On 26 February 2026, the Supreme Court of India, acting on its own motion (suo motu) and without any petitioner before it, issued a remarkable order. It banned a Class 8 social science textbook published by the National Council for Educational Research and Training (“NCERT”), directed the physical seizure of all copies in circulation, ordered the removal of the book from every digital platform, threatened its authors with criminal contempt proceedings, and barred classroom instruction based on its contents — all in response to a chapter that described corruption in the judiciary and case backlogs as institutional challenges. No one had been heard. No one had been charged. The order was issued, as orders do, and the machinery of the state moved to comply.

That the chapter’s substance may be worth defending is an argument I’ve made elsewhere. This post is concerned with something prior and, in some ways, more troubling: the manner in which the Court proceeded. The NCERT affair is a case study in how the Court, when it chooses to act against those who scrutinise it, abandons the very procedural principles it has spent decades developing for everyone else.

Punishment Before Process

The order of 26 February reads, in places, more like a verdict than a preliminary proceeding. The Court described the chapter’s contents as reflecting a “discernible underlying agenda to undermine the institutional authority and demean the dignity of the judiciary.” It characterised the NCERT Director’s written response defending the chapter as “reckless, irresponsible, contemptuous, and motivated.” It concluded, on a prima facie examination, that the publication fell within the definition of criminal contempt under Section 2(c) of the Contempt of Courts Act, 1971.

What is striking is that these characterisations — of motive, of agenda, of culpability — arrived before any of the individuals responsible had been heard. The chapter’s authors were not parties. They had received no notice. They had made no submissions. The Court did not hear them before concluding, in effect, that they had acted deliberately and with malice. The show-cause notice came after the conclusion, not before it — a sequencing that inverts the foundational principle that a person must be heard before adverse findings are recorded against them.

At the March 11 hearing, the Court went further. Having by then identified Michel Danino, Suparna Diwakar, and Alok Prasanna Kumar as the persons responsible for preparing the chapter, it directed that they be “disassociated” from all government-funded curriculum and educational projects across all states, union territories, and universities forthwith. The order extended to any work “involving public funds,” an approach which is quite reminiscent of the one taken by the Court previously in the case of a YouTube podcaster, as discussed here on this blog in the past as an instance of judicial overreach and paternalism.

The Court further observed that these individuals “either lack reasonable knowledge about the Indian judiciary or they deliberately and knowingly misrepresented the facts.” On this basis—a binary that presumes either incompetence or malice, with no third option available—they were effectively barred from a category of professional work, nationally, without trial, without finding, and without appeal.

There is, to be sure, a narrow escape route. The order is “subject to their approaching this Court for modification with an explanation.” It is apparent that this post-facto hearing guarantees no procedural fairness. Rather, this shows the slow establishment of an approach that is characteristically different from the suo-motu jurisdiction of the court, where the Court feels empowered to create a lis, decide upon the matter, and impose punishment, all without adequately hearing the affected parties. Per the Court’s directions, the affected individuals now bear the burden of approaching the very institution that has already characterised their work as contemptuous, to seek relief from an order that presumes their culpability. The choice offered is not between innocence and guilt. It is between submission and exclusion.

Contempt Law and Its Amplitude

The legal vehicle deployed here—criminal contempt under Section 2(c) of the Contempt of Courts Act, 1971—deserves scrutiny in its own right. Section 2(c) defines criminal contempt as the publication of any matter that scandalises or tends to scandalise, or lowers or tends to lower the authority of, any court; prejudices or interferes with the due course of any judicial proceeding; or otherwise obstructs the administration of justice. The provision is breathtakingly wide. “Tends to lower” covers not just deliberate attacks on the institution but any material that a court might, on its own assessment, find to have that tendency.

The Indian Supreme Court has, in other contexts, been among the most sophisticated courts in the world on the dangers of vague and overbroad statutory language. It struck down Section 66A of the Information Technology Act in Shreya Singhal v. Union of India precisely because vague formulations such as “grossly offensive” could be deployed to suppress legitimate expression. The reasoning was that statutory language that leaves the speaker uncertain about the boundaries of permissible expression has a chilling effect that itself constitutes a constitutional harm. The same logic applies, with equal force, to Section 2(c), but it is a logic the Court has never applied to the provision that most directly protects its own reputation.

The result is a structural asymmetry. The Court that insists on definitional clarity in legislation restricting speech about public figures and political questions declines to apply that standard to legislation restricting speech about itself. Every critique of the judiciary, every academic analysis, journalistic investigation, or pedagogical treatment that a bench finds objectionable is potentially actionable. The boundary between legitimate criticism and criminal contempt is drawn, not by a neutral arbiter, but by the institution being criticised. It is the Court that determines, in each case, whether the criticism has crossed the line. That the Court routinely insists it welcomes legitimate criticism, and that its orders are not intended to suppress it, does not resolve the problem. The uncertainty created by the threat of proceedings is itself the suppressive mechanism.

A Court Too Possessive of Its Domain

There is a broader pattern into which this case fits. The Indian Supreme Court has, over the past decade, developed what can only be described as a proprietary attitude toward any question touching the judiciary. Judicial appointments are the exclusive domain of the collegium, a position the Court has defended against legislative and executive encroachment with considerable vigour. Criticism of judicial conduct is managed through contempt jurisdiction. And now, as this case illustrates, the content of educational material about the judiciary is subject to judicial oversight as well.

