In a 148 page judgement that categorically struck down the controversial amendments to the Rules, Justice Gautam S Patel of the Bombay High Court recognized the chilling effect of censorship that could prevail when the business intermediary is held accountable for anti-Central government content. The amendments amount to nothing but censorship of user content, states the judgement. (Para 202)
On January 31, after a wait for almost four months, Justice Patel’s was one among a split judgment delivered in the petitions filed in the Bombay High Court against the 2023 amendment to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules. The amendment, which provide for setting up fact check units empowered to identify and tag what it considers false or fake online news for any activity of the Central government, had been challenged before the High Court by comedian Kunal Kamra and several others. Justice Neena Gokhale disagreed.
As the Patel judgement narrates at the outset, several petitioners (that included the Editors Guild of India and the National Broadcasters & Digital Association) apart from Kamra himself, different voices from different quarters, are united in their complaint about the Central Government’s 2023 amendment to a particular Rule under the Information Technology Act. This amendment, they all say, has just one purpose: to muzzle criticism and questioning of the Central Government as it goes about its ‘business’, to stifle dissent from what the Central Government puts out in regard to affairs of the state.
“Through this amendment, the Central Government has anointed itself as the sole arbiter of what is or what is not ‘fake, false or misleading’. It, and it alone, will decide this, including the entirely ambiguous term ‘misleading’; and when it does, dare any ‘intermediary’ allow such information to continue to be hosted on a social media platform, that intermediary immediately risks losing statutory protection. The entire amendment is overbroad, vague and without controlling guidelines. It does not even provide for an opportunity for the propounder of any information to defend its correctness, and the government becomes a judge in its own cause (hence another dimension of violation of principles of natural justice).” (Para 2)
The findings are conclusions are delivered over nine pages. (Paras 202 – 2015).
Considering the effect of the impugned amendment, Justice Patel states that “What it does is to shift the responsibility for content accuracy from the creator or originator of the content to the service provider or intermediary, an entity that axiomatically has no control over the content at all. But an intermediary is the one that has safe harbour and, as a business entity, is regulated by the government in myriad ways. The impugned amendment makes the government’s chosen FCU 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 said to be ‘fake’. ‘Misleading’ is entirely subjective. And as to ‘truth’ and ‘falsity’, throughout recorded human history there are few, if any, absolute truths. “(Para 202).
“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.” (Para 202)
Most crucially, the judgement examines how exactly the envisaged ‘Fact Checking Unit’ under the ambit of the Central (Union) government will function.
“How the FCU will go about its business is also unknown. We are simply asked to trust it. This is not a question of trust, and especially not of distrust in any particular dispensation. It is simply a matter of setting the impugned Rule against the settled law and seeing whether it passes established Constitutional tests.(Para 203)
“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 is fake, false or misleading; but, more importantly, of the right to place an opposing point of view. We have already seen examples, and they are not entirely hypothetical. The government routinely rebuts criticism. If, in addition, this is now dubbed fake, false or misleading (and there are no guidelines to suggest why it cannot), then criticism and debate are stifled. There is little achieved in saying that the guidelines will come later. There is no assurance of that either; and they should have been in place by now if there was such an intent.” (Para 204)
While recognizing the problem of deep fakes and similar challenges, the judgement observes that these are the very challenges that the nature of a
“democratic republic with ensured freedoms: the cacophony of dissent and disagreement is the symphony of a democracy at work. The view of the FCU or the PIB that a particular piece of information relating to the business of the government is fake or false or misleading should not be allowed to be taken inviolate; most certainly, its publication cannot be subject to penalty and loss of immunity.” (Para 205)
Who, after all checks the government asks the judgement,
“Who, after all, is to fact check the fact checker? Who is to say if the view of the FCU is fake, false or misleading? Quis custodiet ipsos custodes?” (Para 206)
The judgement goes on to elaborate how, the amendments to the rules under challenge seek to insidiously expand the restrictive aspects of Article 19 (2). Besides, they also fail to meet the test laid down by the Supreme Court in Shreya Singhal, 2007.
