Section 66A — Striking Down (Shreya Singhal)

No single provision of the Information Technology Act, 2000 has shaped Indian free-speech jurisprudence as decisively as Section 66A. Inserted by the Information Technology (Amendment) Act, 2008 and notified into force on 27 October 2009, it punished the sending of "grossly offensive" or "menacing" electronic messages with up to three years' imprisonment. For six years it was wielded against cartoonists, students, professors and ordinary citizens, until a two-judge Bench of the Supreme Court in Shreya Singhal v. Union of India, (2015) 5 SCC 1, declared it unconstitutional in its entirety on 24 March 2015. This chapter unpacks the text of the provision, the architecture of the judgment, the doctrines it imported into Indian law, and the curious afterlife of a "zombie" section that police continued to invoke long after its judicial death.

The Provision and Its Origins

Section 66A did not exist in the original Information Technology Act, 2000. It was inserted by the Information Technology (Amendment) Act, 2008 (Act 10 of 2009) and brought into force with effect from 27 October 2009. The amendment was a hurried legislative response to the perceived inadequacy of the parent Act in addressing online harassment, spam and offensive electronic communication, and it was passed by the Lok Sabha without substantive debate in December 2008 in the shadow of the Mumbai terror attacks.

The section read: "Any person who sends, by means of a computer resource or a communication device,— (a) any information that is grossly offensive or has menacing character; or (b) any information which he knows to be false, but for the purpose of causing annoyance, inconvenience, danger, obstruction, insult, injury, criminal intimidation, enmity, hatred or ill will, persistently by making use of such computer resource or a communication device, or (c) any electronic mail or electronic mail message for the purpose of causing annoyance or inconvenience or to deceive or to mislead the addressee or recipient about the origin of such messages, shall be punishable with imprisonment for a term which may extend to three years and with fine." An Explanation clarified that "electronic mail" and "electronic mail message" covered messages or information created or transmitted on a computer, computer system or communication device, including attachments in text, image, audio, video and any other electronic record.

The drafting borrowed loosely from Section 127 of the United Kingdom's Communications Act, 2003, which penalises "grossly offensive" telephonic messages. But the Indian transplant was far broader: it reached every form of electronic information, carried a three-year custodial term against the UK's six months, and strung together a chain of subjective adjectives—annoyance, inconvenience, danger, obstruction, insult, ill will—that had no anchor in any recognised category of criminal harm.

Anatomy of the Vagueness

The constitutional vice of Section 66A lay in its vocabulary. Terms such as "grossly offensive," "menacing character," "annoyance" and "inconvenience" were nowhere defined in the Act, and the Supreme Court found no settled judicial meaning that could lend them precision. What is grossly offensive to one reader may be entirely innocuous, even agreeable, to another. The provision thus made the criminality of speech turn not on any objective standard but on the subjective sensibility of the recipient or of the investigating officer.

Justice Nariman, writing for the Bench in Shreya Singhal, illustrated the problem by contrasting Section 66A with the carefully delimited offences in the Indian Penal Code. Defamation, obscenity and incitement each have a defined contour; "causing annoyance" has none. The Court observed that the section did not even require the information to be false, except in clause (b), so that a true statement of fact, if it annoyed someone, could attract three years' imprisonment. The absence of any requirement of mens rea directed at a recognised harm meant the provision swept within its net political satire, trade-union appeals, religious debate and the daily friction of online disagreement. This is the doctrine of unconstitutional vagueness, drawn into Indian law from American precedent and discussed further in the chapter on definitions under the Act.

The Court grounded the vagueness doctrine in the rule-of-law value that a penal statute must give fair notice of what it forbids. A citizen of ordinary intelligence, the Bench reasoned, must be able to know in advance whether his contemplated conduct is lawful; a law that fails this test invites arbitrary and discriminatory enforcement, leaving the line between the permitted and the forbidden to be drawn after the fact by the very officials who prosecute. Section 66A failed on both counts. It gave no fair warning, and it conferred on the police a roving discretion to decide, post hoc, whether a message had crossed an undefined threshold of offensiveness. The Court contrasted this with provisions whose contours had been settled by long judicial exposition, such as the obscenity standard refined in Aveek Sarkar v. State of West Bengal, (2014) 4 SCC 257, where the "community standards" test gave the term a workable meaning. No such saving gloss was available for "grossly offensive" or "annoyance."

