federal reinforcement it might have expected. The bench’s questions during the six days of hearing, preserved in the reportage of the proceedings and reflected in the opinions, returned repeatedly to a single pressure point: if privacy is not fundamental, what in the Constitution prevents a law authorising the compulsory recording of every conversation in every home, provided the law is duly enacted and fairly administered? No answer offered in argument survived that question intact, and the unanimity of the result was, in retrospect, visible from the moment the question went unanswered.

Dignity, Autonomy and the Grammar of the Plurality

The plurality’s theoretical core is a chain of three links: dignity, autonomy and privacy, each necessary to the next. Dignity names the status of the person as an end rather than an instrument of policy, a status the Court had been building into Article 21 since Francis Coralie Mullin v. Administrator, Union Territory of Delhi in 1981, where the right to life was held to include the right to live with human dignity, and since Maneka Gandhi v. Union of India in 1978, where Article 21 was read to demand that any procedure curtailing liberty be just, fair and reasonable rather than arbitrary or oppressive. Autonomy names the practical content of dignity in a life actually lived: the capacity to form intentions, attach to persons and causes, revise one’s commitments and present oneself to the world under a description one has chosen. Privacy names the condition autonomy requires, the protected space of body, home, correspondence and data within which intentions form before they are performed in public. A state may regulate the performance extensively; the plurality’s holding is that the space of formation is constitutionally defended, and that a law entering the space must justify the entry under the test the verdict prescribes.

This chain explains a feature of the plurality that puzzles readers who approach the verdict as a privacy case in the narrow sense: its long engagement with cases on food, dress, faith and sexuality that seem remote from databases. The engagement is deliberate. The plurality treated the choice of what to eat, illustrated by the litigation over beef bans and the Hinsa Virodhak Samaj line of reasoning, the choice of what to wear, the choice of faith and the choice of intimate partner as instances of one protected genus, the decisional facet, and it placed them under privacy rather than under the enumerated freedoms alone because privacy is the genus that unites them. A restriction on diet can be attacked under Article 19 or Article 25 in a given case; the plurality’s move is to hold that behind the enumerated attacks stands a general protection for the intimate decisions by which a person composes a self. That generality is the feature critics of the verdict most distrust, since a genus-level right multiplies the occasions for judicial intervention, and it is equally the feature defenders most prize, since enumerated attacks alone leave every novel intrusion unprotected until a legislature or a later bench catches up.

The plurality’s handling of the natural right register shows the same care. To call privacy inalienable and inherent is to say that the Constitution recognises a pre-political entitlement, and the plurality said it while anchoring the recognition in text, in structure and in the Assembly’s open drafting, so that the holding would not rest on judicial moral philosophy alone. Gautam Bhatia’s account of the judgment fastens on exactly this discipline: in his reading, the plurality performs the transformative method, taking the Constitution’s commitments seriously as instruments for remaking the relation between state and person, while keeping each step tethered to a clause, a precedent or a structural inference that a later court can check. The opposing scholarly instinct, voiced in commentary that treats the verdict as judicial legislation, holds that recognition of unenumerated rights by structure is amendment by bench, and that the Assembly’s failure to enact the privacy clauses its advisers discussed in 1947 and 1948 should have been decisive against the right. The plurality met that objection on the record: the advisers’ discussions concerned express clauses on search, seizure and correspondence on the American and Irish models, and the Assembly’s choice of open language for Article 21, against the background of those discussions, is evidence of a drafting method rather than of a rejection of the underlying interest. On the documentary record, the plurality has the better of this exchange, and the verdict’s critics have generally shifted their fire from the existence of the right to the breadth of its applications, which is the ground on which the proportionality debate of the preceding sections is fought.

Privacy Across the Freedoms: Speech, Association and Movement

Although Article 21 is the right’s principal seat, the plurality was explicit that privacy draws life from the whole of Part III, and the cross-references do practical work. Freedom of speech and expression under Article 19(1)(a) presupposes thoughts not yet expressed, drafts not yet published and sources not yet revealed; a surveillance regime that records every enquiry a journalist makes and every meeting a dissident attends burdens expression before a single word is printed, and the doctrine of chilling effect, which the Court had entertained in earlier speech cases, acquires in Puttaswamy a constitutional home. Freedom of association under Article 19(1)(c) presupposes membership lists the state cannot demand at will, a point with a lineage in the Assembly’s own debates on political organisation. Freedom of movement under Article 19(1)(d) presupposes journeys that need not be reported in advance to authority, the precise interest Kharak Singh’s trackers had invaded. Privacy is the connective tissue among these freedoms, and the plurality’s location holding, which plants the right in Article 21 and in the freedoms together, means that a state action can fail on privacy grounds even where the enumerated freedom it burdens is hard to isolate.

The press dimension of the right received support from the Rajagopal line of cases and from the structure of the plurality’s informational facet. A newspaper’s unpublished material, a reporter’s notes and the identity of a confidential source sit within spatial and informational privacy simultaneously; compelled disclosure regimes must therefore run the proportionality sequence, naming their aim and proving their measure, rather than resting on the general proposition that investigation overrides confidence. Election law felt the same pressure in the years after the verdict, as litigation over the privacy of donors, voters and candidates asked courts to balance transparency interests against the associational privacy the plurality had described; the pattern’s appearance in fields far from Aadhaar, within three years of the verdict, is the behaviour the architecture reading predicts and the recognition reading cannot explain, since a bare recognition supplies no method for fields the recognising bench never mentioned.

