Contesting State Power

We are firmly in the midst of a new narrative being written. A new narrative that touches all aspects of Indian society, from the historical to the sociological, informational, and inevitably, the constitutional. This post examines one proposition of our new Indian narrative – that the power exercised by a validly elected government should not be contested. At its core, this position consists of two elements: (i) the actions of such a government represent the will of the people; and (ii) when in power, the elected representatives of such a government claim sole authorship over the Indian polity – other groups will have their chance at authorship (if) and when they are elected.

In this post, I question these twin assumptions from the perspective or republican self-government. I argue that for a State to be legitimate, it must create a system of popular control where ‘the people’ are the ultimate arbiters of the efficacy of a government. Elections and ‘the will of the people’ are insufficient controls. Rather, what is required is a system of contestation on every site of public-decision making. I seek to demonstrate how despite elections, contestation by ‘the people’ is not an inconvenience to be tolerated, but a vital control on State power that ensures the continued legitimacy of the State as a whole. I end by briefly commenting on the outer limits of contestation, the case of unjust laws and the fraying legitimacy of the State.

The role of the State

States exist to coordinate a community’s behaviour and achieve a better life for the community’s members (e.g. we all drive on the same side of the road, allowing us to drive more safely). At its most basic, a State may protect its members from external threats. People band together on a patch of land for geo-strategic reasons and determine who is worthy of protection. Long before the Indian Constitution talks about rights or principles, it talks land, territory, and citizenship (Art. 1 to 11). Once these primal concerns are addressed, members begin to fine tune their State to better their lives vis-à-vis each other. See Articles in the Indian Constitution attacking centres of dominating private power such as gender, caste, and religion.

To achieve these goals, the State must be empowered to take action. As citizens, we want the State to have the power to protect our borders, we want the State to exercise coercive power to dismantle the inequalities of gender, caste, and religion. However, empowering the State raises a parallel concern – will the State itself become a source of unfreedom? State power when exercised for private or factional interests can be detrimental to the members of a political community. This dilemma lies at the heart of constitutional law. As James Madison summed it up, “In framing a government which is to be administered by men over men, the great difficulty lies in this: you must first enable the government to control the governed; and in the next place oblige it to control itself” (Federalist 51). Madison responds to the dilemma in his very next line – “A dependence on the people, is no doubt, the primary control on the government; but experience has taught mankind the necessity of auxiliary precautions.

Before addressing ‘the people’, a word on ‘auxiliary precautions’: these are design choices to restrain State power and thus lower the risk of it being manipulated. The two most pertinent being the rule of law and the dispersal of power. Legislating in a principled and general (not particularistic) manner and ensuring laws apply equally to all guards against State power being used in a factional or private interest. This is why selective (‘political’) prosecutions or discriminatory legislation are intuitively inimical to us as citizens, because the coercive power of ‘our’ State is being manipulated. Further, States typically disperse power to limit the ability of any one authority drastically restricting a citizen’s freedom. See checks and balances across branches of government, bicameralism, and federalism as other classic examples of breaking up power.   

Control by ‘the people’

However, as Madison noted, ‘the people’ form the primary control on government. The ability of citizens (as a collective) to continually ensure that the coercive power of the State is being employed for their benefit forms a background justification for tolerating the authority of the State. In other words, the State must prove to be acting in the interests of citizens, else it must be abolished.

As Akhil Amar notes, long before the American second amendment protected individual gun enthusiasts, it enshrined a distrust of large standing armies who may not act in the citizens’ interests (see Myanmar in 2021). The American founders firmly believed that a real republic should be guarded by its own citizens, far more reliable than a motley band of outsiders armed to the teeth. Vesting control with citizens means granting them the power to “alter or abolish the established Constitution, whenever they find it inconsistent with their happiness”. A blog post is not the place to discuss the varying textures of the Indian constitution, but our framers were not blind to the need the State to track the well-being of citizens; see the adoption of universal suffrage, parliamentary accountability, and the nexus between the inviolable right of individuals to approach courts (Art. 32), and the power of courts to strike down a broad range of State action (Art. 12 & 13).

The other side of this coin is that where citizens are controlling the State to act in their interests, citizens have a corresponding obligation to obey its general authority (more on this later). But how do we determine whether the State is acting in the interests of the citizens, especially as reasonable citizens can disagree over what constitutes the citizen’s interests in a plural society? And does obeying the general authority of the State mean accepting its every decision?

Contesting Specific Actions

States represent a compromise between the individual and the community. Individual agency sometimes must give way to facilitate coordination (e.g. for all of us to drive safely, I forgo the right to drive on whichever side of the road I choose). However, correspondingly, the State must recognise the claims of individual and groups to autonomy and opinions. One understanding of how this achieved is consent to government action – individuals consent to their autonomy being limited on specific sites. But as Philip Pettit notes,

If explicit individual consent is required, non-arbitrariness in public decision-making becomes an inaccessible idea. If implicit consent is thought to be enough, and an absence of protest is taken as evidence of implicit consent, then non-arbitrariness in public decision making becomes an idea so accessible as to be empty.”      

Philip Pettit, ‘Republicanism: A Theory of Freedom and Government’ (1997).

Within our democratic framework, one solution to this quandary is contestation. More specifically, when citizens or groups disagree with specific actions of a State (usually through its elected government), they have a right to contest this action. At every site of public-decision making, individuals and groups must be capable of raising questions as to whether government action involves the appropriate considerations to further the citizen’s interests. Although elections are a form of contestation, elections alone represent a far too intermittent and imperfect method of popular control (especially in the age of powerful political parties and an anti-defection regime). Thus, elections are not the exclusive form of citizen control. Transparency in government decision making (facilitative), the right to information (facilitative), public consultations as part of Environmental Impact Assessments, jury trials (before they were abandoned), challenging State action in courts, parliamentary sub-committees, parliamentary debate itself, and finally public protest are all institutionalised forms of contestation that ensure State power is continually used for ‘the people’. As Pettit notes,

The self-ruling demos or people may often run on automatic pilot, allowing public-decision making to materialize under more or less unexamined routines. But what makes them self-ruling is the fact that they are not exposed willy-nilly to that pattern of decision making: they are able to contest decisions at will and, if the contestation establishes a mismatch with their relevant interests or opinions, able to force an amendment.”