The Facts and the Trigger Cases

Shreya Singhal was a public-interest petition filed under Article 32, triggered by a cluster of arrests that had exposed the section's potential for abuse. In November 2012, two young women in Palghar, Maharashtra—Shaheen Dhada and Rinu Srinivasan—were arrested under Section 66A after one of them questioned, in a Facebook post, the shutdown of Mumbai following the death of a political leader, and the other merely "liked" the post. The episode produced national outrage.

Other arrests followed the same pattern. Professor Ambikesh Mahapatra of Jadavpur University was arrested for circulating a cartoon of the West Bengal Chief Minister. Businessman Ravi Srinivasan was booked for a tweet about a politician's son. In each case the conduct was plainly within the realm of political comment, yet the elastic language of Section 66A allowed the police to treat it as a cognisable offence. Shreya Singhal, then a law student, filed the lead petition in 2012; several other petitions, including one by the People's Union for Civil Liberties (PUCL), were heard together. The petitioners challenged not only Section 66A but also Section 69A and the blocking rules, and Section 79 governing intermediary liability.

The Bench and the Judgment

The matter was decided on 24 March 2015 by a Division Bench of Justice J. Chelameswar and Justice Rohinton Fali Nariman, the judgment being authored by Justice Nariman. The Court's central conclusion was unambiguous: Section 66A is struck down in its entirety as violative of Article 19(1)(a) of the Constitution and not saved by Article 19(2). It was also held to be "not saved" under the public-order, defamation or incitement heads, because it bore no proximate or rational nexus to any of the eight permitted grounds of restriction.

Crucially, the Court declined to read down the provision or to sever its offending parts. Counsel for the Union had urged that the section could be narrowed by judicial interpretation to save it, and gave an assurance that it would be administered with care. The Bench refused both invitations. It held that the assurance of a particular government could not cure a constitutional defect, since a statute, once enacted, does not depend for its validity on the temperament of the executive that administers it. And it held that where a statute is so vague that no reasonable construction can cure its overbreadth, the only constitutional course is to strike it down. The judgment thereby cemented the principle that the chilling effect of a vague penal speech law is itself a constitutional injury, independent of any actual prosecution.

The Court also dealt with the Union's argument founded on the presumption of constitutionality—the principle that a duly enacted law is presumed valid and the burden lies on the challenger. While accepting the presumption as a general rule, the Bench held that it operates with reduced force in the domain of fundamental freedoms, and cannot rescue a provision whose very terms are incapable of objective application. In free-speech cases, the Court said, the ordinary deference to legislative wisdom yields to the heightened scrutiny that the guarantee in Article 19(1)(a) demands.

Discussion, Advocacy and Incitement

The doctrinal heart of Shreya Singhal is its tripartite classification of speech into discussion, advocacy and incitement. Justice Nariman held that mere discussion or even advocacy of a cause, however unpopular, lies at the very heart of the Article 19(1)(a) guarantee and cannot be curtailed. It is only when speech crosses into incitement—and incitement with a proximate connection to one of the Article 19(2) grounds, such as public order—that the State may step in to restrain it.

This framework borrowed openly from American First Amendment jurisprudence, in particular the "clear and present danger" lineage running from Schenck v. United States through Justice Brandeis's concurrence in Whitney v. California, culminating in the imminent-lawless-action test of Brandenburg v. Ohio. The Court found that Section 66A penalised information that need not incite anyone to do anything; "written words may be sent that may be purely in the realm of discussion or advocacy of a particular point of view." Because the section captured pure discussion and advocacy without any requirement of incitement, it failed the constitutional test at the threshold.

Public Order and Article 19(2)

The Union had argued that Section 66A was a reasonable restriction in the interest of public order. The Court rejected this on the authority of its own foundational precedents on Article 19(2). In Romesh Thappar v. State of Madras, AIR 1950 SC 124, the Court had held that public order is an aspect distinct from, and narrower than, the security of the State, and that a restriction must have a proximate, not a remote or fanciful, connection with the apprehended disorder.

