The Bombay High Court Dismisses the Ministry of Truth

In 2023, the Indian central government amended the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 [“the IT Rules 2021”] to establish a Fact Check Unit [“FCU”]. The FCU was tasked to monitor online content related to ‘any business of the Central Government’ and order the takedown of any information that it considered ‘fake or false or misleading.’¹⁾ The FCU itself was envisaged as a public body and a part of the central government. If intermediaries failed to comply with the FCU’s orders, the amendment established liability for intermediaries for the content posted on their platform by third-party users. This is a crucial departure from otherwise existing liability shields for online platforms. Typically, those liability rules tie liability only to speech – but not the intermediation of speech. Simply put, platforms like Facebook or Twitter are, as a matter of widely shared principle, not liable for the speech of their users. As it seems, the Indian central government wanted to depart from that principle for content concerning, or, practically speaking, criticizing the Indian State. As the FCU would be the last arbiter of what could be said online in India about the central government, the amendment instituted what could be called a ‘Ministry of Truth’.

Questions regarding the constitutionality of that amendment quickly reached the courts. Eventually, a two-judge bench of the Bombay High Court delivered a split verdict on the constitutionality of the amendment in Kunal Kamra v. Union of India [Opinion of Judge 1, Opinion of Judge 2] earlier this year. A third judge of the High Court has now delivered their opinion, striking down the FCU on various grounds.

I agree that the Bombay High Court was right to strike down the FCU. The amendment was governmental overreach bordering on aggrandizement. However, I argue that the Court did not go far enough. By restricting the scope of analysis to the idea of governmental determination of information as ‘fake, false or misleading’, the Court failed to appreciate the fundamental question underlying the impugned amendment and closed future possibilities of governmental power to unilaterally order the takedown of online speech by threatening the extension of liability to the intermediaries. The Court’s response, in my understanding, is an incomplete answer to the problem of the unconstitutional restrictions on online speech by the government.

Given free speech’s fundamental importance for the functionality of any democratic polity, I argue that we should look beyond the intricacies of the specific case and explore current challenges to freedom of expression in contemporary Indian constitutionalism and its political context. Based on that, the blog post seeks to offer a comprehensive defense of a broad, and anti-interventionist conception of free speech in India. Such considerations will prove especially important if the government escalates the matter to the Supreme Court or redesigns the law.

Against this backdrop, I briefly discuss the decision of the Bombay High Court in Part 1. Part 2 shows that under the design of the IT Rules 2021 and the Indian Constitution, there remains a possibility for the government to resurrect the FCU. I, therefore, argue that the only way to comprehensively protect online free speech and effectively restrict governmental interference is by approaching the question from the point of view of the procedure to be followed while ordering the takedown of speech.

The decision of the Bombay High Court

Detailed commentaries and relevant documents about the case and the three judicial opinions can be found here, here, here, here, and here. Rather than engaging extensively with the three opinions, I will briefly summarize the grounds on which the majority of the Court found the impugned provision unconstitutional and then explore its approach, particularly with respect to Article 19(1)(a) – the right to freedom of speech under the Indian Constitution.

The majority made a long series of arguments to attack the constitutionality of the impugned rule. It noted that the words ‘fake,’ ‘false,’ and ‘misleading’ are vague, and the provision suffers from overbreadth. It highlighted the chilling effects of the rule and the potential self-censorship (both direct and indirect) that the authors will indulge in. It also observed that the rule violates the principle of natural justice by vesting the central government with the final say on online speech made about their affairs and by affording no opportunity to hear the author before their speech is censored. The majority also held that the provision violates the right to equality as it creates a discriminatory regime on two grounds. First, the rule only targets online speech but not offline speech, which is absurd as the dangers of false, fake, or misleading speech should remain the same irrespective of whether it appears online or in print. Second, the rule treats speech about the business of the central government more sacrosanct than other forms of speech without any sound basis for such a distinction. The majority also held that the rule violates the freedom to carry on a business or trade as recognized under Article 19(1)(g), as it hampered the businesses of media agencies that operate online and share their work with the people through intermediary platforms. More fundamentally, the majority noted that the parent legislation – the Information Technology Act of 2000 – does not provide any power to the central government to draft rules establishing a Fact Check Unit.

The final argument that the majority recognized against the impugned rule was about freedom of speech. The Court noted that, unlike the American Constitution, the Indian Constitution specifically envisages a list of eight grounds on which the government could restrict citizens’ right to free speech. This list is exhaustive and is provided under Article 19(2). The Court found that as the fake, false, or misleading speech acts are not recognized as legitimate grounds under Article 19(2), the rule cannot be constitutionally sustained.