The equality limb completes the cross-Part III structure. Article 14 forbids arbitrary state action, and the plurality treated privacy invasions distributed by algorithm or by administrative habit as capable of arbitrariness even where a statute is neutral on its face: a profiling system that sorts citizens by caste, faith or neighbourhood for differential treatment engages privacy and equality in one act. This intersection matters for the data economy, in which classification is the basic operation and its burdens fall unevenly on communities with the least capacity to contest them. The verdict’s insistence that privacy is the right of the welfare recipient and the pavement dweller as fully as of the property owner, stated by the plurality against the Union’s suggestion that privacy is a concern of the affluent, is an equality proposition dressed as a privacy proposition, and it arms later courts against the two-tier settlement in which the prosperous buy privacy as a service while the dependent surrender it as a condition of subsistence.

From Verdict to Statute: The Srikrishna Process and Data Protection

A constitutional right against the state does not by itself regulate the data economy, and the bench knew the limit of its own instrument. The committee under Justice B. N. Srikrishna, a former judge of the Supreme Court, was appointed by the Ministry of Electronics and Information Technology in July 2017, days before the nine-judge bench began its hearings, with a mandate to study data protection and draft a law. Its white paper of November 2017 framed the design questions in the verdict’s fresh vocabulary, and its report of 27 July 2018, A Free and Fair Digital Economy, arrived with a draft Personal Data Protection Bill that translated the plurality’s facets into regulatory machinery: consent as the default basis of processing, purpose limitation, collection limitation, rights of access, correction and erasure for the data principal, obligations of security and accountability for the data fiduciary, and a Data Protection Authority armed with penalties scaled to global turnover. The report invoked Puttaswamy as its constitutional charter and treated the proportionality test as the standard against which exemptions for state processing would be judged.

The draft’s treatment of the state drew the sharpest scholarly fire, and the fire is instructive about the verdict’s limits. Exemptions permitting government agencies to process data on grounds of national security and public order, drafted in wide terms, were criticised by digital rights advocates, Apar Gupta of the Internet Freedom Foundation among the most persistent, as a legislative attempt to win by exemption what the Union had lost in argument: a zone in which the proportionality sequence does not run. The revised bill introduced in the Lok Sabha in December 2019 retained wide state exemptions while enlarging the proposed authority’s governmental complexion, and its reference to a joint parliamentary committee extended the contest into 2020 without resolution. The sequence vindicates Mehta’s caution of September 2017 from an unexpected direction: the verdict’s architecture held firm in the courts while the statutory machinery that was supposed to complete it became the site of the real struggle, and the outcome of that struggle, in committee rooms rather than courtrooms, will decide how much of the architecture governs daily practice.

For the architecture reading, the statutory sequel supplies a final confirming datum. Observe the terms in which the Srikrishna debate was conducted: consent, purpose limitation, proportionality, safeguards, exemption, accountability. Every one of those terms is a term of the verdict’s art, carried from the plurality and the Kaul concurrence into a legislative process the bench did not control. A judgment remembered as a recognition would have supplied a slogan for that debate; Puttaswamy supplied its working concepts, and the exemptions controversy is intelligible as a fight over the third component of the map, the restriction test, conducted by other means. The fight also locates the verdict’s genuine vulnerability with precision. If Parliament enacts a data law whose state exemptions swallow the test, the right’s practical content will contract toward the pre-2017 position while its formal architecture stands untouched, and future historians of the doctrine will record that the architecture survived its builders’ intentions. The verdict cannot prevent that outcome; what it guarantees is that the contraction, if it comes, will be measurable against a stated standard, in open argument, rather than invisible.

The Man Who Lent the Case Its Name

Constitutional architecture is usually discussed in the passive voice, as though doctrines build themselves, and the Puttaswamy record resists that habit because its originating act was one retired judge’s refusal. K. S. Puttaswamy had served on the Karnataka High Court and had retired decades before the Aadhaar enrolment centres reached Karnataka; in 2012, at an age past eighty-five, he filed the writ petition that challenged the scheme’s constitutional foundations, objecting that a programme presented as voluntary was becoming a practical condition of access to pensions, subsidies and bank accounts, and that the biometric core of the programme took from the body what no subsequent apology could return. The petition’s framing is the seed of everything the nine-judge bench later built: it attacked the absence of a statute behind the scheme, the absence of consent worthy of the name, and the absence of any recognised right that could discipline the collection, thereby forcing the Union to defend the third absence, which it did, at the price of the 2015 argument that undid its position.

Other petitioners widened the record in the connected matters. A former soldier and engineer, S. G. Vombatkere, challenged the scheme’s expansion from a different direction, and petitioners drawn from the worlds of social security administration, banking and civil liberties placed before the Court the operational evidence about exclusion, authentication failure and database error that the plurality’s equality-inflected passages absorb. The consolidated record therefore contained, by 2017, both the high constitutional question and the low administrative evidence, and the verdict’s refusal to treat privacy as a luxury of the secure is anchored in that double record: the judges had read the affidavits of pensioners denied grain by failing fingerprints alongside the Attorney General’s submissions on constitutional structure, and the plurality’s sentence protecting the privacy of every person, rich or poor, is written against those affidavits.

The petition also illustrates the institutional patience the architecture required. Filed in 2012, referred in 2015, heard by nine judges in 2017 and answered on the Aadhaar merits only in 2018, the challenge occupied six years of the Court’s life and the final years of its principal petitioner’s ninth decade. Doctrine built at that tempo acquires a deliberateness that faster litigation rarely achieves; the six opinions of 24 August 2017 read as documents composed for a long shelf life, addressed to benches and legislatures the authors would never sit with, and the care taken over the extent of each overruling, over the staging of the restriction test and over the taxonomy of facets is the care of judges who expected to be read in 2047 as attentively as in 2017.

