Defending an unpopular ban -- and some other unfashionable views

(Brief Note: It is rather late, and I have an early internship tomorrow. Consequently, I have omitted hyperlinks. I hope to add them later, although I know myself well enough to be sceptical. Most sources should be easily locatable. My apologies; I could probably have added them in the time taken to write this.)

This post examines the Delhi High Court's decision to uphold the ban on Telegram in the context of the NEET-UG examination leak controversy. It argues that while the ban undoubtedly implicates free speech and access to information, that fact alone is not a sufficient reason to strike it down. The right to freedom in India has never been absolute, and the reasonable restrictions clause exists precisely because competing state interests demand accommodation. Some balance must be struck between the harms caused and the harms prevented. The State, therefore, ought to be afforded some margin of appreciation in determining where that balance lies. 

As readers will have gathered by now, the remainder of this post is a defence of the ban and the judgement. Before getting there, however, it may be useful to begin with a quick rundown of what the Court held.

The Judgement in Brief

The Court in Telegram was dealing with two principal questions: (1) whether the blocking order suffered from non-application of mind; and (2) whether a temporary platform-wide ban on Telegram could satisfy the requirements of proportionality.

1. On the first, the Court rejected Telegram's objections. It held that the impugned order adequately disclosed the Government's satisfaction that Telegram was being used for the dissemination of examination-related misinformation with potential implications for public order (Paras 22, 24). Given the emergency nature of the proceedings, the reasons contained in the order were considered sufficient, more particularly when read alongside the subsequent post-decisional hearing and final order confirming the ban (Paras 25-28).

2. The second issue concerned proportionality. Telegram argued that the Government could have continued targeting specific channels, bots, and accounts rather than disabling access to the platform as a whole (Paras 38, 41). The Court, however, accepted the Government's account and noted the platform's large public channels, bot ecosystem, audience migration mechanisms, mirror channels, and message editing functionality – all of which allegedly allowed unlawful activity to rapidly reappear despite repeated takedowns (Paras 42-45). In these circumstances, narrower interventions were found to be ineffective (Para 45).

Hearing First, Acting Later(!)?

Of the various objections raised against the ban, the procedural one is probably the strongest. There is something extremely troubling about the State waking up one morning and disabling access to a platform used by millions of people before those affected have had a full opportunity to contest the decision. The basic idea that people should ordinarily be heard before adverse action is taken against them is deeply embedded in public law. If one encountered the facts of this case in the abstract — a platform blocked first and heard later — one would be forgiven for being sceptical.

The problem, however, with much of the criticism directed at the judgement is that it asks the wrong question. Prior hearings are almost always preferable to post-decisional hearings. But the relevant question is whether a prior hearing was realistically compatible with the objective that the Government was attempting to achieve. To put it differently, what exactly was the State supposed to do?

Rule 9 is an emergency framework and its entire justification lies in the possibility that waiting for the ordinary process might itself frustrate the purpose of intervention. One may disagree with the Government's assessment of the facts. One may even think the threat was overstated. But if one accepts – even for a moment – that the factual premise upon which the order proceeded — i.e. the examination fraud networks were actively operating in the immediate run up to a national examination involving more than twenty lakh candidates — then the demand for a complete hearing before any action whatsoever begins to look somewhat unrealistic – and frankly – a bit stupid. 

The easiest way to see this is through a simple counterfactual. Assume that the Government had done precisely what critics now suggest it should have done. Assume it had issued notice, invited responses, scheduled hearings, considered written submissions, and completed the ordinary process before acting. Suppose, during that period, the alleged fraud ecosystem continued operating and the examination was conducted amidst the same atmosphere of uncertainty that had already led to controversy. Would anyone then argue that procedural fidelity justified inaction? The answer appears fairly obvious. 

The larger point of natural justice is fairness. And fairness is context dependent. In some situations fairness requires a hearing before action. In others, fairness is adequately protected by a hearing immediately afterwards. 

When – “I Disagree” Becomes – “Non-Application of Mind”

The allegation of non-application of mind is a fairly common tendency in public law litigation of equating disagreement with reasoning to absence of reasoning. The two are simply not the same thing.