Philip Pettit, ‘Republicanism: A Theory of Freedom and Government’ (1997).

(Unfortunately, the systemic weakness of many of these forms of contestation has left Indian citizens with few options to retain control over the government other than taking to the streets.)

This stands in stark contrast to populist reasoning that the elected representatives are the sole authors of a nation’s fate. What emerges is system where an elected government has a mandate to act in the citizens best interests, and individual and groups operating through various inflection points within our constitutional democracy keep them honest to this goal. The process of contestation attempts to bridge the all-important gap between individual/group autonomy and the needs of the collective by arriving at a compromise where neither is unduly subordinated. Perfect consensus may not always emerge, but the process of contestation ensures that an individual’s/group’s autonomy is not supressed for the sole reason that its views are at variance with an elected government or ‘popular will’. Rather, compromise is achieved based on publicly-defensible norms.

How should the State be contested?

Recall that where a State is legitimate and generally pursues the citizens interests, citizens have a corresponding obligation to obey the State. How do we square this with the above discussed right of citizens to contest the actions of the State? Simply put, I can continue to respect my State’s general authority while still challenging the exercise of the State’s specific authority. The framers of our Constitution were all too conscious of this crucial distinction. For example, in explaining why the offence of sedition should be scrapped, Ananthasayaman Ayyangar distinguished ‘an attack on the government’ from ‘the entire State being overthrown’ (See Bhatia, ‘Offend, Shock and Disturb’). The former could never be criminalised because, in systems of self-government (which India was about to become), the people’s right to argue that government action was not in the interests of citizens is inherent and a vital control on State power. Based on this rationale, the bar for sedition is not even that seditious speech may lead to violence (this can be dealt with by ordinary criminal law), but rather that the speech must advocate the overthrow of the State itself – i.e. contest its general authority.

It follows that even beyond the institutionally recognised methods of contestation, a citizen’s right to contest specific actions of elected governments continues insofar as they do not contest the general authority of the State itself. This may include breaking a law citizens view as unjust. For example, Pettit notes that the member of a civil disobedience movement may break a perceivedly unjust law but still accept the authority of the State to prosecute and the courts to judge them. When the authority of the State is divided into general and specific, and contestation is understood not as an inconvenience, but as inherent to self-government, the hollowness of a particular grammar of patriotism is laid bare – most notably the use of the term ‘anti-national’ to those critical of a government, or the flagrant use of the sedition provision. 


To recap, in a system of self-government, citizen control over State power forms the background justification for obeying the State’s laws. Viewed from the perspective of autonomy, we are willing to tolerate our loss of autonomy if the State (through its coordinating power) grants us even more autonomy by furthering our interests. In addition to the rule of law and dispersal of power, citizen control is not achieved not solely by electing representatives, but rather through a system of public decision making that allows citizens to contest the specific authority of the State on various sites. Crucially, elections are not an infinite mandate to exercise the coercive power of the State but a conditional power. At any point, the demos may rise up and contest. Where citizens do contest, they may do so through institutional means or even by breaking the law and suffering the penalty. But in a legitimate State they must continue to respect general authority (i.e. not advocate the overthrow of the entire state).

This last point must be caveated given our present circumstances. The overall legitimacy of a State is always a matter of degree. It is therefore possible that that within a largely legitimate State, actions may be taken that “breach the conditions of legitimacy that the regime generally respects”. For example, Tarunabh Khaitan’s recent paper (here) distinguishes between actions that may ordinarily be taken by an elected government and actions that debase liberal democratic constitutionalism itself. Where the fabric of the State itself is changed to rob citizens of continued control, the State may lose a degree of legitimacy and should be prepared for challenges to its general authority.

Note: The intention of this post was solely limited to highlighting the contestatory tenets that underlie our system of self-government and rebutting arguments on the dispositive nature of elections. Readers interested in the comprehensive scholarly accounts of contestatory democracy and republican self-government may refer to: (i) Pettit, ‘Republicanism: A Theory of Freedom and Government’; (ii) Pettit, ‘On the People’s Terms’; (iii) and Raz, ‘The Morality of Freedom’ (on the political obligations of citizens).

Safoora Zargar and Disha Ravi: A Tale of Two Bail Orders

The order granting bail to Disha Ravi in a sedition case is remarkable not so much because of its outcome, but because of the short shrift that it gives to the State’s hysterical accusations of conspiracy-by-Google-Docs. In ordinary circumstances, this would not be remarkable either – judicial skepticism towards the State’s claims of far-reaching conspiracies to justify keeping people in jail, when there exists no evidence linking them to actual violence, should be par for the course. However, that has conspicuously not been the case in recent times, at all levels of the judiciary. Consequently, what makes the bail order remarkable is how (sadly) uncharacteristic it is. Indeed, the order stands in stark contrast to the order of 4th June 2020, that denied bail to Safoora Zargar, in what have come to be known as “the Delhi Riots cases.” A comparison between the two, therefore, merits scrutiny.

In the aftermath of Disha Ravi’s bail, it did not escape public attention that both orders were handed down by the same judge. It is important to clarify, however, that the purpose of this post is not to serve up some kind of a gotcha! conclusion, or point to judicial hypocrisy. As all human beings, judges themselves evolve over time and reflect on their own judgments and orders, and one can hope – in light of yesterday’s order – that future cases before the same judge would follow his more recent convictions on liberty, free expression, and skepticism of State power, rather than his older ones. The comparison is merited, however, the underlying State case in both circumstances was remarkably similar (and indeed, appears to be following a specific legal “toolkit”, if one may use that word), and the verdicts in both cases reflect two starkly opposed judicial approaches to the question of personal liberty.

