Shreya Singhal v Union of India: Impact on Cyber Law

By Advocate Panchanand Shaw, Bar Council of West Bengal · Published 4 Nov 2026 · Cyber law

Shreya Singhal v Union of India is the decision that took the most feared provision out of India's cyber law. On 24 March 2015, in Writ Petition (Criminal) No. 167 of 2012 and connected matters, the Supreme Court struck down section 66A of the Information Technology Act, 2000, holding that it violated the freedom of speech and expression conferred by Article 19(1)(a) of the Constitution and was not saved by any of the grounds listed in Article 19(2). The case turned on the wording of a single short section, and its reasoning governs every online-speech prosecution filed since.

Understanding the ruling is practical, not academic. Complainants still ask for section 66A to be applied to an abusive message, police officers still sometimes cite it from habit, and accused persons need to know what replaced it. This guide explains what section 66A said, why it fell, what survived, and how the conduct it once covered is dealt with today.

What section 66A actually said

Section 66A punished the sending, by means of a computer resource or a communication device, of information that was grossly offensive, or that the sender knew to be false and sent for the purpose of causing annoyance, inconvenience, danger, obstruction, insult, injury, criminal intimidation, enmity, hatred or ill will. Its most striking feature was what it did not require. It did not require the message to be obscene, defamatory, a threat, or an incitement to an offence. Annoyance and inconvenience, by themselves, were enough, and the section carried a punishment that could be used to arrest a person over a single post.

The provision sat inside a statute meant to regulate computer misuse. The Statement of Objects and Reasons of the amendment that introduced it referred to new crimes such as publishing sexually explicit material, voyeurism, e-commerce fraud and identity theft. Section 66A was the one clause in that group that dealt with expression rather than conduct, and it was drafted in open-ended language that left the police, rather than the statute, to decide what was punishable.

The petitions were filed after a series of arrests over social-media posts in 2012, which brought the section into national controversy. The challenge was not confined to section 66A. The petitioners also questioned section 69A and the blocking rules, section 79 and the intermediary guidelines, and comparable state provisions, of which section 118(d) of the Kerala Police Act, 1960 was the example before the Court. The judgment therefore settled the shape of online-speech regulation in several respects at once, not merely the fate of one clause.

Why the Supreme Court struck it down

The three-tier test

The Court began by distinguishing three levels of expression: discussion, advocacy and incitement. Discussion and advocacy of a point of view, however unpopular, are protected speech; only incitement, where a person urges others to commit an offence, falls outside the protection and may be restricted under Article 19(2). Section 66A, the Court found, had nothing to do with incitement. A message that was merely annoying or inconvenient did not ask anyone to commit a crime, so the section could not be defended as a restriction grounded in public order.

Vagueness

Terms such as "grossly offensive", "annoying" and "inconvenient" were undefined and open-ended. A law that makes a citizen's liability depend on an officer's sense of what is offensive gives no guidance in advance about what is permitted. Vagueness of that kind is fatal, because citizens cannot regulate their conduct and the police are left with arbitrary discretion.

Overbreadth and chilling effect

Because the section swept in speech with no connection to any Article 19(2) ground, it was overbroad, and its reach produced a chilling effect: a person who fears arrest over a mildly offending post may say nothing at all. The Court also rejected the argument that an assurance of reasonable administration could save a badly drafted law. A clause capable of being used outside Article 19(2) cannot be rescued by promises about how it will be used.

An argument based on Article 14 - that speech on the internet was treated differently from other media - failed. The Court held that there is an intelligible difference between internet speech and print or broadcast, and that a legislature may create separate offences for the internet. The section fell under Article 19(1)(a), not Article 14. In the same judgment the Court struck down section 118(d) of the Kerala Police Act, 1960, for identical vagueness.

What the Court upheld

The judgment was not a wholesale rejection of the IT Act. Section 69A, which empowers the Central Government to block public access to information in the interests of the sovereignty and integrity of India, defence, security of the State, friendly relations with foreign States, public order, or preventing incitement to a cognizable offence, was upheld. The Court found it narrowly drawn, requiring a reasoned order in writing, with procedural safeguards under the Information Technology (Procedure and Safeguards for Blocking for Access of Information by Public) Rules, 2009.

Section 79, the intermediary exemption, was also upheld, but sub-section (3)(b) was read down. An intermediary loses the exemption only where it fails to remove or disable access to material after it has actual knowledge from a court order, or after notification by the appropriate Government, and only where that knowledge relates to conduct within Article 19(2). The Court reasoned that intermediaries cannot be asked to judge millions of complaints themselves. The Intermediary Guidelines Rules, 2011 were read down on the same lines.

What replaced section 66A?

Nothing replaced it, and that is the point. The conduct people imagine section 66A covered is now dealt with by other provisions, depending on what was actually done.