The Comparative Shelf the Plurality Built From

The plurality’s comparative method deserves separate treatment because it is the part of the architecture most easily mistaken for decoration. The opinion’s tour begins in 1890, with Samuel Warren and Louis Brandeis arguing in the Harvard Law Review that the common law already contained a right to be let alone, and it moves to Brandeis’s dissent in Olmstead v. United States in 1928, where wiretapping by federal agents was condemned in an opinion that lost the case and won the century. Katz v. United States in 1967 then supplies the plurality with its modern American anchor: the United States Supreme Court held that the Fourth Amendment protects people rather than places, and Justice Harlan’s concurrence framed the test of a reasonable expectation of privacy that Indian courts had already borrowed in the telephone tapping cases. Griswold v. Connecticut in 1965 contributes the structural method, the location of an unenumerated right in the penumbras formed by enumerated guarantees, which the plurality adapts to Part III with the substitution of dignity for the American opinion’s marital frame.

Beyond the American shelf, the plurality and the concurrences drew on the jurisprudence of the European Court of Human Rights under Article 8 of the European Convention, which protects respect for private and family life, home and correspondence, and which that court had applied to data retention, workplace monitoring and the press; on the German Federal Constitutional Court’s Census Act judgment of 1983, which derived informational self-determination from the general right of personality and held that a state may not compile a total registry of the person; and on authority from Canada and South Africa in which privacy appears as a dimension of dignity under charters drafted, like India’s, for societies emerging from systems of domination. The German census doctrine is the comparative source with the deepest affinity to the Aadhaar problem, since the German court confronted a state enumeration exercise defended on planning grounds and answered that a person who cannot know who holds which knowledge about him will adapt his behaviour to the unseen record, chilling the participation on which democratic order depends. The plurality’s adoption of that anxiety, translated into the language of autonomy, is the point at which comparative law stops being ornament in the verdict and becomes load-bearing.

The method has a discipline that distinguishes it from cherry-picking. Each foreign authority is introduced at the point of the argument it supports: Warren and Brandeis for the common law root, Olmstead for dissent as prophecy, Katz for expectation, Griswold for structure, the German court for data, Strasbourg for the balancing of press and private life under a qualified right. A critic can dispute any single borrowing, and scholars have disputed the plurality’s reading of the American cases as too smooth a story for a divided field, yet the architectural function of the shelf is independent of any single volume: the bench demonstrated that privacy as a fundamental right is the standard equipment of the post-1945 constitutional democracies and that India’s absence from that company since 1954 was the anomaly requiring justification. The Union’s argument, which treated the right as an exotic import, could not survive that demonstration, because the plurality showed the import had been sitting in the framers’ own library, in the Irish and American models the Assembly studied, since 1947.

Why the ADM Jabalpur Correction Matters Beyond Privacy

The overruling of the Emergency judgment is filed in the map as the second component of a privacy verdict, and its significance overflows that filing. ADM Jabalpur was never a privacy case; it was the case in which the Court held that the right to life itself could be placed beyond judicial protection by executive proclamation, and its overruling inside Puttaswamy restored a premise on which every other right in Part III silently depends, namely that the rights are the Constitution’s recognition of entitlements the state did not create and cannot recall. The plurality’s language on the point is the most quoted in the judgment after the holding itself: the judgments of the majority in the 1976 case were seriously flawed, the dissent of Justice Khanna was correct, and life and liberty are not the bounty of any government. Coming from a bench that included the son of a judge of that majority, the correction performed an institutional act that a mere distinguishing could not: the Court confessed error in terms, on the record, in a case large enough to carry the confession.

The correction also completed a statutory repair that had stood unfinished since 1978. The Forty-fourth Amendment, enacted by the Janata Parliament in the Emergency’s aftermath, had amended Article 359 so that the enforcement of Articles 20 and 21 could not be suspended even during a proclaimed Emergency, and had thereby prevented a repetition of the ADM Jabalpur order. Yet the amendment left the ADM Jabalpur reasoning formally alive: a future court, reading the reports, would have found the sole-repository theory of Article 21 still citable, its authority dented by amendment but never disowned by the judiciary that had authored it. Puttaswamy supplied the disowning. Doctrine and text were aligned at last, forty-one years after the failure and thirty-nine years after the amendment, and the alignment is a precondition of the first component’s security, since a privacy right nested in Article 21 is only as safe as Article 21 itself.

There is a final, quieter dimension. Justice Khanna died in 2008, three years after a portrait of him was unveiled in the Supreme Court courtroom where he had dissented, an honour the institution grants to few judges below the rank of Chief Justice, a rank his dissent had cost him. The 2017 plurality, in vindicating the dissent by name, converted a portrait’s tribute into doctrine, and every subsequent citation of Puttaswamy carries Khanna’s vindication inside it. When later benches invoke the verdict for propositions about data or sexuality, they cite a judgment whose second component is a monument to the proposition that a judge’s duty runs against the executive of the day even when the text has been arranged to excuse surrender. The architecture reading captures this too: the verdict’s plan includes, as a load-bearing element, a theory of judicial obligation under pressure, without which the remaining components are drawings rather than defences.