Admittedly, there is a significant legal difference between the two cases, in that while Disha Ravi was “only” accused of sedition, Safoora Zargar was booked under the UAPA, whose Section 43(D)(5) throws up significant barriers to granting bail, barriers that have been made worse by the Supreme Court’s notorious Watali judgment. A perusal of the two bail orders reveals, however, that in this context, the distinction is largely irrelevant. Even though Disha Ravi’s counsel specifically argued that sedition is a milder offence that – even on conviction – allows judges to levy only a monetary fine by way of sentence, this did not form the basis of the bail order. Even the classic bail conditions – whether the accused is a flight risk, and whether the accused can tamper with the evidence or the witnesses – occupy only a small space in the Disha Ravi bail order; for the most part, the learned ASJ focuses on whether, on the facts presented by the Prosecution, a prima facie case of sedition is made out or not (and finds that it isn’t). This is exactly how he proceeded in the Safoora Zargar case – and indeed, under 43(D)(5) of the UAPA, bail can be granted if no prima facie case has been made out.

This is where the similarities between the two cases become important. In both cases, there were certain incidents of violence. Neither Disha Ravi nor Safoora Zargar had engaged in violence, been present at the scene of the violence, or incited violence (there were vague claims that Safoora Zargar had given “inflammatory speeches”, but these were not adverted to by the Court, and did not form part of its order). For this reason, in both cases, the Prosecution attempted to pin the actions or speeches of others onto the accused, by alleging the existence of an overall conspiracy, of which the accused were a part.

How did the Court deal with this? In Disha Ravi’s case, it very correctly noted that “conspiracy cannot be proved merely on the basis of inferences. The inferences have to be backed by evidence” (para 22). Dealing with the Prosecution’s contention that Disha Ravi was in contact with secessionists, the Court then noted:

“… it is not mere engagement with persons of dubious credentials which is indictable rather it is the purpose of engagement which is relevant for the purpose of deciding culpablity. Any person with dubious credentials may interact with a number of persons during the course of his social intercourse. As long as the engagement/interaction remains within the four corners of law, people interacting with such persons, ignorantly, innocently or for that matter even fully conscious of their
dubious credentials, cannot be painted with the same hue. In the absence of any evidence to the effect that the applicant/accused agreed or shared a common purpose to cause violence on 26.01.2021 with the founders of PJF, it cannot be presumed by resorting to surmises or conjectures that she also supported the secessionist tendencies or the violence caused on 26.01.2021, simply because she shared a platform with people, who have gathered to oppose the legislation.” (para 22)

Now this is entirely correct, and follows a long line of Supreme Court precedent that has clarified that if you’re going to damn someone by association, it has to be active association (in UAPA cases, active membership of banned organisations). This is crucial, because without limiting the chain of association, just about anyone who has ever shared a platform or interacted in any form with any dubious individual can be brought into the net of criminality through the legal device of a “conspiracy”, and incarcerated for years without trial. But this is also what the Court very conspicuously did not do Safoora Zargar’s case; in fact, in that case, the Court said:

“Further, even if no direct violence is attributable to the applicant/accused, she cannot shy away from her liability under the provisions of the said Act [UAPA]. When you choose to play with embers, you cannot blame the wind to have carried the spark a bit too far and spread the fire.”

But this is precisely the endlessly extendable guilt-by-association doctrine that the Court (correctly) rejected in Disha Ravi’s case. Safoora Zargar was damned specifically for (allegedly) having associated with people who engaged in violence, without any evidence against her – either by way of commission of violence, or by way of incitement. In Disha Ravi’s case, the Court (correctly) insisted that allegations of conspiracy require evidence, and found evidence lacking, as there was no common intention and no proof of anything beyond allegations of mere association; in Safoora Zargar’s case, the Court began by accepting the allegations of conspiracy, and then, on that basis, imputed the actions of some of the alleged conspirators to Safoora Zargar as well, because of their mere association. Ultimately, the issue boils down simply to this: in one case, the Court insisted that the State back up its claims against the specific, criminal culpability of an accused, including as part of a larger conspiracy; in the other case, it didn’t.

The difference in the two approaches is particularly stark on the key legal issue of the relationship between disaffection in violence. In Disha Ravi’s Case, after citing the judgment in Kedar Nath Singh, the Court noted:

Evidently, law proscribes only such activities as would be intended, or have a tendency, to create disorder or disturbance of public peace by resort to violence. ‘Violence’ seems to be the gravamen of the charge. (para 20)

In Safoora Zargar’s case, after citing the judgment in Kedar Nath Singh, the Court noted:

Therefore, evidently, law frowns upon any act which has a tendency to create disorder or or disturbance of law and order or incitement to violence. Therefore, mere violence is not the gravamen of the charge under u/s 2(o) of the UAPA.

The substitution of the word “any” with “only”, and the removal of the word “not” led the Court to a 180-degree turn, when considering the same judgment (Kedar Nath Singh) and the same legal term (disaffection, albeit in different laws). The distinction is crucial, because how strongly the Court requires the causal link to be established between the accused and the act of violence is the fact upon which everything else turns. Disha Ravi was (correctly) granted bail because the Court asked for evidence. Safoora Zargar was (wrongly) denied bail because the Court relied on a metaphor.

As I clarified at the beginning, the point of this post is not to play gotcha! games; at the same time, inconsistency in judicial pronouncements in matters relating to personal liberty presents a serious challenge to the rule of law. The Disha Ravi case shows that notwithstanding the Supreme Court’s Watali judgment, it is entirely possible for a trial Court judge to examine the basics of the State’s “conspiracy” claim, and find them utterly lacking in evidence or particulars, thus completely breaking the chain of causation between the accused and the act of violence. All it needs is a dose of judicial skepticism, some probing questions, and clarity on the legal position that whether it is sedition of the UAPA, violence (and incitement thereof) is the gravamen of the charge, and even the Prosecution’s prima facie case requires to pass at least that basic threshold before the bar under 43(D)(5) is invoked. If that is done consistently, then we will have no more Safoora Zargar cases, and no more imprisonment by metaphor.