The consequences of that shift matter. Most BNS provisions covering insult and defamation are far narrower than section 66A ever was, several are non-cognizable or compoundable, and a complainant must be able to point to a specific ingredient - a threat, a false imputation of fact, an incitement - rather than merely to hurt feelings. That is precisely the balance the Supreme Court intended.

What it means for a cyber crime complainant

If a message has abused or upset you but done nothing more, the honest answer is that the criminal law may offer no remedy, because the Court deliberately removed a provision that turned annoyance into an offence. If the message contains a threat, a false imputation of fact intended to harm reputation, an incitement, sexual content, or a demand for money, then there is a real cause of action, and the complaint should be built around the ingredient that is present. Where content is defamatory, criminal defamation under section 356 of the BNS is available, and a civil suit for defamation remains a separate option. Where material is obscene or sexually explicit, the IT Act sections 67, 67A and 67B apply, and for content relating to a child the POCSO Act, 2012 adds its own offences and reporting duties.

What it means for an accused person

The judgment is a defence in itself. Any complaint or charge that cites section 66A is bad from the start, because the section no longer exists. Even where a substitute provision is cited, the defence should test whether its ingredients are made out: criminal intimidation needs a threat of injury to person, reputation or property; defamation needs an imputation of fact, made or published, intended or likely to harm reputation; public mischief needs something more than an unpopular opinion. An accused facing such a charge should also remember that several of these offences are bailable and non-cognizable, which affects arrest and the forum. Where an arrest is threatened over a social-media post, bail and anticipatory bail before the appropriate court are the immediate remedies, and a bail application should be prepared on the footing that the underlying provision is weak.

Where the law is still unsettled

The ruling left genuine grey areas. It did not settle how far a platform may be asked to remove content, and the IT Rules have been amended since, most notably the 2021 rules that add due-diligence and grievance machinery. It did not resolve every tension between a defamation charge and the constitutional protection of speech. And it did not, and could not, prevent the filing of FIRs under other sections over online speech; whether such a case survives depends on how narrowly the police and courts read the substituted provision. On these points the position is genuinely developing, and any assessment should be made on the current text of the rules rather than on a recollection of the 2015 judgment alone.

How these complaints are drafted today

Because the old catch-all is gone, the way an online-speech complaint is framed has changed. A workable complaint now does four things.

1. Identifies the exact words, image or recording complained of, quoted verbatim, and the platform on which it appeared.
2. States the specific ingredient that is present - a threat, a false imputation of fact, an incitement, or obscene content - and matches it to the section relied on.
3. Attaches a properly certified copy of the electronic record, so that the post can be proved under section 63 of the Bharatiya Sakshya Adhiniyam, 2023.
4. Explains why an offence, and not merely the sender's opinion, is made out, because a vague complaint is the fastest route to a closure report.

Common mistakes

• Asking the police to register a case under section 66A. It does not exist, and a complaint built on it will fail.
• Assuming that because section 66A is gone, nothing can be done about abusive online messages. Threats, defamation, obscenity and cheating remain punishable.
• Confusing the removal of content with prosecution. Blocking under section 69A and an FIR are different processes with different tests.
• Treating the judgment as one that protects incitement. It does not; the Court drew the line at incitement deliberately.
• Ignoring the certification of the electronic record of the offending post, which must satisfy section 63 of the Bharatiya Sakshya Adhiniyam, 2023 when produced as secondary evidence.

When to involve a lawyer

A complainant who has received a threatening or defamatory message, and an accused booked over a social-media post, both benefit from advice on what the facts actually support. The work is to identify the specific ingredient, to choose between a criminal complaint, a civil defamation action and a platform grievance, to handle the electronic-evidence certification that decides whether the post can be proved at all, and to act quickly on bail where an arrest is threatened. Our cyber crime lawyers handle both sides of online-speech matters, and the wider defence position is set out under criminal defence.

The judgment did not give the internet immunity from the criminal law. It insisted that any restriction on speech be precise enough for a citizen to know, before speaking, what is prohibited.

Frequently asked questions

Information for general guidance only and not legal advice. Provisions cited follow the Information Technology Act 2000, the Bharatiya Nyaya Sanhita 2023 and the Bharatiya Nagarik Suraksha Sanhita 2023.

Related articles

• Subramanian Swamy v Union of India: cyber defamation
• K.S. Puttaswamy v Union of India: privacy and cyber crime
• State of Haryana v Bhajan Lal: quashing a cyber crime FIR
• Ravinder Singh v State of Punjab: when is it cheating?
• Jacob Mathew v State of Punjab: negligence and liability
• Arjun Panditrao Khotkar v Kailash Kushanrao Gorantyal

Cyber crime lawyer in your city · Bank account freeze help · Contact