In each of these domains, the Court’s position is the same: questions about the judiciary are for the judiciary to resolve. The legislature may not alter the appointment process. The executive may not comment on judicial accountability. Textbook authors may not present students with documented facts about corruption and case backlogs without risking prosecution. No principled argument can justify this position. It is, in fact, institutional self-interest dressed in constitutional language.

This is particularly incongruous for a court that has written extensively on separation of powers and on the importance of each branch operating within its constitutionally described domain. The Court’s own jurisprudence on the Basic Structure doctrine, on legislative competence, on the limits of executive action, all of it rests on the premise that no institution is above constitutional constraint. But the practice of the contempt jurisdiction, as deployed in this case, suggests that when the institution being constrained is the Court itself, the analysis changes.

The Self-Censorship That Does Not Require an Order

More than the immediate consequences faced by the three individuals named in the March 11 order, the potential chilling effect of this judicial inquiry is deeply concerning. This episode will force the recalibration, quiet and widespread, of what academics, authors, curriculum developers, and journalists understand to be safely writable about the Indian judiciary.

The Court has said, both in its orders and from the bench, that it does not intend to suppress legitimate criticism. But the institutional signal sent by this proceeding is categorical: write about judicial corruption in a textbook that reaches millions of schoolchildren, and you risk being publicly characterised as either ignorant or malicious by the Chief Justice of India, barred from government-funded work, and placed under the shadow of criminal contempt proceedings. No one needs to receive that signal twice. The next curriculum developer, the next researcher, the next journalist considering whether to document what former Chief Justices have themselves acknowledged in public—judicial corruption, case pendency, transparency deficits—will know what awaits them. They will make adjustments accordingly. None of this will require a further order.

Self-censorship of this kind is not easily remedied. It leaves no paper trail. The books that are not written, the chapters that are not included, the analyses that are softened before publication, these absences are not recorded anywhere. But they are, over time, the most significant outcome of cases like this one. A free press and a free academy are not silenced by prosecution alone. They are silenced by the demonstrated willingness of the powerful to prosecute, even occasionally, and even selectively. The demonstration, not the frequency, is what governs behaviour.

The Recourse Problem

The order of 26 February provides no mechanism for the textbook’s authors, or for NCERT, to seek review before a different forum. The High Courts are not available as contempt of the Supreme Court is a matter for the Supreme Court. There is no appeal from a suo motu order of this Court. The “modification” route offered in the March 11 direction requires the affected individuals to petition the same bench that has already recorded its view of their conduct. It is a process designed to produce apologies, not acquittals.

NCERT’s trajectory illustrates the dynamic precisely. On 26 February, the Director defended the chapter in writing, a response the Court described as reckless and contemptuous. By 10 March, NCERT had issued an “unconditional and unqualified apology” and confirmed the withdrawal of the entire book. At the March 11 hearing, the Court noted this apology and questioned whether it was genuine or merely an attempt to evade legal consequences. The institution that expressed a considered view in writing had, within two weeks, abandoned that view entirely. Not because it was persuaded. Because it had no viable alternative.

This is not how accountability is meant to work in a constitutional democracy. Accountability flows from independent review, from adversarial process, from the possibility of a finding of no wrongdoing. What this proceeding has produced, instead, is compelled contrition, an apology extracted not by argument but by the demonstrated consequences of not apologising. That the Court then scrutinised the apology for sincerity completes an almost Kafkaesque circuit: the institution compelled to apologise is then judged on the quality of its remorse by the institution that compelled the apology.

The Protector as Prosecutor

There is a foundational problem that runs beneath all of the specific procedural failures identified above. It is this: the institution whose reputation is at issue is the same institution that determines whether that reputation has been unlawfully damaged, that issues the order, that decides whether the apology is sufficient, and that will ultimately adjudicate any contempt charge. There is no external check, no independent arbiter, no forum in which the person accused of scandalising the Court can obtain a disinterested hearing. The Court is, simultaneously, the complainant, the investigator, the judge, and the sentencer.

Procedural fairness in every other domain of Indian law rests on the separation of these functions. The right to be heard before an adverse order, the right of appeal to an independent tribunal, the prohibition on a judge sitting in her own cause; these are not bureaucratic formalities. They are the structural guarantees that prevent authority from collapsing into arbitrariness. The contempt jurisdiction, as currently constituted, exempts the Court from all of them when it chooses to act in its own defence.

The Indian Supreme Court has, as a historical matter, been an institution of genuine consequence. One that has protected fundamental rights against an overbearing executive, held the democratic process to constitutional standards, and provided redress to those the other branches ignored. That record is real and should not be dismissed. But it is also not a licence. An institution that protects others from the exercise of unchecked authority cannot exempt itself from the same constraint. The guarantee of one’s rights cannot be made contingent on the mood of the institution one wishes to criticise. That is not constitutionalism. It is patronage.

When the protector becomes the prosecutor, the citizen is left with a choice the Constitution was designed to eliminate: speak freely and risk the wrath of the institution, or stay silent and keep your career. In a democracy, that should not be a choice anyone has to make.