The Bench leaned heavily on Superintendent, Central Prison, Fatehgarh v. Ram Manohar Lohia, AIR 1960 SC 633, where the Court insisted on a "proximate" rather than "remote or far-fetched" link between the speech and the public-order consequence, likening the required nexus to a spark in a powder keg. Measured against this standard, Section 66A failed comprehensively: causing "annoyance" or "inconvenience" has no proximate connection with public disorder, and the section did not even require any tendency to disturb public tranquillity. The provision therefore could not be salvaged under the public-order head, nor under defamation, decency or any other clause of Article 19(2).

The Court was equally emphatic that the State's interest must be tested against each enumerated ground separately. It went head by head through Article 19(2): the section had nothing to do with the sovereignty or integrity of India, the security of the State, friendly relations with foreign States, decency or morality, contempt of court, defamation or incitement to an offence. A message that is merely "annoying" might be none of these things, yet the section made it punishable. The mismatch between the offence created and the grounds that could justify it was, in the Court's words, complete; the section could not be brought within Article 19(2) by any process of legal alchemy.

Overbreadth and the Chilling Effect

Beyond vagueness, the Court invoked the related American doctrines of overbreadth and chilling effect. A law is overbroad when, in addition to legitimately proscribable conduct, it sweeps within its ambit a substantial quantum of protected speech. Section 66A was overbroad because its sprawling adjectives criminalised vast swathes of constitutionally protected expression alongside any genuinely harmful messages.

The chilling effect concerned the Court still more. Because citizens could not know in advance whether their post would be read as "grossly offensive" or merely as robust opinion, the rational response was self-censorship. The very existence of the section, the Court held, deterred legitimate speech—a constitutional harm that materialised even where no prosecution was ever launched. The Court distinguished Indian doctrine from the American position in one respect: it acknowledged that overbreadth challenges in India are ordinarily confined to free-speech cases, but held that within that domain the doctrine applied with full force. This reasoning links closely to the larger constitutional architecture of the Act discussed under electronic governance.

The Medium and the Article 14 Challenge

A subtler argument in the case concerned Article 14. The petitioners contended that Section 66A was discriminatory because it singled out speech on the internet for harsher treatment than identical speech in print or in person, where no comparable offence existed. The Union answered that the internet is a distinct medium—global in reach, instantaneous, anonymous and capable of viral amplification—so that messages sent over it stand on a different footing.

The Court accepted that the internet possesses features that constitute an intelligible differentia: minimal cost of dissemination, a vast and indeterminate audience, persistence of content and the ease of anonymity. On that ground it held that the Article 14 challenge, taken on its own, did not succeed, because a legislature may rationally treat a uniquely powerful medium differently. But this conclusion did not save Section 66A. Having struck the provision down under Article 19(1)(a), the Court made clear that an intelligible differentia under Article 14 cannot rescue a law that independently violates the free-speech guarantee. The medium may justify a tailored regulation; it cannot justify a vague and overbroad one.

Procedural Misuse and the Arrest Problem

Part of what made Section 66A so dangerous in operation was its procedural classification. Although the question of cognisability had been the subject of an advisory issued by the Central Government in January 2013—requiring approval of a senior police officer before arrest under the section—the Court noted that executive advisories are no substitute for constitutional validity. An advisory can be ignored, withdrawn or varied; it does not narrow the text of the offence, and a citizen's liberty cannot rest on the self-restraint of the police.

The Court treated the pattern of arrests—of students, cartoonists and ordinary commentators—as evidence of the section's inherent capacity for abuse rather than as mere instances of misapplication. This is an important doctrinal move: the Bench held that a provision must be judged by what it permits, not merely by how it has so far been used. Even a well-intentioned executive could, under the cover of Section 66A, suppress legitimate dissent, and the possibility of such suppression was itself fatal. The reasoning resonates with the broader principle, found throughout the scheme of secure electronic records and signatures and the Act's penal provisions, that powers affecting liberty must be hedged by clear and objective standards.

Section 118(d) of the Kerala Police Act

Travelling along with the challenge to Section 66A was a challenge to Section 118(d) of the Kerala Police Act, 2011, which penalised any person who caused annoyance to another "in an indecent manner" by statements or verbal or comment or telephone calls or by any means of communication. The Court found this provision to suffer from the identical vice of vagueness as Section 66A—the criminal threshold turned on the subjective notion of "annoyance in an indecent manner" with no ascertainable standard.