To establish non-application of mind, one must show that the authority either failed to consider relevant material, considered wholly irrelevant material, or arrived at its conclusion through a process so mechanical that it effectively ceased to be a decision at all. The record, however, shows that there had been weeks of engagement between Telegram, MeitY, NTA and other agencies. And the Government, in turn, also considered reports relating to the persistence of examination fraud, the repeated re-emergence of channels despite takedowns, etc. Indeed, the the non-application argument produces a somewhat odd result. If the Government had simply asserted "public order" and proceeded no further, the criticism would have held. But that is not what happened here. The order identifies a specific harm, explains why existing interventions were thought inadequate, links the proposed restriction to the objective sought to be achieved, and subsequently subjects that reasoning to review through the statutory process (Paras 22-30). 

That may or may not be persuasive – for the critics – but it is nevertheless reasoning.

Then Came Proportionality

We all know by now that proportionality roughly consists of four familiar limbs: (i) a legitimate objective, (ii) a rational nexus between means and ends, (iii) necessity, and (iv) balancing. The Court adopted precisely this from Anuradha Bhasin and applied it to the facts before it. What follows is an examination of that analysis. Some of it tracks the Court's reasoning. Some of it develops arguments the Court only hints at. And some of it responds to the criticisms that have emerged in the aftermath of the decision.

1. Legitimate Aim

The first limb can be disposed of rather quickly. It is difficult to identify any serious critic of the judgment who disputes the legitimacy of the State's objective. The real dispute begins at the level of means and not the ends. 

2. Rational Nexus 

The rational nexus requirement is – again – a relatively low threshold, and rightly so. Constitutional courts are generally not in the business of demanding perfect solutions. As long as there is a logical connection between the measure adopted and the objective sought to be achieved – the Government is good to go. 

In Telegram, the Government's case was that Telegram's platform design enabled the rapid dissemination and regeneration of examination fraud through bots, mirror channels, audience migration mechanisms, and editing features. To this end – then – restricting access to the platform bears an obvious connection to the objective. 

Now one may think the response was excessive. But excessive is not the same thing as irrational. For that you already have the limbs of necessity and balancing (aka least restrictive measure). 

3. Necessity

Every restriction on rights can be met with a less restrictive alternative if one is willing to describe the alternative at a sufficient level of abstraction. The question, however, is whether that alternative is capable of achieving the State's objective to roughly the same degree. What needs to be understood, therefore, is that the test of necessity requires the Government to adopt the least restrictive measure that can reasonably be expected to work.

In this case, channel-specific takedowns, content moderation, targeted removals, could of course have addressed the problem on a more theoretical level. But whether they could do so with comparable effectiveness and in the relevant time space is what necessity explains. 

4. Balancing

The balancing stage asks whether the cost imposed upon rights is disproportionate to the public interest pursued by the restriction. The task of the court is therefore to determine whether the sacrifice demanded of liberty is justified by the objective sought to be achieved.

Again, the Telegram ban appears considerably more balanced than many critics suggest. The restriction undoubtedly burdened the speech and associational interests of millions of users. But the restriction was also temporary, narrowly tied to a specific event, and directed towards preserving the integrity of a national examination process involving over twenty lakh candidates. The question, then, is whether the temporary loss of access to a particular platform imposed a greater cost than the potential harm sought to be prevented. On the facts before the Court – where the restriction was limited in duration, the objective was substantial, and the consequences of inaction were hardly trivial – the Court’s conclusion appears an ordinary exercise in balancing.

One Plus One – Equals Two

The Court’s conclusions on both questions point in one direction. If there was an application of mind, and if the restriction satisfied proportionality, then the judgement is – broadly speaking – a defensible one. One may of course quibble with aspects of the reasoning, and may even prefer a different outcome. But neither of those criticisms are enough to establish infirmity.

There is, however, a much clearer way of understanding why the Court reached the result that it did. While the judgement never says so explicitly, much of its reasoning appears to rest on an idea familiar to comparative constitutional law – i.e. ‘the margin of appreciation’. It is to that idea that I now turn to.

