The Digital Public Square meets the Digital Baton

We are living through a time of great flux on the aspect of legal regulation of speech. The rules that societies had developed while living in a pre-digital age of newspapers and soap-box orators appear ill-suited to deal with speech and expression in the metaverse. It has prompted states to hyperactively innovate with fresh strategies to regulate speech and expression in the public sphere, and this feeling of standing atop shifting sands has become increasingly acute over the past decade.

India, of course, is no exception to this. The digital transformation of India has been a key plank of government policy at the national level in the new millennium; perhaps most ambitiously seen with the contentious roll-out of a national identity scheme, the ‘Aadhaar’. The conversations around this identity scheme culminated in the legal recognition of a fundamental right to privacy by India’s Supreme Court in 2017. At the same time, the past decade witnessed two terms of the Bharatiya Janata Party led alliance. This pre-existing impetus on making inroads in the digital sphere has met with the trend of state’s innovating their regulation of speech, culminating in what an author dubbed the building of an “Orwellian Framework” by 2023.

This retrospective cannot undertake a comprehensive review of all the developments in free speech law in India over the past decade. What I propose to do, instead, is focus on developments where free speech intersected with criminal law. Regulation of what the state identifies as problematic speech through administrative penalties or pure criminal law continues to underline most litigation on aspects of free speech law across time, which makes looking at this slice of the developments in law both exciting and broadly representative of the legal trends.

Operating within this framework, this post looks at developments across the three arms of the State—beginning with parliament, I turn to the executive, and finally to the courts—in how they have dealt with “hard cases” in the past decade and developed the law on free speech and expression in the process.

Parliament

Article 19(1)(a) of the Constitution of India guarantees free speech and expression, but at the same time, under Article 19(2), elaborates the areas of activity in which the government can interfere with this guarantee—allowing laws that place reasonable restrictions, in the interests of maintaining public order, the sovereignty and integrity of the country, friendly relations with foreign states, decency or morality, defamation and contempt of court.

The Indian Parliament has made several important contributions over the past decade in the domain of free speech, with the government ending its second term by passing three new criminal laws to replace the pre-independence criminal codes of India (though not yet brought into force). The new general law of crime (the Bharatiya Nyaya Sanhita) does not make any big changes to the earlier position in respect of criminalizing speech by, say, redefining obscenity or creating a new hate speech law as some had sought.

I would argue, though, that far more significant developments on the legislative front have been made outside of Parliament, when the executive utilizes existing laws both in terms of enforcing pre-existing legal regimes and creating new ones. The former I will discuss in the next section, so let us focus on the latter here. By the executive creating new regimes, I refer to the delegation of legislative functions to the executive, a common legal practice in most constitutional systems. This routine feature has been utilized to devastating effect to regulate speech and expression especially in the digital sphere. So, while the parent law, the Information Technology Act 2000, has not been amended as such in the last decade, the rule-making power under this law has witnessed frenetic activity.

Introducing sweeping changes through delegated legislation and not through the floor of the House reduces the legitimacy of the law-making exercise by preventing discussion and debate, besides promoting a sense of uncertainty through all-too-frequent changes being introduced. These issues are well demonstrated by the Information Technology (Intermediaries Guidelines and Digital Media Ethics Code) Rules 2021. The rules replaced a 2011 regime with sweeping changes and sought to bring the digital publication of news under some regulation as well for the first time—notice how such a fundamental shift was not carried out through statute but by delegated legislation. A key change in the 2021 regime has been to widen compliance requirements for intermediaries to not be held liable when it comes to problematic speech, which incentivizes take-downs of any potentially problematic speech because the risk of losing protective cover is too great for business. Doing so through delegated legislation meant that Parliament was deprived of opportunity to debate whether such heightened regulation was merited or not. It also meant reduced judicial review as there is a presumption drawn from administrative law that courts are not experts in designing and enforcing rules, which are the domain of the executive.

The 2021 rules have witnessed successive amendments in the past three years, but it is arguable that the full import of these changes has not yet been felt because of legal challenges filed in 2021 itself, which remain pending and forced the government to temporarily halt the roll-out of the new regime. Thus, for instance, the most recent iteration of changes to the rules introduced a ‘Fact Check Unit’ in a bid to stymie disinformation online. The move sought to compel intermediaries to take down any information about the “business” of the central government labelled misleading by the government itself, where failure to do so would potentially strip the intermediary of its legal protection. Copycat legislation appears to be on its way at the state level as well. Again, this feature has not yet been implemented due to a legal challenge that remains pending as of today.

The Executive

India’s rankings on almost all global indices measuring the protection of free speech have sharply fallen over the past decade. This, in large part, is due to how existing laws, and new ones, have been enforced by the executive. The synergy of old and new is best expressed in how all governments across India have embraced the legal strategy of shutting down the internet to deal with actual or threatened public disorder, drawing their powers to do so from an 1885 statute which was repealed only in December 2023. It has reached a stage where India has been billed as the internet shutdown capital of the world.

Journalists writing stories critical to the establishment, at both state and national levels, have been increasingly targeted for simply doing their job by using broadly worded anti-terror and anti-money laundering statutes, besides other crimes. Students demonstrating and voicing critical opinions have been prohibited from taking to the streets, often prosecuted for doing so. Such steps invite serious scrutiny and debate, not to mention necessarily involving a branch of the state (the courts) to review any acts of the executive. Perhaps this is why the prosecution model for chilling free speech has slowly given way to the widespread use of content takedown powers by the executive, which are not subject to strict judicial review.