Accordingly, the Court struck down Section 118(d) of the Kerala Police Act insofar as it related to such communication, holding it equally violative of Article 19(1)(a) and unsaved by Article 19(2). The point is examination-relevant because it demonstrates that the Shreya Singhal ratio is not confined to the IT Act: it is a transferable constitutional standard against vague speech-criminalising statutes wherever they appear in Indian law.

Section 69A and the Blocking Rules Upheld

The judgment is often remembered solely for what it struck down, but its careful upholding of Section 69A is equally important. Section 69A empowers the Central Government to issue directions for blocking public access to information through any computer resource, on grounds congruent with Article 19(2)—sovereignty, integrity, defence, security of the State, friendly relations with foreign States, public order, or to prevent incitement to a cognisable offence.

The Court upheld both Section 69A and the Information Technology (Procedure and Safeguards for Blocking of Access of Information by Public) Rules, 2009. It reasoned that the blocking power was hedged by adequate safeguards: it could be exercised only on the enumerated grounds relatable to Article 19(2), reasons had to be recorded in writing so that the order could be tested in judicial review, and the rules provided a committee-based procedure with an opportunity of hearing. The contrast with Section 66A was stark: where 66A was untethered from Article 19(2), Section 69A was carefully moored to it, and where 66A was vague, 69A was procedurally disciplined.

Section 79 and Intermediary Liability Read Down

The third limb of the decision concerned intermediary liability. Section 79 grants intermediaries a safe harbour from liability for third-party content, subject to conditions. Section 79(3)(b) required an intermediary, upon receiving "actual knowledge" of unlawful material, to expeditiously remove or disable access to it, failing which the safe harbour was lost.

The Court read down Section 79(3)(b), holding that "actual knowledge" must mean knowledge received through a court order or a notification by an appropriate government agency that the material is unlawful, and that the unlawfulness must relate to the grounds in Article 19(2). It declined to leave intermediaries to adjudicate, on millions of private complaints, whether content was lawful—a regime that would have produced over-removal and a private chilling effect. Rule 3(4) of the Intermediary Guidelines Rules, 2011 was read down correspondingly. This reading was later superseded in part by the Intermediary Guidelines and Digital Media Ethics Code Rules, 2021, but the constitutional principle of Shreya Singhal continues to discipline how those rules are interpreted.

The Zombie Afterlife: PUCL's Follow-Up

A striking-down does not, in practice, erase a section from the police mind. For years after 2015, police across several States continued to register first information reports under Section 66A, often because the bare text remained printed in commercial copies of the Act with no annotation of its invalidity. Researchers documented well over a thousand fresh cases registered under the dead provision, a phenomenon commentators dubbed the "zombie" section.

In 2019, in People's Union for Civil Liberties v. Union of India, the Supreme Court directed that copies of the Shreya Singhal judgment be circulated to all High Courts and that the invalidity of Section 66A be brought to the notice of the police and trial courts. When fresh data in 2021 showed continued prosecutions, the Court expressed astonishment and issued further notice to the Centre. By October 2022 it had issued a set of directions requiring that Section 66A be annotated as struck down in all published versions of the Act and that pending prosecutions under it be withdrawn. The episode is a cautionary illustration that constitutional invalidation requires administrative follow-through to become real.

Doctrinal Legacy

Shreya Singhal is now the leading Indian authority on the constitutionality of speech-restricting statutes. Its tripartite test of discussion, advocacy and incitement has been applied in later free-speech litigation, and its insistence on a proximate nexus with Article 19(2) grounds has become the standard against which new penal speech provisions are measured. The judgment also marks the firm reception into Indian law of the doctrines of vagueness, overbreadth and chilling effect, previously treated with caution.

For the judiciary aspirant, the case repays close study not merely as an IT Act decision but as a constitutional law landmark sitting alongside Romesh Thappar and Ram Manohar Lohia in the Article 19 canon. It illustrates the technique of reading down (Section 79), of upholding with reasons (Section 69A), and of outright striking down (Section 66A and Section 118(d) of the Kerala Police Act) within a single judgment—a rare and instructive demonstration of the full toolkit of constitutional adjudication. Read together with the chapters on the introduction to the Act and on electronic governance, it situates the IT Act firmly within the framework of fundamental rights rather than treating it as a purely technical statute.