Read(ing) Telegram through a Margin of Appreciation

The margin of appreciation doctrine — most commonly associated with the European Court of Human Rights — draws upon a fairly simple insight. Courts review rights restrictions. They do not govern. Readers may better appreciate this point by referring to – at least briefly – Alexander Bickel's famous counter-majoritarian difficulty (unelected judges second guessing decisions taken by democratically elected institutions). For there will inevitably be situations where the State possesses informational and institutional advantages that courts lack. Questions involving predictive assessments of risk, public order concerns, technological regulation, and the more complex policy choices are obvious examples.

In such situations, while judicial review still exists; courts must not behave as though they are the primary decision makers. A margin of appreciation simply recognises that more than one permissible answer exists. Think about the proportionality doctrine. The necessity limb – for instance – does not come with an instruction manual identifying the single least restrictive measure in every case. Judges ultimately have to make evaluative judgements under conditions of uncertainty. 

And if this is accepted, Telegram becomes much simpler to understand. Was the threat serious enough? Had narrower interventions become ineffective? Did the platform's design create enforcement difficulties? How much weight should be assigned to the risk of examination fraud? These are not purely legal questions. They are mixed questions of fact, policy, technology, and prediction. 

This matters because courts possess no special competence in answering them. More importantly, there is no suggestion anywhere in the record that the Government acted with a coloured intention to suppress speech for its own sake. If the objective is legitimate, the reasoning is genuine, and the measure falls within a range of reasonable responses, courts need not insist on discovering the single best answer themselves. They can acknowledge that the Constitution permits more than one answer and move on.

The Implied Freedom of Political Communication (Thoughts and Observations)

Unlike the Constitution of India, with its entrenched catalogue of fundamental rights in Part III – or constitutions such as those of South Africa, Kenya, Germany, Canada, Ireland, and the United States – the Constitution of the Commonwealth of Australia contains no such bill of rights. To most readers of this blog – accustomed to constitutions announcing rights in elaborate detail and courts spending much of their time deciding what those rights mean – this may appear extremely odd. Yet it is perhaps less surprising than it first seems to be.  

Many modern bills of rights are products of particular historical experiences. They were responses to something that had gone badly wrong. The American constitutional tradition, for instance, is difficult to understand without some appreciation of slavery, segregation and the long struggle over civil rights. South Africa's constitution bears the unmistakable imprint of apartheid and the political realities that accompanied its demise. The same is broadly true of many post-colonial constitutions, which were often drafted against the memory of arbitrary government. Having seen what the state is capable of when left unchecked, constitutional designers frequently sought to place certain matters beyond ordinary politics. Rights guarantees – in this sense – tell us something about the anxieties of the societies that produced them.

The Commonwealth of Australia, however, emerged from a rather different constitutional tradition. Its Constitution was enacted by the British Parliament and drafted largely within the assumptions of nineteenth-century constitutionalism. In many respects it reflects an older – more conventional – understanding of constitutional law. 

[A ‘constitution’ constitutes. It creates institutions, distributes powers, establishes relationships and provides the framework through which public authority is exercised. At its most basic level, constitutional law is the law governing the state itself. And so, to that extent, the classical constitutional project is fundamentally one about institutional design. 

Montesquieu's familiar tripartite division of legislative, executive and judicial authority captures much and most of this intuition.]

And the Australian Constitution remains remarkably faithful to this older model. One searches in vain for anything resembling the First Amendment of the US Constitution or Article 19(1)(a) of the Indian Constitution.

This, then, raises an obvious question. If Australia lacks an express constitutional guarantee of freedom of speech, how does one explain the existence of what has come to be known as the implied freedom of political communication? The answer, I believe, is this: what matters is not what a constitution expressly says but what it necessarily presupposes. For it was from this idea that the implied freedom of political communication emerged.

The doctrine finds its origins in Sections 7 and 24 which requires that members of the Commonwealth Parliament be "directly chosen by the people". Section 64 links ministers to Parliament through the system of responsible government. Section 128 places constitutional amendment in the hands of the electorate itself. Together, these provisions create an order founded upon representative and responsible government. For (1) a people cannot meaningfully choose representatives if they are unable to discuss political matters; (2) Governments cannot be held accountable if criticism of public affairs is prohibited; and (3) Nor can citizens exercise constitutional functions, such as voting in referenda, if political information is unavailable. 