Vikash Kumar v UPSC: An Important Judgment on Disability and Reasonable Accommodation

Editor’s Note 1Posts about the contemporary Supreme Court may be read in the context of the caveats set out in this post (link).

Editor’s Note 2: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations (link) against a former Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances (e.g., the introduction of structural mechanisms to ensure accountability)].

On the 11th of February, a three-judge bench of the Supreme Court handed down an important judgment in Vikash Kumar vs Union Public Services Commission. The Appellant was a civil services aspirant who suffered from writers’ cramp. However, his application for writing the Civil Services Examination with the assistant of a scribe was rejected by the UPSC, on the basis that he did not have a “benchmark disability”, as defined by the Persons With Disability Act, 2016 (in this case, limited to blindness, locomotor disability, or cerebral palsy, to the extent of 40%). On a perusal of the relevant rules and the Government Office Memorandum issued by the Ministry of Social Justice and Empowerment, the Court found that while individuals with a benchmark disability were entitled to a scribe, those individuals who did not fall within those categories could also be allowed to have one, in the event that they could produce a certificate to that effect from an approved government institution. Indeed, while the UPSC argued in Court that it was not permitted to deviate from the rule of benchmark disability in allowing for a scribe, the Ministry of Social Justice itself observed that there did exist non-benchmark disabilities that could significantly hamper writing ability. The Court then found that this was buttressed by the statutory policy as well: the PwD Act had a separate category for benchmark disability, that was limited to the issue of reservations; it was therefore unwarranted to deny other rights to the residual category of non-benchmark persons with disability. As Chandrachud J., writing for the Court, observed:

The second concept which is embodied in Section 2(s) is that of a person with disability. Section 2(s) unlike Section 2(r) is not tagged either with the notion of a specified disability or a benchmark disability as defined in Section 2(r). Section 2(s) has been phrased by Parliament in broad terms so as to mean a person with a long term physical, mental, intellectual or sensory impairment which in interaction with various barriers hinders full and effective participation in society equally with others. (paragraph 26)  

Having established that the mere absence of a benchmark disability was not sufficient cause to deny the appellant the benefit of a scribe, the Court then went on to engage with the statutory purpose of the PwD Act. The Court noted that the purpose of the PwD Act was to deepen the social commitment to equality, and impose positive obligations – both upon the State and upon the private sector – to ensure that its objectives were fulfilled. For this reason, as an interpretive matter, limiting certain rights ipso facto to persons with a benchmark disability was clearly contrary to the statutory purpose:

Except in the specific statutory context where the norm of benchmark disability has been applied, it would be plainly contrary to both the text and intent of the enactment to deny the rights and entitlements which are recognized as inhering in persons with disabilities on the ground that they do not meet the threshold for a benchmark disability. (paragraph 41

Next, the Court considered the concept of reasonable accommodation, at the heart of the PwD Act. The Court noted that, in accordance with the PwD Act, disability was primarily a social construct, in the sense that the barriers imposed upon disabled individuals were because of the way society constructed itself, with a certain concept of able-ness as the norm (a good example of this is the use of stairs – and not ramps – as default structures to connect levels within a building). Consequently:

The principle of reasonable accommodation acknowledges that if disability as a social construct has to be remedied, conditions have to be affirmatively created for facilitating the development of the disabled. Reasonable accommodation is founded in the norm of inclusion. Exclusion results in the negation of individual dignity and worth or they can choose the route of reasonable accommodation, where each individuals’ dignity and worth is respected. Under this route, the “powerful and the majority adapt their own rules and practices, within the limits of reason and short of undue hardship, to permit realization of these ends.” (paragraph 45)

Chandrachud J. made four important points about the principle of reasonable accommodation, as set out under the PwD Act: first, it was an individualised principle, which meant that the needs of individuals would have to be considered on a case-by-case basis; secondly, as the purpose of the PwD Act was to advance equality, the burden would lie upon the entity denying reasonable accommodation, rather than the one seeking it; and thirdly, the obligation was immediate in nature – i.e., the right to reasonable accommodation was directly enforceable, and not subject to gradual or incremental fulfilment; and fourthly, reasonable accommodation required meaningful dialogue – or engagement – with the affected individual to determine how best to overcome the barrier in question (paragraphs 44 – 46). Each of these is a crucial interpretive finding in the context of the PwD Act, and the consequences remain to be worked out in the fullness of time.

Importantly, the Court also noted that its 2019 judgment in V. Surendra Mohan v State of Tamil Nadu, which had been severely criticised for its refusal to allow a visually disabled person from becoming a judge, was no longer good law, as it failed to take into account the principle of reasonable accommodation. In particular, Chandrachud J. observed:

By definition, reasonable accommodation demands departure from the status quo and hence ‘avoidable complications’ are inevitable. The relevant question is whether such accommodations would give rise to a disproportionate or undue burden. (paragraph 54 )

Two important inferences follow from this. First, the kind of situation that was upheld as lawful in Mohan – where a blanket 50% visual impairment bar was imposed on qualification for judicial service – would be ipso facto unconstitutional, as it would make reasonable accommodation – in its individualised component – unenforceable; and secondly – other then the evidentiary burden lying upon the entity (State or private sector) denying reasonable accommodation, the legal standard to be met would be that of showing that reasonable accommodation would cause an “undue” or “disproportionate” burden. While this legal standard doesn’t entirely address the basic issue of a world where the norm is that of able-ness (because, by allowing denial of accommodation in “undue burden” cases, it retains able-ness as the norm), it goes some way towards doing so – and much will depend on how future judges interpret the term “undue burden.”

In conclusion – and in this specific case – the Court rejected the Union government’s arguments of potential misuse (using the striking analogy that the solution to copying in an exam using “chits” was not to impose a dress code that would make it impossible for some people to write the exam altogether), directed that the Appellant be allowed a scribe, and also directed the Ministry of Social Justice and Empowerment to frame guidelines on the issue of scribes, taking into account the individualised character of reasonable accommodation.

Vikash Kumar marks the Supreme Court’s first serious engagement with the concept of reasonable accommodation under the PwD Act, and sets out some important principles to help ensure that the Act can fulfil its role in advancing substantial equality under the Constitution. It lays a strong and durable foundation for future cases to build upon.