The Middle Line Revisited: Gobind to Selvi in Detail

The smaller-bench cases that filled the decades between Kharak Singh and the reference repay a closer reading than summary allows, because the nine-judge bench treated them as materials to be organised rather than errors to be cleared. Gobind v. State of Madhya Pradesh, decided in 1975, arose from surveillance regulations of the Madhya Pradesh police that authorised history sheets, close watch and domiciliary visits for habitual offenders. Justice Mathew’s opinion is a study in judicial ambivalence: it assumed in the petitioner’s favour that a privacy right of some dimension exists under Article 21, imported the American compelling interest standard for its invasion, and then upheld the regulations on the ground that the class of persons watched was narrow and the public interest in preventing crime substantial. The opinion closed with an exhortation that the right be developed case by case, an invitation the Court accepted for forty years. Read after Puttaswamy, Gobind’s ambivalence is diagnostic: a bench of three judges could gesture at the right, borrow a test and decide a case, yet could not certify the foundation, and the gesture-plus-upholding pattern repeated across the decades is what a missing architecture looks like in practice.

R. Rajagopal v. State of Tamil Nadu in 1994 moved the right into the law of expression. The case concerned the proposed serialisation, in a Tamil magazine, of the autobiography of a convicted prisoner known as Auto Shankar, which prison authorities sought to restrain. The Court held that the right to privacy includes a right to be let alone in matters of family, marriage, procreation and motherhood, that publication based on public records stands outside the right, and that public officials act at the risk of comment on their public conduct. The opinion’s sources were American tort doctrine filtered through Indian constitutional text, and its significance for 2017 lies in its method: privacy appears as a working rule of decision between private parties and the press, developed without any engagement with the Sharma denial, as though the denial concerned a different subject. PUCL v. Union of India in 1997 then converted the right into procedure. Faced with telephone tapping under Section 5(2) of the Indian Telegraph Act, the Court read privacy into Article 21, held tapping a serious invasion of the right, and, finding the statutory procedure empty, prescribed safeguards of authorisation, review committee oversight, duration limits and record destruction that operated as law until Parliament should act. The PUCL safeguards are the direct ancestor of the procedural-guarantee limb of the Puttaswamy test: the 2017 bench generalised a device invented in 1997 for telephones into a standing requirement for covert or continuous invasions of every kind.

The line completed itself in the decade before the reference. District Registrar and Collector, Hyderabad v. Canara Bank in 2005 held that documents a customer lodges with a bank carry a privacy interest that a tax statute’s inspection power must respect in proportion. Suchita Srivastava v. Chandigarh Administration in 2009 protected reproductive choice as a dimension of personal liberty for a woman with intellectual disability whose pregnancy the state proposed to terminate, and did so in the vocabulary of autonomy that the plurality would later adopt wholesale. Selvi v. State of Karnataka in 2010 barred involuntary narcoanalysis, polygraph and brain-mapping, grounding mental privacy in the protection against self-incrimination and in Article 21’s substantive content, and thereby extended the right to the interior of the mind, the facet furthest from the search-and-seizure world of Sharma. National Legal Services Authority v. Union of India in 2014, recognising transgender persons’ right to self-identified gender, treated identity as an aspect of dignity and self-expression that the state must record as declared rather than assign by inspection. Arranged in sequence, the middle line had already decided, by 2014, that privacy protects the home, the telephone, the bank file, the womb, the mind and the declared identity of the person. The nine-judge bench’s task was not invention; it was ratification with foundations, and the verdict’s authority draws strength from the fact that the doctrine it certified had been load-tested in litigation for four decades.

The Test at Work: Proportionality Before and After 2017

The restriction test has its own Indian prehistory, which the plurality acknowledged and which explains the test’s rapid uptake after 2017. In Modern Dental College and Research Centre v. State of Madhya Pradesh in 2016, a Constitution Bench had already adopted a four-prong proportionality standard for restrictions on the Article 19(1)(g) freedom of profession, drawing on the structured doctrine developed by courts in Germany, Canada and Israel and systematised in Aharon Barak’s treatise Proportionality, published in 2012. The four prongs, proper purpose, rational connection, necessity and balancing in the strict sense, were therefore part of Indian constitutional method before the privacy reference was heard. The Puttaswamy plurality’s threefold formulation, law, legitimate aim and proportionality, and Justice Kaul’s addition of procedural guarantees, should be read as the adaptation of that standing method to a right without an internal limitation clause. Article 19 carries its restrictions in its own text, in clauses (2) to (6); Article 21 carries none, and the test in Puttaswamy is the Court’s answer to the question of where the limits on the Article 21 privacy right are to come from: not from legislative grace, and not from judicial intuition case by case, but from a stated sequence that any bench can run and any litigant can contest.

Running the sequence against a familiar regime shows its teeth. Take the telephone tapping framework the PUCL judgment constructed. Legality is satisfied by Section 5(2) of the Telegraph Act, a statute of the legislature rather than an executive circular. Legitimate aim is satisfied by the statutory grounds of public emergency and public safety, interests the Constitution recognises. Proportionality then does the discriminating work: a tap authorised without recorded reasons, renewed without review, or retained without destruction of intercepts fails the necessity and safeguard limbs even though it passes the first two stages, and the PUCL safeguards survive inside the Puttaswamy framework as the content of those limbs for this technology. The same sequence, applied to a bulk data retention regime of the kind the Aadhaar litigation concerned, separates the questions that polemic tends to fuse: whether identification for welfare is a legitimate aim, which the 2018 majority and the 2017 plurality both accepted in principle; whether universal biometric enrolment is a suitable and necessary means, where evidence about alternatives and error rates decides; and whether the balance between administrative gain and the surrender of bodily identifiers holds at the scale of a billion enrolments, where the dissent and the majority of 2018 parted. A test that can locate disagreement with that precision, stage by stage, is doing analytical work no slogan about recognition can perform.