Beyond these arguments, there is also a long commentary in the two majority opinions about how a legal restriction in the name of curbing fake, false, or misleading speech could easily transition into a strong tool in the hands of the government to curb every form of dissent and the development of opposing public discourse. Such a restriction on free speech is nothing but censorship in the name of public interest. As Justice Patel aptly summarized:

“202. What troubles me about the impugned 2023 amendment, and for which I find no plausible defence is this: the 2023 amendment is not just too close to, but actually takes the form of, censorship of user content … The impugned amendment makes the government’s chosen [Fact Check Unit] the sole authority to decide what piece of user-content relating to the undefined and unknowable ‘business of the government’ is or is not fake, false or misleading. The lack of definition of these words: business of the government; fake; false; and misleading makes the amendment both vague and overbroad. Anything might be the business of government. Anything could be ‘fake’. ‘Misleading’ is entirely subjective. And as to ‘truth’ and ‘falsity’, throughout recorded human history there are few, if any, absolute truths. Perceptions, perspectives, possibilities, probabilities – all will to a greater or lesser extent colour what one chooses to believe or hold or chooses not to believe or hold. The assumption that there are absolute truths to even the business of government, even if we knew what that included and what it did not, is unsubstantiated.

204. By shifting responsibility for user content to the vulnerable segment, viz., the intermediary, the amendment of 2023 effectively allows the government, through its FCU, to be the final arbiter not just of what is or [not is] fake, false or misleading; but, more importantly, of the right to place an opposing point of view. …”

While Justice Patel makes references to scholars who approach free speech from the value of democracy, like Alexander Meiklejohn, he essentially adopts a purely Miltonian approach. Noting that there are certain forms of false speech that the government could legitimately prohibit, for instance, those that incite violence [par. 164], he holds that a rule vesting powers in the government to effectively prohibit every speech act that it considers fake, false, or misleading is dangerous, both for the reasons of delegitimizing individual cognitive capabilities to determine the truthfulness of the ideas/opinions for themselves and for logistical reasons regarding the rightful implementation of the law. For instance, Justice Patel notes,

“206. Who, after all, is to fact check the fact checkers? Who is to say if the view of the FCU is fake, false or misleading? …

214. Finally, I believe it is unthinkable that any one entity – be it the government or anyone else – can unilaterally ‘identified’ (meaning picked out and decided) to be fake, false or misleading. That surely cannot be the sole preserve of the government. The argument that the government is ‘best placed’ to know the ‘truth’ about its affairs is equally true of every citizen and every entity.”

Surprisingly, Justice Patel makes not a single reference to Areopagitica in his otherwise lucid, educative, and powerful opinion.

However, one aspect that perplexed me about the two majority opinions is how they failed to address an important question underlying the impugned provision, i.e., can the government assume the power to unilaterally restrict speech acts made online? Is the exercise of such a unilateral power in consonance with the right to free speech? I explore this aspect in detail in the next section, contextualizing it with the nature of the internet and the Indian constitutional and IT laws regime.

Governmental control of online speech

In drafting the impugned rule, the government repeated an already familiar mistake. Back in 2009, the Indian government had introduced Section 66A to the Information Technology Act of 2000, criminalizing online speech acts that are ‘grossly offensive,’ ‘knowingly false,’ or have ‘menacing character.’ In Shreya Singhal v. Union of India (2015), Section 66A was struck down by the Supreme Court of India for a set of similar reasons as Kunal Kamra.

However, consider a situation where the government chooses to reintroduce the impugned rule about the FCU by replacing the phrase ‘fake, false or misleading’ with the eight grounds mentioned under Article 19(2). The rule would then read as follows:

“the intermediary … shall make reasonable efforts by itself, and to cause the users of its computer resource to not host, display, upload, modify, publish, transmit, store, update or share any information in the interests of the sovereignty and integrity of India, the security of the state, friendly relations with foreign states, public order, decency or morality or in relation to contempt of court, defamation or incitement to an offence.

This rewriting would effectively be a watered-down version of the existing Section 69A of the IT Act, which empowers the central government to order the takedown of online speech acts in the interest of a few of the eight grounds of Article 19(2). When its constitutionality was challenged before the Supreme Court along with that of Section 66A in Shreya Singhal, the Court upheld the provision for two reasons. First, the Court noted that Section 69A’s scope does not extend beyond permissible restrictions under Article 19(2). Second, a series of procedural safeguards are included under Section 69A, which the central government must follow before ordering the takedown. These procedural safeguards are elaborated in greater detail under the Information Technology (Procedure and Safeguards for Blocking for Access of Information by Public) Rules, 2009, and are somewhat similar to those provided under Article 9 of the EU Digital Services Act. The Supreme Court held:

“It will be noticed that Section 69A, unlike Section 66A, is a narrowly drawn provision with several safeguards. First and foremost, blocking can only be resorted to where the Central Government is satisfied that it is necessary to do so. Secondly, such necessity is relatable only to some of the subjects set out in Article 19(2). Thirdly, reasons have to be recorded in writing in such blocking order so that they may be assailed in a writ petition under Article 226 of the Constitution.”