Guest Post: Union of India v K.A. Najeeb – a Ray of Hope for UAPA Undertrials?

Editor’s Note 1Posts about the contemporary Supreme Court may be read in the context of the caveats set out in this post (link).

Editor’s Note 2: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations (link) against a former Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances (e.g., the introduction of structural mechanisms to ensure accountability)].

[This is a guest post by Bharat Harne.]

The Supreme Court (SC) on 1st February 2020 delivered an important judgment, Union of India v KA Najeeb (K.A. Najeeb), related to granting of bail in Unlawful Activities Prevention Act  (UAPA) 1967 cases. The Court held that any constitutional court has the power to grant bail to people accused of offences under UAPA irrespective of Section 43-D (5), so as to enforce the right to speedy trial which is guaranteed under Article 21 of the Constitution.  The judgment has been hailed as the step in the right direction, given that the stringent provisions of Section 43-D (5) makes it almost impossible for a person to secure a bail for an offence under UAPA and under-trials languish in jail while the trial drags on for years. In this post I will highlight the importance of the judgment in the context of the operation of UAPA on ground and will try to answer some of the questions which K.A. Najeeb has thrown up.

UAPA – A Tool of Oppression

 The UAPA creates an alternate criminal justice system where the Code of Criminal Procedure (CrPC) does not apply and there are little safeguards for the accused. Empirical research has shown that two-thirds of the accused end up getting acquitted. However, the criminal trial drags on for years and most of the accused end up serving significant amount of time in jail before the trial concludes. This is primarily because of Section 43(D)-5 of the UAPA. According to Section 43(D)-5 a person accused of an offence under UAPA cannot be released on bail if, on a perusal of case diary or the report made under s. 173 of CrPC, the court is of the opinion that there are reasonable grounds for believing that the accusation against such person is prima facie true. Notice that the standard of prima facie is extremely low. In NIA v Zahoor Ahmad Shah Watali held that the to satisfy the standard of prima facie no elaborate scrutiny or dissection of the material is required. Simply put, the court merely has to rely on the words of the investigating agency and see whether the allegations fit the offences. In view of such stringent bail provisions and lengthy trials there was no way for undertrials to get bail even though they ultimately they might end up getting acquitted. Thus, an accusation under UAPA becomes as good as conviction and a way for the State to punish people without subjecting them to a fair trial.

The lack of any interim relief paved the way to a particularly odious practice i.e., informal plea bargaining. Since undertrials languished in jails for years and years they end up serving a significant portion of the sentence of the crime they are accused of even before the trial has concluded. This leads to the accused reaching an informal arrangement with the prosecutors wherein the former plead guilty, resulting in a conviction that is usually for the period already served as undertrial or a reduced sentence. It is not unlikely that quite a few accused who were actually innocent pleaded guilty just so that they can get out of prison. The fact that a person is forced to plead guilty for a crime that they did not commit is illustrative of how UAPA is doing grave injustice to the accused and violating their right to liberty.

It is in this context that the decision of the SC becomes very important. The court has held that bail can be granted to an undertrial irrespective of Section 43-D (5) of the UAPA if the court finds that right to speedy trial under Article 21 is being violated. In the instant case the accused had spent nearly five years in prison out of maximum 8-10 for which he could potentially get convicted. Moreover, 276 witnesses who were left to be examined when the matter came before the SC. It held Section 43-D (5) is not a bar when right to speedy trial is violated and accordingly granted bail in this case. The decision of the court is welcome, however the decision itself has opened up a few questions which I will now explore.

A Few Concerns

First, it is not exactly clear what is the standard used by the Court to arrive at the finding that right to speedy trial has been violated. The court seemed to have been led by two considerations – a) the period of time spent in jail and b)- the possibility of early conclusion of trial. With respect to the first condition the accused had spent five out of the maximum of 8-10 years for which he could be convicted. However, the court stopped short of giving any principled reasoning which could be used in future cases to decide whether the right to speedy trial has been violated. Currently s. 436A of CrPC (which does not apply to UAPA) states that if an accused has spent half the period of time out of the maximum period of punishment specified for that offence, the accused has to be released on bail. This provision was not a part of the original CrPC and was added in 2005. Although the Statement of Object of Reasons to the CrPC amendment does not talk about s. 436A, the rule of giving bail on spending half the time specified for that offence is prison emerges out of the jurisprudence of SC in cases concerning right to speedy trial. In Hussainara Khatoon v Home Secretary, State of Bihar (which laid the foundation of right to speedy trial)- the court observed how several undertrial prisoners has spent more than one half of the maximum punishment of which they could be convicted. The court directed the government to appoint lawyers for such undertrials and file an application for bail. In Supreme Court Legal Aid Committee v UoI, another case dealing with right to speedy trial, the court issued a number of directions with respect to pending cases. One such direction was to release those undertrials who were accused of an offence under NDPS Act carrying maximum punishment of five years or less and have spent time in jail which is not less than half the punishment provided. It was only after this line of cases that s. 436A was added to the CrPC. It can be safely be assumed that the legislature was inspired by jurisprudence on right to speedy trial. It is therefore submitted that the rule contained in s. 436A of CrPC i.e., half the maximum punishment as undertrial can be a good standard for UAPA cases where the right to speedy trial is being prayed because the source of this rules itself is jurisprudence on right to speedy trial. If not as a fixed rule, it should at least give a presumption that the right to speedy trial has been violated and then the burden should shift on the state to justify continued detention.