The staging also allocates burdens in a way that changes litigation behaviour. At the legality stage the state produces its statute or fails. At the aim stage the state names its purpose in terms a court can test for constitutional legitimacy, and purposes that cannot be named in open court, such as the general convenience of the administration, fail quietly at this stage in well-run challenges. At the necessity stage the evidential initiative shifts: the state must show that a less invasive adequate alternative was considered and why it falls short, which forces ministries to generate, before litigation, the comparative assessments that good administration should generate anyway. At the balancing stage the court states the exchange rate it is applying between the aim and the right, in reasons that appellate benches and scholars can attack. The post-2017 unevenness documented in the adjudication section consists, in case after case, of a bench collapsing one of these stages, accepting an aim without naming it or a necessity claim without evidence; the stages give the critic the coordinates of the collapse. That is the deepest sense in which the verdict is architecture: it built the grid on which its own successors can be graded.

Judgment Day: The Order of 24 August 2017 and Its Reception

The operative act of 24 August 2017 was short enough to be read aloud in minutes, and its brevity against the five hundred and forty-seven pages of reasoning is a lesson in how this Court separates decision from justification. Chief Justice Khehar announced that the bench was unanimous and read the order signed by all nine judges. The order declared that the right to privacy is protected as an intrinsic part of the right to life and personal liberty under Article 21 and as a part of the freedoms guaranteed by Part III of the Constitution. It recorded that the decisions in M. P. Sharma and Kharak Singh, to the extent they held that such a right is not protected by the Constitution, stand overruled. It disposed of the reference and returned the Aadhaar challenges to the appropriate bench for decision on their merits in the light of the declared law. Three sentences of operative text thus performed the verdict’s first two components, location and overruling, while the reasoning that supported them and the test and pattern that completed them waited in the opinions published with the order.

The reference format explains the order’s shape. The nine judges had been asked questions, not given a dispute to dispose of, and their formal task ended with the answers; no direction issued against Aadhaar, no enrolment was disturbed, and the programme whose challenge had generated the case continued under interim orders until the five-judge bench ruled in 2018. Commentary that scored the day as a defeat or victory for Aadhaar therefore misread the instrument, a confusion the Aadhaar-context reading of the verdict inherits. The day’s true transaction was doctrinal: a proposition about the Constitution’s structure, contested since 1954, was settled by the largest bench the question could command, and every future privacy case in India would begin from the order’s three sentences rather than from Sharma’s silence.

The political reception performed an irony the opinions had prepared. The Union government, whose Attorney General of 2015 had denied the right’s existence, welcomed the verdict through the Law Minister, Ravi Shankar Prasad, who stated that the government had itself taken the position that privacy is a fundamental right and that the ruling affirmed the administration’s commitment to digital governance with safeguards. Finance Minister Arun Jaitley, who had piloted the Aadhaar Act through Parliament in 2016, addressed the Rajya Sabha on the judgment’s implications and accepted the proportionality framework as the standard the Act would meet. Opposition speakers claimed the verdict as a rebuke of the government’s litigation stance and of the identification programme’s compulsory drift. The competing claims of paternity are themselves evidence for the architecture reading: a unanimous nine-judge statement of structure belongs to no party, and politicians of every camp reached for its vocabulary within hours because the vocabulary had become, in a single morning, the obligatory language of Indian rights talk. A judgment that merely recognised a right in a pending dispute would have been scored and filed; a judgment that rearranged the architecture was quoted in Parliament the same week.

The press and the academy sorted themselves along the lines the adjudication section has traced. Editorial pages treated the verdict as a landmark of the first rank, several reaching for the comparison with Kesavananda Bharati that the most-consequential framing of this series also employs. The scholarly journals took longer and argued harder, over the natural right register, over the test’s staging, over the omission of any direct ruling on the surveillance statutes, and those arguments, conducted through 2018 and 2019 in the Indian Journal of Constitutional Law, the Economic and Political Weekly and the National Law School journals, constitute the disagreement this article has adjudicated. What none of the receptions disputed, from the Law Minister to the verdict’s sternest academic critics, was the fact of settlement on the existence and location of the right; the argument had already moved to application, which is to say, to architecture in use.

The Verdict in the Long Arc of the Founding Settlement

Return, at the end of the analysis, to the founding frame in which this series reads every constitutional event. The Constituent Assembly that drafted Part III between 1946 and 1949 worked in the shadow of a colonial state that had governed through surveillance regulations, sedition law and preventive detention, and the Assembly’s members argued over the rights chapter with that machinery in living memory; the account given in the constitutional rights framework records how the Assembly chose open-textured guarantees over enumerated detail. Puttaswamy is the late vindication of that drafting choice. The Assembly could not have listed informational privacy in 1949, because the information economy did not exist to be imagined; what the Assembly could do, and did, was to write liberty in language wide enough for a nine-judge bench in 2017 to find the unlisted right inside the listed one, seventy years after the text was adopted. The verdict thus closes a circuit opened at the founding: the framers trusted structure over enumeration, Sharma distrusted the trust, Kharak Singh half-repaired it, the Emergency broke it, the Forty-fourth Amendment patched the text, and Puttaswamy restored the structural method as doctrine.