“Of equal concern is the attempt in the impugned amendment to impermissibly expand the remit of Article 19(2). This, as I have noted, is directly contrary to the government’s own stand before the Supreme Court in Kaushal Kishore, where it argued that Article 19(2) is exhaustive. I fail to see how the amendment is within the scope of Shreya Singhal either. To the contrary: it fails every test set out in that decision, especially for overbreadth and vagueness. The impugned amendment is ultra vires Article 19(1)(a), Article 19(2), Article 19(1)(g), Article 19(6), Article 14, violates the principles of natural justice and is also ultra vires Section 79 of the IT Act. (Para 207)
Clearly rooting for the protection of all fundamental rights, he holds,
“As a general and perhaps even inflexible rule, I would suggest that every attempt to whittle down a fundamental right must be resisted root and branch. The slightest possibility of a fundamental right abridgment cannot be allowed to stand. Every attempt to limit any fundamental right must be demonstrably confined to its permissible limits within Articles 19(2) to 19(6). Everything else is illegitimate. For between the ‘abyss of unrestrained power’ and the ‘heaven of freedom’ lie these three Articles of our Constitution: Articles 14, 19 and 21. These are the famous words of YV Chandrachud CJI in Minerva Mills v Union of India & Ors. (Para 208)
“The submission on Article 14 and the invalid classification is, in my view, correctly placed. As I have noted, information relating to the business of the Central Government is a subset fully included in the 2022 amendment. There is no particular reason why information relating to the business of the Central Government should receive ‘high value’ speech recognition, more deserving of protection with a dedicated cell to identify that which is fake, false or misleading, as opposed to precisely such information about any individual or news agency. There is material, indicated above, about what is called the epistemic apocalypse. We have seen that other than point to instance of ‘fakery’, there is no material at all of any particular ‘public interest’ or ‘national interest’ peril — and these are not even within the permissible parameters of Article 19(2). Consequently, it follows that separating out the business of the Central Government for preferential treatment is class legislation, not a rational or permissible classification. I will accept that it is not open to intermediaries to disclaim all responsibility: no Petitioner has suggested that. Wholesale abandonment of all responsibility is irresponsible. But that only reinforces the point that all information deserves equal treatment. (Para 209)
The issue of governmental overreach and vagueness raised by the Petitioners have been upheld.
“The questions of over breadth and vagueness are indeed troubling. I do not believe it is any answer to suggest that, though we do not have it now, in the fullness of time the FCU will evolve some sort of working protocol, guidelines or yardsticks. These are being, in that self-applied to handpicked content and without any indication of the processes to be followed it is difficult to accept that the impugned Rules are sufficiently narrowly tailored. These issues are discussed in Shreya Singhal, and those findings apply with equal force here. (Para 210)
“Further, the lack of contemporaneous guidelines raise the other question, never fully answered. As I have noted, the entire argument of the Union has more or less proceeded on the basis that all users are individuals. But, as we have seen immediately, that is entirely incorrect. Users are also entities such as news outlets and journals. Not only do they have their own fact-checking systems, but they and their individual writers publish in print and online. The decisive test must surely be that if the material in print cannot be subjected to FCU checking and compelled deletion, there is no reason why, merely because the exact same material also appears online it is susceptible to unilateral determination of fakeness, falsity or being misleading.” (Para 211)
“I have not been able to accept the submission, perhaps implicitly suggested, that there has been no violation — in the form of censorship or unilateral takedown as yet — and therefore there is no call for interference; or that the assault is on ‘mere possibility of abuse’. It is not my reading of the law that petitions can only lie infringement by infringement. The entire discussion in Shreya Singhal on ‘the chilling effect’ militates against an acceptance of any such submission, for the finding of the Supreme Court is clearly directed towards the anticipated future impact of a rule. The very words ‘chilling effect’ suggest only this.” (Para 212)
The judgement also rejects the incorrect contention that the government is somehow sole repository of determination that citizens get “correct information.”
“More importantly, this argument is founded on the entirely incorrect theory that the government is somehow parens patriae; it is duty bound to ensure that citizens receive only ‘correct information’ (or what the government considers correct information); that the reasonable reader is infantile and cannot decide for herself or himself; and so on to the end of the chapter. This is again circular, for it is posited on the assumption that government-related information is somehow special and deserving of extra protection. This sits at odds with the fact that the biggest megaphone and the loudest voice is that of the government: if there is one entity that does not need such protection, it is the government. It already has an ‘authentic’ voice; possibly, the most authentic voice. And it has so far been unafraid to use it.” (Para 213)
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. Paradoxically, complaints of a grievous nature (pornography, child abuse, intellectual property violations) can only be taken down only after following a grievance redressed procedure; yet anything relating to the business of the Central Government can be ‘identified’ as fake, false or misleading by the FCU — and cannot be hosted. (Para 214)
Having made these cogent arguments in the conclusions and at length before, finally, in Para 2015, Justice GS Patel strikes down the “2023 amendment to Rule 3(1)(b)(v) of the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021” (Para 2015). The solicitor general of India (SGI) has assured the Court that no further steps of implementation would take place before the matter is finally heard, afresh, by a new bench.
The entire judgement may be read here.
Related:
Serious flaws in the Digital Personal Data Protection Act
Registration of digital fact-checking sites with govt may be a part of the future Digital India Bill
India’s 2023 bad laws: Impact on Individual Freedoms and Indigenous Rights in a weaponised state
Broadcasting Bill adverse to freedom of speech & freedom of press: EGI