Second, the court also took into consideration the time which it would take to conclude the trial. The court was of the opinion that in the present case since 276 witnesses were yet to be examined the trial will take long to finish and the accused ought to be released. It will be noticed that in the present case the court is concerned with a particular stage of criminal justice process, i.e., trial. However, it is well settled that right to speedy trial extends to all stages including investigation, inquiry, trial, appeal, revision and re-trial. In the context of UAPA it is important to pay attention to one particular stage i.e., investigation. This is because the NIA which is India’s anti-terror agency deliberately slows the investigation to keep the accused in prison for as long as possible. This is done by filing chargesheets and supplementary chargesheets with long gaps, which ensures that the trial is kept in suspended animation and the accused is in prison. This is especially done when the agencies know that their case is weak and will not stand a trial. This strategy has been adopted by the agencies in Delhi Riots case and Bhima Koregoan case. In both these cases the accused have been mostly denied bail. Since the right to speedy trial includes the stage of investigation the courts in the future will have to take this reality into account and expand the scope of SC’s decision in K.A. Najeeb. The SC’s jurisprudence on speedy trial and long period of investigation itself does not inspire much hope. In Rahubir Singh v State of Bihar the police of Bihar were accused of delaying the investigation to keep the accused in jail at any cost. The accused were a group of people caught while they were secretly attempting to cross the Indo-Nepal border at the height of Sikh militancy in 1984. It turned out that one of the persons was himself suspended from Indian Police Services for his anti- India activities. The court observed that the investigative agencies were justified in extending the investigation because the case involved ‘suspected conspiracies bristling with all manners of complexities’ and therefore even though there were ‘lulls’ in investigation it cannot be said that right to speedy trial was violated. While these considerations might have been true in that case, it is well known that arguments of ‘national security’, and ‘conspiracies involving complex investigation’ are a slippery slope to giving the executive free hand to trample on the liberties of its citizens. This exceptionalism might well play a role in future cases because the UAPA itself deals only with such suspected conspiracies and threats to national security and the Indian judiciary is notorious for buying into such arguments of ‘national interest’ and ‘national security’ too easily.

Third, India’s bail jurisprudence is notoriously inconsistent and discriminatory. For example, the MP High Court did not even apply the well settled principles of bail law in Munawar Faruqui case and inexplicably relied on Fundamental Duties to deny bail. Similarly, while bail was granted to Arnab Goswami in Article 32 petition because the courts thought the police was misusing its power, journalist Siddique Kaplan is languishing in jail and his Article 32 petition was rejected and when Arnab Goswami case was cited, it was simply remarked that every case is different. If constitutional courts fail to apply even well settled principles of bail law to regular cases, only time will tell how K.A. Najeeb will be applied in the future especially because the judgment does not even lay down any concrete principle on which bail is to be given and therefore remains susceptible to inconsistent and unprincipled application.


K.A. Najeeb has the potential to remedy injustice that is caused by stringent bail condition under s. 43-D(5) of UAPA.  However, this potential can be realized only if there is principled application of the judgment. In K.A. Najeeb the court relied on two considerations the period of time in jail as undertrial and the time left for conclusion of trial in order to determine whether right to speedy trial was violated but failed to specify any principled rule to decide future cases. In this post firstly, I have argued how the law can take shape in the future firstly showing how the rule in s. 436A of the CrPC can be a good indicator if not a brighline rule for violation of right to speedy trial; secondly, I have argued that the courts need to take into account the delaying tactics used by investigative agencies to effectively enforce right to speedy trial. In conclusion it is submitted that only if we have some settled principle based on which bail can be given in UAPA cases can we hope to effectively enforce right to speed trial and remedy the injustice caused by UAPA.

Guest Post: From Foreigners Tribunal in Assam to Justice Pushpa Ganediwala – A Question of Decisional Independence?

[This is a guest post by Talha Abdul Rahman.]

Reportedly, in Assam’s Foreigners Tribunal, the renewal of annual contract of its Members depends upon the ‘conviction rate’. In other words, it means that if the members of the Foreigners Tribunals declare a higher number of persons as foreigners, higher are their chances of being retained on their posts.  Elsewhere,  I have criticised Foreigners Tribunals in Assam for a variety of reasons, including being an affront to decisional independence.

‘Decisional Independence’ is a facet of ‘independence of the judiciary’ where every individual judge is (expected to be) insulated from external forces that may influence her decisions. This includes factors that may move her conscience to please the Government or a party to the litigation. As a doctrine, this even requires a judge to be able to think and apply her mind independently from other judges on the bench.  Thus, when members of the Foreigners Tribunals are expected to have a certain ‘conviction rate’ just to be able to remain in employment or are expected to please the body ‘supervising’ their function, it naturally means that their decision making process lacks the required hallmark – independence. This fear of ‘reprisals’ for deciding cases according their conscience denudes the legal system of its legitimacy.

The situation of a member of the Foreigners Tribunal whose contract is not renewed and the situation of Justice Pushpa Ganediwala whose judicial appointment to be made permanent has been withdrawn by Supreme Court is qualitatively comparable. A judge cannot be punished for wrong judgments which are otherwise within her jurisdiction. This is because when a judgment is bad, there are ways to have it corrected. I have no quarrel that the reasons given by Justice Ganediwala’s judgment in her judgment in the recently reported POCSO Cases  are perverse – and in the same breath, all of us can count at least half a dozen other judgments of High Courts and Supreme Court whose reasoning we find perverse.  However, to punish a judge for taking a particular view on the matter is not supported by the Constitution. 

It would be a good moment to also recall that Justice R. N. Agarwal and Justice U. R. Lalit were not confirmed as judges of the High Court for their actions in blocking the State actions during the Emergency. Between the Emergency and today, we are still doing the same thing our judges – albeit for seemingly different reasons.

One may argue that non-confirmation of an additional judge of a High Court is not a ‘punishment’. It is much like an employee on probation who has not been confirmed – there is no vested right.  To this,  I have two preliminary replies: a judge’s position is different from an ordinary employee, and second, once the services of a probationer have been confirmed, an avenue of withdrawal of confirmation may not necessarily be available under the service rules.  And, on facts, can one really say that the action of withdrawal of recommendation is not punitive?