The namable claim can be restated with the evidence complete. The 2017 Puttaswamy verdict established right-architecture; the architectural pattern shapes the post-2017 rights trajectory. Privacy entered 24 August 2017 as a contested inference from decided cases and left it as a located, faceted, test-governed fundamental right, with the Emergency’s contrary theory overruled in terms and the Subba Rao dissent of 1962 absorbed into orthodoxy. The benches of September 2018 did not have to invent a method for Aadhaar, for Section 377 or for Sabarimala, because the method was waiting for them in the plurality’s paragraphs. Whatever the unevenness of later application, and the record of unevenness is real and documented, the standard against which application is measured stands fixed in a unanimous nine-judge verdict that no smaller bench can displace. That is what architecture means in a constitutional court: not that every later room is well built, but that the foundations, the load-bearing walls and the plan are settled, and that argument about the building proceeds from the plan.

Frequently Asked Questions

Q: What is the Puttaswamy verdict?

The Puttaswamy verdict is the unanimous judgment of a nine-judge bench of the Supreme Court of India in Justice K. S. Puttaswamy (Retired) v. Union of India, delivered on 24 August 2017. Chief Justice J. S. Khehar and Justices Chelameswar, Bobde, Agrawal, Nariman, Sapre, Chandrachud, Kaul and Nazeer held, through six opinions, that privacy is a fundamental right protected as an intrinsic part of Article 21 and of the freedoms in Part III of the Constitution. The bench overruled M. P. Sharma and Kharak Singh to the extent of their denials of the right, disowned the Emergency-era reasoning of ADM Jabalpur, and prescribed a test of legality, legitimate aim and proportionality for any state invasion of the right, before returning the Aadhaar challenges to a smaller bench.

Q: Is privacy a fundamental right in India?

Yes. Since 24 August 2017 privacy has ranked as a fundamental right under the Constitution of India, enforceable against the state through writ petitions under Article 32 in the Supreme Court and Article 226 in the High Courts. The nine-judge bench in Puttaswamy held that the right is an intrinsic part of the right to life and personal liberty in Article 21 and draws support from the freedoms guaranteed across Part III, including speech, association and movement under Article 19. The right is not absolute: a law invading privacy survives if it rests on valid legislation, pursues a legitimate state aim and is proportionate to that aim, with procedural safeguards against abuse. Before 2017 the position was unsettled, because M. P. Sharma in 1954 and Kharak Singh in 1962 had declined to recognise a general right, and Puttaswamy overruled both decisions to the extent of that refusal.

Q: What is the proportionality test?

The proportionality test is the sequence a court applies to decide whether a state invasion of privacy is constitutionally permitted. In the Puttaswamy plurality’s formulation the invasion must first rest on a law, since executive action without legal authority fails at once. It must second pursue a legitimate state aim of the kind the Constitution accepts as a ground for limiting freedoms. It must third be proportionate, meaning the measure is suited to the aim, no more invasive than the aim requires, and balanced so that the public gain is not bought at a cost to the right that the aim cannot justify. Justice Kaul’s concurrence added procedural guarantees against abuse as a further requirement. The test descends from the structured proportionality doctrine of German, Canadian and Israeli courts and from the four-prong standard the Supreme Court had adopted in Modern Dental College in 2016.

Q: Did Puttaswamy overrule ADM Jabalpur?

Yes, in substance and by name. The Puttaswamy plurality held that the majority judgments in ADM Jabalpur v. Shivkant Shukla of 28 April 1976 were seriously flawed, endorsed Justice Khanna’s lone dissent as the correct statement of law, and rejected the 1976 theory that Article 21 is the sole repository of the right to life and liberty, so that a Presidential order could suspend the right’s protection altogether. The plurality reasoned that life and liberty are inalienable entitlements the Constitution recognises rather than grants, from which it follows that no proclamation can extinguish them. The overruling completed a repair the Forty-fourth Amendment had begun in 1978 by placing Articles 20 and 21 beyond suspension during an Emergency; the amendment had changed the text, and Puttaswamy withdrew the judicial reasoning that had once blessed the suspension.

Q: Who is K.S. Puttaswamy?

K. S. Puttaswamy is the retired judge of the Karnataka High Court whose writ petition of 2012 gave the case its name. Past eighty-five years of age when he filed, he challenged the Aadhaar identification programme on the grounds that enrolment presented as voluntary was becoming a practical condition of pensions, subsidies and banking, that the programme rested on executive action without a governing statute, and that the compulsory surrender of fingerprints and iris scans invaded a privacy interest the Constitution should protect. His petition, joined by other challengers including S. G. Vombatkere, produced the reference of August 2015 on whether privacy is a fundamental right, the nine-judge bench of July 2017 and the verdict of 24 August 2017. The Aadhaar merits themselves were decided by a five-judge bench in September 2018 in the judgment that bears his name as well.

Q: What does Article 21 say?

Article 21 of the Constitution of India provides that no person shall be deprived of his life or personal liberty except according to procedure established by law. The text is a single sentence of twenty words in the Constitution’s Part III chapter of fundamental rights, and it applies to every person, citizen and non-citizen alike, unlike the freedoms of Article 19 which are confined to citizens. Through interpretation the Supreme Court has read into its spare language the rights to livelihood, health, a clean environment, education at the elementary stage, speedy trial and, after the Puttaswamy verdict of 24 August 2017, privacy as an intrinsic part of life and personal liberty. Since the Forty-fourth Amendment of 1978, the enforcement of Article 21 cannot be suspended even during a proclaimed Emergency, a protection Article 20 shares and no other fundamental right enjoys in the same absolute terms.

Q: How does Puttaswamy affect Aadhaar?