The question before us then is, can a law that does not include procedural safeguards but limits the governmental powers to takedown online speech acts to Article 19(2) grounds be a valid exercise of power? While one may find hints about the unconstitutionality of such a law in Shreya Singhal and the two majority opinions in Kunal Kamra, they all fail to address this question directly. Rather, Shreya Singhal makes the following observation, which leaves the readers uncertain about the true scope of the state’s power to restrict online free speech:

“Merely because certain additional safeguards such as those found in Section 95 and 96 [of the Code of Criminal Procedure] are not available does not make the Rules constitutionally infirm. We are of the view that the Rules are not constitutionally infirm in any manner.”²⁾

This intellectual exercise that we are engaged in by way of a hypothesis is not fictional. For instance, consider Rule 3(1)(d) of the IT Rules 2021. It provides that

“an intermediary … shall not host, store or publish any unlawful information, which is prohibited under any law for the time being in force in relation to the interest of the sovereignty and integrity of India; security of the State; friendly relations with foreign States; public order; decency or morality; in relation to contempt of court; defamation; incitement to an offence relating to the above, or any information which is prohibited under any law for the time being in force …” [Rule 3(1)(d)]

Compared to Section 69A, this provision adds new grounds on which the government could order the takedown of online speech. While grounds such as decency, morality, contempt of court, and defamation find a mention under Article 19(2) of the Constitution, they are not included under Section 69A. Rule 3(1)(d), therefore, creates a regime wherein the government or any of its designated agencies could unilaterally direct the takedown of online content in the absence of any Section 69A procedural safeguards. It is also significant to note that the last portion of the rule (underlined for easy reference) expands the scope of governmental powers even beyond Article 19(2), and now the government could unilaterally direct the intermediaries, without following any procedure, to take down information that it considers as prohibited under any law.

The implications of such a regime are severe. First, consider this scenario. One of the due diligence obligations for the intermediaries under the IT Rules 2021 is to ‘cause the users of its computer resources to not host, display, upload, modify, publish, transmit, store, update or share any information that, – … is … patently false or untrue or misleading in nature … or is insulting other nations’ [Rule 3(1)(b)]. In simple terms, intermediaries are prohibited under law from carrying false, untrue, or misleading information. What if the government now uses Rule 3(1)(d), particularly the part wherein the intermediaries are not to host information that is prohibited under any law, and orders the takedown of information about its business that it considers as fake, false, or misleading? Such takedown orders need not be restricted only to content about the business of the government; rather, they can extend to the entire domain of online speech.

Second, as Article 19(2) of the Constitution allows the government to restrict speech on broad grounds like ‘public order’, ‘morality’, and ‘friendly relations with foreign states’, a unilateral exercise of power to order the takedown of speech would necessarily translate into a unilateral power to set the scope, manner, and content of online discourse. For a government that cannot desist from unlawfully jailing people, gagging people’s speech adds to its toolkit as an easy way to effectively control the public discourse. This would, in essence, resurrect the FCU.

Third, such legal regimes result in the institutionalization of a sense of fear in the citizenry, where people tend to censor anything that could even remotely sound critical of the ruling dispensation in the expectation of either legal or extra-legal rebuke. Instances of people deleting their tweets and posts abound.

Moreover, the fact that none of the intermediaries approached the courts to challenge either Rule 3(1)(d) or the amendment to Rule 3(1)(b)(v) that created the FCU shows the sense of reluctance among them to fight the state. They would not want to risk their safe harbor by posing any challenge, even if principled, to the authority of the State. As a national daily reported on the grimness of the situation,

“Companies appear reluctant to challenge rule 3(1)(d) in courts. ‘There is no appetite to take the government on. Companies don’t want to be seen as challenging the government as that has very real consequences in terms of investments, contracts, approvals, etc. There is a distinct and clear fear,’ the first expert said.”

Lastly, unilateral power to order the takedown of speech acts puts a burden on the citizens to challenge such governmental actions in the absence of the knowledge of the grounds on which their content was censored. Procedural protections, such as sharing a reasoned order justifying the takedown, allowing time to respond, etc., create friction to the brazen exercise of power. It is only by centralizing the aspect of just procedures that one can seek comprehensive protection of online free speech. While I realize that procedural safeguards may not render the desired results, they do create a regime of slower governmental interference, a chance to judicially and publicly shame unreasoned/badly reasoned governmental orders, and narrow down the scope of wide Article 19(2) grounds such as ‘public order’ or ‘friendly relations with foreign states’. Otherwise, as shown above, it would not be difficult for the government to resurrect the essence of FCU.

Concluding Remarks

As Nussbaum argues, legal regimes grounded in fear pose existential dangers to our democratic societies. Judicial response to such actions of governmental aggrandizement must be comprehensive and should address the issues at the most foundational level in such a way that it effectively denudes every avenue to the government to impede our fundamental rights. Moreover, a Miltonian argument will only have a limited appeal in today’s context – the question is much bigger than that of truthfulness or falsity in online speech. The impugned rule and other provisions discussed above are attempts to make dents to the idea of a healthy public discourse where certain narratives are meant to remain beyond challenge. This scheme of two-level online content moderation, where the speech is effectively moderated first by the intermediaries and then by the government, must, therefore, be resisted.