It was on this basis that the High Court first recognised the doctrine Lange v Australian Broadcasting Corporation. The Court there was careful to emphasise that the freedom did not derive from abstract notions of liberty, individual autonomy, or even democratic theory in the broad sense. Rather, it arose from the text and structure of the Constitution itself. The freedom existed because the constitutional system required it. What is important to note, however, is that the implied freedom is not an individual right. It does not confer upon citizens a personal entitlement to speak. It is simply a limitation upon governmental power. The judicial inquiry is therefore whether a law burdens communication relevant to the maintenance of Australia's system of representative and responsible government (as opposed to more conventionally – in a rights based sense – whether an individual's freedom of expression has been infringed).

In Lange, the Court articulated a two-stage test:

1. Does the law effectively burden communication about governmental or political matters in its terms, operation or effect? 

2. If it does, is the law reasonably appropriate and adapted to achieving a legitimate objective compatible with the constitutionally prescribed system of representative and responsible government?

The test has since undergone refinement, particularly through decisions such as McCloy v New South Wales and Brown v Tasmania, where the Court increasingly adopted the proportionality doctrine. 

This foundation helps explain both the reach and the limitations of this ‘implied’ freedom.

1. On one view, the doctrine is remarkably powerful. It has enabled courts to invalidate legislation despite the absence of an express bill of rights. 

2. On another view, it remains narrow since freedom is tied to the functioning of representative government, and not because speech is intrinsically valuable.

That distinction becomes particularly important in contemporary controversies surrounding social media regulation. Australia's decision to restrict access to major social media platforms for those under sixteen has already generated a noticeable ripple effect across a number of jurisdictions. France has moved towards an under-15 ban, Denmark is pursuing restrictions for those under 15, while Greece, Norway, Poland, Slovenia, Austria, Malaysia and Indonesia have all either announced, proposed, or begun implementing comparable age-based restrictions. More recently, the United Kingdom has also signalled its intention to adopt an Australian-style model. 

Now, in jurisdictions possessing constitutional free speech guarantees, the challenge would be comparatively simple. One begins by asking whether expression has been restricted and whether the restriction can be justified. A simple question of proportionality. In Australia, however, the approach cannot simply proceed on the basis that expression has been burdened. The argument must establish something more. It must demonstrate that the burden affects communication relevant to representative and responsible government. 

And this is precisely where much of the current debate lies. Proponents of the legislation tend to characterise the measures as regulatory interventions aimed at child welfare, mental health, and online safety. Opponents, meanwhile, seek to frame social media as an increasingly indispensable platform for political discourse, civic engagement and public participation. The outcome depends largely upon which characterisation prevails.

Conclusion

The implied freedom emerged in a world where political communication largely meant newspapers, public meetings and election campaigns. Contemporary political life, however, has increasingly blurred that understanding. Social movements, political campaigns and public discourse frequently emerge from spaces that would once have been regarded as outside conventional political processes. Nepal's recent elections, where political mobilisation was driven in significant part through platforms such as Discord and TikTok, illustrate the point. It is therefore unsurprising that questions continue to arise regarding the future development of the doctrine. 

If political participation increasingly takes place through digital platforms, should constitutional protection expand accordingly? If representative government depends upon contemporary forms of communication, can constitutional doctrine continue to rely upon assumptions developed in a different technological age?

Australian constitutional law has yet to provide entirely satisfactory answers. For the moment, however, the implied freedom remains what it has always been: not a constitutional right to free speech, but an implication derived from a constitutional system committed to representative and responsible government. It protects communication not because speech is free, but because democracy requires it.

[Update] CLP Piece on the SIR Judgement

I have written a piece for the Constitutional Law and Philosophy blog on the recent SIR judgement. My analysis takes up the Supreme Court's conclusion that the ECI may scrutinise citizenship status for the purposes of electoral eligibility. I argue that the Court's reasoning travels only so far as its express premises permit; beyond that, it generates tensions and contradictions that the judgement fails to resolve. I also highlight the extent to which the Court's analysis rests upon a rather selective memory of the past and lived experience. The title, borrowed from Michel Gondry's 2004 film, reflects that theme.

(You can access it here:  

https://indconlawphil.wordpress.com/2026/06/04/eternal-sunshine-of-the-spotless-judgment-citizenship-despite-everything-in-adr-vs-union/)

Best,