Puttaswamy decided the premise on which the Aadhaar litigation turned without deciding the Aadhaar case itself. By holding privacy to be a fundamental right and prescribing the test of legality, legitimate aim and proportionality, the nine-judge bench of August 2017 fixed the standard the five-judge bench would apply to the Aadhaar Act when the merits were heard. In the Aadhaar verdict of 26 September 2018 the majority used that standard to uphold the programme’s core for welfare delivery while striking down Section 57 on private sector use, restricting retention of authentication records and reading down disclosure provisions, and Justice Chandrachud’s dissent used the same standard to argue the Act should fall entirely. The 2017 verdict also meant that every expansion of Aadhaar into new services thereafter had to be justified as a proportionate invasion of a fundamental right rather than defended as a mere administrative convenience.

Q: Why did the privacy question require a bench of nine judges?

Bench size in the Supreme Court follows the strength of the precedent under challenge. The authorities denying a general fundamental right to privacy were M. P. Sharma, decided by eight judges in 1954, and Kharak Singh, decided by six judges in 1962, and a bench may not overrule a decision of a larger or equal bench. When the three-judge bench hearing the Aadhaar challenges concluded in August 2015 that the correctness of those authorities had to be settled first, and when the Union of India rested its case on them, the only constitutionally clean route was a bench larger than eight. A five-judge Constitution Bench in July 2017 therefore directed the reference to nine judges, who heard argument from 19 July to 2 August 2017 and answered unanimously on 24 August, overruling the older decisions to the precise extent of their denials.

Q: What did M.P. Sharma decide in 1954?

M. P. Sharma v. Satish Chandra, decided on 15 March 1954 by a bench of eight judges, concerned search and seizure of documents from companies and traders under investigation. The petitioners argued that compelled production of their papers invaded a privacy protection comparable to the Fourth Amendment of the United States Constitution. The bench rejected the claim, holding that the framers of the Indian Constitution had not written a privacy guarantee into Part III, that the power of search and seizure is a necessity of social defence against crime, and that the American analogy failed because the Indian text omitted what the American text contained. The decision stood for sixty-three years as the largest-bench authority against a fundamental right to privacy, until the nine judges in Puttaswamy overruled it to the extent that it denied constitutional protection to privacy.

Q: What argument did Attorney General Mukul Rohatgi present against a fundamental right to privacy?

Arguing for the Union of India in August 2015, Attorney General Mukul Rohatgi submitted that the Constitution of India contains no fundamental right to privacy at all. His case rested on three supports: the framers had considered privacy-type protections and declined to enact a general guarantee; the eight judges in M. P. Sharma and the six judges in Kharak Singh had refused recognition, and smaller benches lacked the authority to displace them; and privacy, so far as Indian law protected it, was a common law interest that Parliament could define, qualify or override by statute. The submission forced the reference to a larger bench, because the Aadhaar challenges could not proceed while the largest benches on the reports denied the right they invoked. By the hearings of July 2017 the Union’s position had softened under a successor Attorney General, yet the bench answered the 2015 argument in its full form and rejected it unanimously.

Q: What did the bench say about the claim that privacy is an elitist concern?

The bench rejected the claim directly. In the course of the Aadhaar argument the Union’s side had suggested that privacy is a concern of the prosperous, of little weight to citizens whose pressing need is food, subsidy and state recognition, and that a poor person would surrender data without hesitation in exchange for welfare. The plurality answered that privacy protects the poor with the same force as the rich: the choices a welfare recipient makes about food, faith, association and family are as intimate as any property holder’s, and a settlement under which subsistence is priced in surrendered rights would divide the Constitution’s protection by income. Justice Nariman’s concurrence treated the beneficiary of welfare as a holder of rights rather than a recipient of largesse. The point mattered beyond rhetoric, because Aadhaar’s subjects are disproportionately the dependent poor, and a right that thinned as income fell would have failed the very population the programme enrols.

Q: What is informational privacy in the opinions?

Informational privacy is the facet of the right concerned with control over data about a person: which particulars are collected, for what purpose, by whom they are held, to whom they are disclosed and how long they are retained. The Puttaswamy plurality treated the facet as central to the right in an information age, in which profiles assembled from fragments surrendered for unrelated purposes can be used to predict, sort and influence the person described. Justice Kaul’s concurrence developed the theme furthest, warning that data collected for one purpose can be mined for another, that profiling assigns persons to categories they never chose, and that protection requires statutory safeguards binding private holders of data as well as the state. The facet is the bridge between the 2017 verdict and data protection law: the Srikrishna committee’s draft bill of July 2018 translated informational privacy into consent requirements, purpose limitation and a regulatory authority.

Q: How did the verdict influence the Navtej Johar judgment of 2018?

Navtej Singh Johar v. Union of India, decided on 6 September 2018, read down Section 377 of the Indian Penal Code so that consensual sexual relations between adults were no longer criminal, and its five judges built the analysis on Puttaswamy as premise. Sexual orientation was placed within the decisional facet of privacy the plurality had named at the right’s core; identity was protected through the dignity anchor; and the colonial provision was measured against the proportionality discipline and found indefensible, since the state could name no legitimate aim that criminalising adult intimacy served. Justice Chandrachud, author of the Puttaswamy plurality, extended its logic from protection against intrusion to recognition of identity, while Justice Indu Malhotra’s opinion framed the outcome as an apology the legal order owed. The thirteen months between the two verdicts show the 2017 pattern in operation: facet identified, location fixed, test applied, contrary precedent discarded.

Q: What did the Srikrishna data protection committee recommend in 2018?