In Supreme Court Advocates-on-Record Assn. v. Union of India, (2016) 5 SCC 1, the Hon’ble Supreme Court has held that:

714. The independence of the judiciary takes within its fold two broad concepts: (1) Independence of an individual Judge, that is, decisional independence; and (2) Independence of the judiciary as an institution or an organ of the State, that is, functional independence. In a lecture on Judicial Independence, Lord Phillips [ Former President of the Supreme Court of the United Kingdom and Lord Chief Justice of England and Wales] said: “In order to be impartial a Judge must be independent; personally independent, that is free of personal pressures and institutionally independent, that is free of pressure from the State.”

718. The entire package of rights and protections ensures that a Judge remains independent and is free to take a decision in accordance with law unmindful of the consequences to his/her continuance as a Judge. This does not mean that a Judge may take whatever decision he/she desires to take. The parameters of decision-making and discretion are circumscribed by the Constitution, the statute and the Rule of Law. This is the essence of decisional independence, not that Judges can do as they please.

726. Generally speaking, therefore, the independence of the judiciary is manifested in the ability of a Judge to take a decision independent of any external (or internal) pressure or fear of any external (or internal) pressure and that is “decisional independence”. It is also manifested in the ability of the institution to have “functional independence”. A comprehensive and composite definition of “independence of the judiciary” is elusive but it is easy to perceive.

The judgment in Supreme Court Advocates-on-Record Assn (supra) makes it rather clear that any action that may be destructive of decisional independence would not pass the muster of the basic structure test of independent judiciary. The precedent of withdrawing recommendation to make an additional judge permanent does not appear to be happy one. Admittedly, the case of Justice Pushpa Ganediwala is a hard case. Since hard cases make bad law, I think the issue now, for the future, is not so much of the withdrawal, but of what factors that went in approving her appointment in the first place?

Postscript: At some point, the Supreme Court must look at the working conditions  of contractually appointed members of the Foreigners Tribunals of Assam and of the Foreigners Tribunals itself. I mentioned it in the beginning, not only for its shock value – but because it is a unique tribunal that is being systemically and systematically constrained to declare Indians as foreigners rendering them Stateless – and that is the real point of this article. The opinions produced by Foreigners Tribunals in Assam are as shocking as judgments of Justice Pushpa Ganediwala  but in Assam the authors of such perverse judgements are rewarded by renewing their contracts. Their members compete to be the highest wicket takers – a pejorative for declaring persons foreigners.

Guest Post: A Critique of the Supreme Court’s Farm Act Order – II

[This is a guest post by Aakanksha Saxena.]

The Supreme Court recently passed an order in discrete batches of petitions arising out of (1) the Farmers’ Produce Trade and Commerce (Promotion and Facilitation) Act, 2020; (2) the Essential Commodities (Amendment) Act, 2020; and (3) the Farmers (Empowerment and Protection) Agreement on Price Assurance and Farm Services Act, 2020, (‘farm laws’), and, the protest by farmers against these laws. Challenges have been filed to the constitutional validity of the laws. An initial petition filed was a PIL by a law student seeking removal of the protesting farmers from the Delhi borders. Another batch of these petitions deserves mention is that it reportedly “supports the validity” of the farm laws – something which is a fundamental presumption in the domain of judicial review.

The Order in question presents myriad concerns. While appreciating that the farmer’s protests have been peaceful, it is insinuated in the same vein that they may be a site of fomenting units which are banned for their secessionist tendencies – on the basis of a mere averment in an intervention application, which was nonetheless pressed by the Attorney General (although, in the final order, directions have been passed to file an affidavit). Further, while noting specifically the absence of counsel representing the farmers, the Bench has proceeded to pass orders indubitably affecting their rights. It appears that the Court has recognised the need for stakeholder participation and consultation as an essential step in policy making, by appointing the expert Committee which is to receive views from all the stakeholders involved; however, what failed to be appreciated is that such a process was integral and ought to have been implemented much prior in time than this hearing, and in any event prior to the enactment of the farm laws. Given the large-scale protests and evident grievances raised by the farmers, the decision-making process leading up to the farm laws could have benefited from stakeholder participation and could also perhaps have avoided agitation and litigation of this nature. The failure to conduct stakeholder participation has clearly led to a situation enabling intervention by the judiciary in the realm ordinarily required to be occupied by the legislature and / or policy makers. Much has been written and said about the impact of the Court’s Order on the farmers’ protests and the natural political posturing and consequences, but this post shall deal with one terse paragraph of the Order which touches on the aspect of judicial review, i.e., whether the Hon’ble Court could pass an interim stay of the impugned Farm Laws.

A reading of the Order discloses that the reasons which weighed with the Bench for staying the operation of the farm laws could be – that negotiations between the farmers’ groups and the Government had been fruitless, and an expert committee to act as a negotiator between the sides would “create a congenial atmosphere”, that some of the farmers bodies’ agreed to go before a Committee, and that senior citizens, women, and children would then be discouraged from protesting which was posing grave risk to these groups. There is not a single statement or suggestion in the Order that the farm laws may, prima facie, be unconstitutional.

Juxtaposed against the Court’s reasoning for the stay, this post seeks to examine the constitutional standards laid down in our jurisprudence for the stay of legislation, particularly economic policy legislation (the standard being somewhat less stringent than laws touching on fundamental and civil rights). It needs to be emphasised that this case was apparently made out by the Attorney General, and this post shall deal with some of those very same precedents that were cited at the Bar, being the inescapable law laid down by the apex court.

In its decision in Siliguri Municipality v. Amalendu Das, the Court was at pains to point out to the concerned High Court the need for self-discipline when it came to interim orders of stay, when the question arose in relation to tax recovery under narrowly applicable state legislation. The Court stressed on the need for a bench to consider the exigencies of the situations and strike the delicate balance. Subsequently, the Court while considering a challenge to Section 9 of the Reserve Bank of India Act, in Bhavesh D. Parish v. Union of India, then went on to consolidate the standard of judicial review by completely deferring to the legislature on economic policy and specifying that any interdiction by courts therein could lead to ramifications which could even retard progress by years. It was expounded that the Court ought only interfere where it was satisfied that a view in the legislation was such that it was “not possible to be taken at all”. This has been reiterated in the case of both legislation and executive policy, when the Court in Bajaj Hindustan Ltd. v. Sir Shadi Lal Enterprises Ltd. clearly held that the court must leave the authority to decide its full range of choice within the executive or legislative power, and in matters of economic policy, the court gives a “large leeway” to the executive and the legislature. In Swiss Ribbons v. Union of India, the Supreme Court while considering a constitutional challenge to provisions of the Insolvency and Bankruptcy Code, expounded this principle as a “Judicial Hands off qua economic legislation”, which flows from SCOTUS Justice Holmes’ celebrated dissent in Lochner. It became the established position that legislatures may do as they feel fit unless restrained by constitutional prohibition, which prohibitions courts cannot extend merely by reading into them conceptions of public policy.