The statutory scheme on ordering takedowns essentially empowers the government to order an intermediary to takedown any content, without necessarily giving prior notice to the user, and most certainly without ever making any of these proceedings publicly available. It is not difficult to see why it would appear an attractive tool to deal with problematic speech—in response to a parliamentary question in the Rajya Sabha it was disclosed that between 2018 and October 2023, over 36,000 URLs were taken down under the IT Act, 2000.

The Courts

Which brings me to the last, and most substantial section of this post, evaluating the role of the courts. It is not an overstatement to suggest that the judiciary has been equally responsible for the slow but steady deterioration of free speech protections over the past decade by its repeated failures to respond to issues with the necessary promptitude. The fait accompli jurisprudence of evasion, as Gautam Bhatia has called it, is closely followed by the actual jurisprudence itself proving to be a mixed bag for securing free speech in the face of arbitrary executive power.

Countless examples exist to support the first claim. Besides the challenges to the Information Technology 2021 Rules mentioned above, consider the validity of Facial Recognition Technology being used by police. A challenge filed before the Telangana High Court has been pending for over two years, and the challenge filed before the Madras High Court has been pending for six months. Besides big-ticket issue-based litigation, there are many smaller cases in the system where litigants have challenged their individual grievances, which requires courts to interpret how the law should be read, but courts simply are not able to decide the petitions in a timely fashion. Thus, petitions were filed before the Supreme Court challenging orders to take down the BBC series “India: The Modi Question”, where important questions of interpreting the takedown regime were raised, but the Court has yet to decide the same one year on from the takedowns. By virtue of no timely court orders, the government can continue to enforce the legal provisions with alacrity, and by the time any meaningful judicial remedy is secured, the damage to free speech is usually done.

In respect of the second claim, there are many examples, but I’ll focus on three—two from the Indian Supreme Court and one from the High Court of Karnataka, in that order. In 2016, India’s Supreme Court upheld the validity of criminal defamation, concluding that free speech guarantees could not trump a constitutionally protected right to reputation, and a defamation offence balanced these interests appropriately. The Court’s reasoning for retaining a crime of defamation with the possibility of imprisonment as against purely civil remedies for what is a private injury was notable in its refusal to engage with the proportionality doctrine—why imprisonment?—as well as its implicit faith in state mechanisms. The possible chilling effect on speech did not trouble the Court, as it felt a need to retain the crime because it viewed civil remedies proving insufficient; without any data to back that claim.

This outcome surprised some as it was preceded by a hugely significant decision of the Court in 2015, which struck down Section 66-A of the Information Technology Act, which punished ‘offensive’ online speech. In Shreya Singhal v. Union of India, a public interest litigation, the Indian Supreme Court struck down Section 66A as being unconstitutionally vague in its proscription of such speech, finding that the phrase ‘offensive’ was a catch-all one which would subsume innocent speech within its folds. The judgment called for a restrictive reading of the limitations upon speech in-built within the Constitution. Speech that fell short of incitement to public disorder would, for the Court, be speech worthy of constitutional protection.

And yet, there was a catch. While Shreya Singhal struck down Section 66-A, at the same time, it upheld the validity of statutory provisions and delegated legislation enabling the takedown of online content, concluding that the legal regime offered sufficient opportunities for aggrieved persons to review takedown orders. And it is this takedown regime which was then used to devastating effect by governments to censor swathes of critical speech online without publishing orders or data. In a remarkable turn of events, Twitter (now X Corp) approached the Karnataka High Court in July 2022 to challenge how the national government had been exercising its powers to block content on the platform.

If Shreya Singhal upheld the online takedown regime as an abstraction on an assumption that it held enough opportunities for legal review, the Karnataka High Court went ahead and upheld the regime after being shown that the opportunities for review were chimerical in nature. While Twitter has challenged this 2022 decision, the reasons offered by the High Court while dismissing its petition (with costs) warrant discussion as they show how courts have applied the constitution’s rights-restrictions matrix. The court noted that the constitution permitted restricting speech where it was a reasonable restriction for the grounds mentioned earlier in this post, and concluded that in context of online speech anything short of a blanket power to the executive to takedown content, was simply too risky considering how inflammable online speech could be. And, after all, a court could not second-guess the executive’s call on whether the speech was inflammable or a threat to national security in the first place. So, the takedown of posts, accounts, and even hashtags was legal. In terms of procedural reasonableness for exercising this power the court again advocated for an approach appropriately deferential to the executive because of the nebulous nature of speech online—sticking to notice and objections was just too clunky where a new dummy account could be created in seconds to re-agitate the same problematic speech.

Conclusion

The value a society and its laws place on protecting free speech is arguably most keenly felt where that speech takes a critical turn. Which is why the history of this field is littered with prosecutions and penalties being levied against problematic speech, inviting courts to draw the lines between what is protected and what is not. The past ten years in India demonstrate that when faced with speech that is critical of government policy or state action, the state has become increasingly hesitant to let it remain on air. What is perhaps most alarming for the health of democracy is that, in most cases, there is often a synergy across the three arms of the State that curbing problematic speech is the best course of action to follow.

(Special thanks to Gayatri Malhotra for her help with research on some aspects of the post)