The committee chaired by Justice B. N. Srikrishna, appointed by the Union government in July 2017 while the nine-judge bench sat, reported on 27 July 2018 under the title A Free and Fair Digital Economy, accompanied by a draft Personal Data Protection Bill. It recommended that processing of personal data rest on consent or on narrowly defined alternative grounds, that collection be limited to stated purposes, that persons hold rights of access, correction and erasure against those who hold their data, that holders bear duties of security and accountability as fiduciaries, and that a Data Protection Authority enforce the scheme with penalties scaled to turnover. The report treated the Puttaswamy verdict as its constitutional charter and the proportionality test as the standard for state exemptions, which critics argued the draft itself set too wide. A revised bill reached the Lok Sabha in December 2019 and went to a joint parliamentary committee.

Q: Which foreign authorities did the plurality draw upon?

The plurality built its comparative argument in layers. From the United States it took Warren and Brandeis’s 1890 Harvard Law Review essay The Right to Privacy, Justice Brandeis’s dissent in Olmstead v. United States in 1928 defending the right to be let alone, Katz v. United States in 1967 with its protection of reasonable expectations of privacy, and Griswold v. Connecticut in 1965, which located an unenumerated right in the penumbras of enumerated guarantees and supplied the structural method the plurality adapted to Part III. From Europe it drew on Article 8 of the European Convention on Human Rights and its case law on data and private life, and on the German Federal Constitutional Court’s Census Act judgment of 1983, which derived informational self-determination from the right of personality. Canadian and South African authority appears on privacy as a dimension of dignity under post-dictatorship and post-apartheid charters.

Q: What is the difference between the plurality opinion and the unanimous result in Puttaswamy?

The result in Puttaswamy was unanimous: all nine judges agreed that privacy is a fundamental right, and the order of the Court, signed by every member of the bench, states that single proposition and the overrulings that accompany it. The reasoning was not unanimous in form, because the bench produced six opinions. The plurality opinion of Justice Chandrachud, joined by Chief Justice Khehar and Justices Agrawal and Nazeer, speaks for four judges and carries the fullest statement of the right’s location, facets and test. Justices Chelameswar, Bobde, Nariman, Sapre and Kaul wrote concurrences that reach the same result by partly different routes, adding the three-facet scheme of repose, sanctuary and intimate decision, the natural right grounding, the taxonomy of interests and the digital age warnings. Later courts treat the plurality as the working text because it is the largest single bloc of reasoning inside the unanimous outcome.

Q: How does the verdict bear on privacy claims against private parties?

The fundamental right declared in Puttaswamy runs primarily against the state, since Part III of the Constitution binds state action, and the verdict did not by itself create a general damages claim against a company or a neighbour who invades privacy. The bench nevertheless marked the horizontal problem expressly. Justice Kaul’s concurrence warned that private holders of data, platforms, employers and data brokers can invade informational privacy at a scale rivaling government, and that a right confined to state action would miss the characteristic invasions of a digital economy. The plurality likewise noted that privacy claims against non-state actors would require legislative and common law development. That development is the office of data protection legislation of the kind the Srikrishna committee drafted in 2018, and of the older tort and publication doctrines the Rajagopal judgment of 1994 had already built between private parties and the press.

Q: What three facets of privacy did Justice Chelameswar identify?

Justice Chelameswar’s concurrence organised privacy as three related freedoms. The first is repose, the freedom to be left alone, the negative liberty against intrusion into body, home and correspondence that the older cases on surveillance and search had protected in fragments. The second is sanctuary, the protected space around the intimate life of the person, covering family, marriage and the relationships within which identity forms, into which the state may not enter without grave justification. The third is intimate decision, the freedom to make the choices by which a life is authored, in faith, in partnership, in reproduction and in the presentation of the self. The scheme matters because each facet answers to a distinct kind of state action: intrusion, occupation and command. Justice Chelameswar also warned against defining privacy so widely that it swallows every liberty interest, a caution later benches cite when they assign a claim to its facet before applying the test.

Q: How does privacy connect Article 21 with the freedoms of Article 19?

The plurality located privacy primarily in Article 21 while holding that the right also lives in the freedoms of Part III, and the connection is functional. Speech under Article 19(1)(a) presupposes thoughts formed unobserved, drafts unpublished and sources unrevealed, so surveillance of enquiry and association burdens expression before any word is printed. Association under Article 19(1)(c) presupposes membership the state cannot catalogue at will, and movement under Article 19(1)(d) presupposes journeys that need not be reported to authority in advance, the interest invaded by the tracking struck at in Kharak Singh. Article 21 supplies the right’s seat because life and personal liberty name the person as such; Article 19 supplies reinforcement because each enumerated freedom describes an activity privacy shelters. The double location means a state action can fail on privacy grounds even where the particular freedom it burdens is difficult to isolate, and it explains why the order of the Court placed the right in Article 21 and in Part III together.

Q: What did the verdict leave for Parliament to complete?

The verdict fixed the constitutional standard and left the machinery to legislation, in three fields especially. The first is data protection: consent rules, purpose limitation, rights against data holders and a regulatory authority, drafted by the Srikrishna committee in 2018 and pending before a joint parliamentary committee in the revised bill of 2019. The second is surveillance law: interception in India still rests on the Indian Telegraph Act of 1885 and the safeguards the PUCL judgment prescribed in 1997, and a statute matching the Puttaswamy requirements of legality, proportionality and procedural guarantee remained unwritten through 2020. The third is remedies against private invaders of privacy, which the concurrences assigned to legislation and to the development of tort doctrine. The division of labour is deliberate in the opinions: the Court declared what the Constitution forbids and requires, while the design of authorities, penalties and procedures belongs to Parliament, whose delay leaves the declared right dependent on case-by-case litigation.