In Health for Millions, the concerned High Court had stayed the operation of Rules in Article 226 writ petitions, by issuing a single-paragraph order stating that ad-interim relief was granted since the Union of India had failed to appear and/or file reply. The Supreme Court strongly reiterated that passing such stay orders was at odds with the need for a court to refrain from staying the operation of legislation, unless the court is convinced that the legislation is patently unconstitutional and factors such as balance of convenience, irreparable injury, and public interest, favour interim stay. It therefore became necessary to set aside the interim impugned order given that the High Court did not consider any of these requirements.

The judgement in Dr. Jaishri Laxmanrao Patil v. The Chief Minister & Anr. referred to in the Order also needs distinguishing for several reasons – it was in a civil appeal, from an order of the Bombay High Court i.e. a constitutional court which had adjudicated upon the validity of the legislation in question, arguments were extensively heard both in support of and against the order under challenge, merit was found in the argument to refer the question of law to the constitution Bench, and it was in that context that the Court found it was not restrained in passing orders to cover the interregnum before the larger Bench presided. At the very least the Bench in Jaishri Laxmanrao Patil engaged with arguments made for and against the stay of the operation of the law, examined the law and the exigencies, and reached a prima facie view on its validity before staying its operation. However, the same cannot be said of the Farmers’ Laws Order, where ostensibly, the hearing was being held in order to ameliorate the heightening tensions and pressure on the government arising out of what were admittedly peaceful, non-violent protests. Of highest concern should be the manner in which the stay was granted in the face of overwhelming binding precedent requiring a constitutional court to expressly reach a prima facie view of unconstitutionality of legislative measures before staying their operation and/or passing any interim measures. In view of the case made out for urgent hearing, the Supreme Court instead of hearing the challenge, chose to appoint a negotiator, and a committee of experts as a negotiator at that. It does not fall to the Supreme Court to direct such committee appointments and stakeholder participation at the stage of a constitutional challenge. It cannot fall from the Supreme Court to stay the operation of a law, de hors a prima facie view of its unconstitutionality, and merely in order to facilitate a political negotiation. The potential effect on future courts is unimaginably dangerous.

Notes From a Foreign Field: The European Court of Human Rights on Russia’s Website Blocking [Guest Post]

[This is a guest post by Gurshabad Grover and Anna Liz Thomas.]

From PUBG to TikTok, online services are regularly blocked in India under an opaque censorship regime flowing from section 69A of the Information Technology (IT) Act. Russia happens to have a very similar online content blocking regime, parts and processes of which were recently challenged in the European Court of Human Rights (‘the Court’). This blogpost summarises the human rights principles applied by the Court to website blocking, and discusses how they can be instructive to petitions in the Delhi High Court that challenge arbitrary censorship in India.

Challenges to Russia’s Website Blocking Practices

On 23 June 2020, the Court delivered four judgements on the implementation of Russia’s Information Act, under which content on the internet can be deemed illegal and taken down or blocked. Under some of these provisions, a court order is not required, and the government can send a blocking request directly to Roskomnadzor, Russia’s telecom service regulator. Roskomnadzor, in turn, requests internet service providers (ISPs) to block access to the webpage or websites. Roskomnadzor also notifies the website owner within 24 hours. Under the law, once the website owner notifies the Roskomnadzor that the illegal content has been removed from the website, the Roskomnadzor verifies the same and informs ISPs that access to the website may be restored for users.

In the case of Vladimir Kharitonov, the complainant’s website had been blocked as a result of a blocking order against another website, which shared the same IP address as that of the complainant. In Engels, the applicant’s website had been ordered by a court to be blocked for having provided information about online censorship circumvention tools, despite the fact that such information was not unlawful under any Russian law. OOO Flavius concerned three online media outlets that had their entire websites blocked on the grounds that some of their webpages may have featured unlawful content. Similarly, in the case of Bulgakov, the implementation of a blocking order targeting extremist content (one particular pamphlet) had the effect of blocking access to the applicant’s entire website. In both the cases of Engels and Bulgakov, where court proceedings had taken place, the proceedings had been concluded inter se the Prosecutor General and server providers, without the involvement of the website owner. In all four cases, appeals to higher Russian courts had been summarily dismissed. Even in those cases where website owners had taken down the offending content, their websites had not been restored.

The Court assessed the law and its application on the basis of a three-part test on whether the censorship is (a) prescribed by law (including foreseeability and accessibility aspects of the law), (b) necessary (and proportionate) in a democratic society, and (c) pursuing a legitimate aim.

Based on the application of these tests, the Court ruled against the Russian authorities in all four cases. The Court also held that the wholesale blocking of entire websites was an extreme measure tantamount to banning a newspaper or a television station, which has  the collateral effect of interfering with lawful content. According to the Court, blocking entire websites can thus amount to prior restraint, which is only justified in exceptional circumstances.

The Court further held that procedural safeguards were required under domestic law in the context of online content blocking, such as the government authorities: (a) conducting an impact assessment prior to the implementation of blocking measures; (b) providing advance notice to website owners, and their involvement in blocking proceedings; (c) providing interested parties with the opportunity to remove illegal content or apply for judicial review; and (d) requiring  public authorities to justify the necessity and proportionality of blocking, provide reasons as to why less intrusive means could not be employed and communicate the blocking request to the owner of the targeted website.

The Court also referenced an earlier judgment it had issued in the case of Ahmet Yildirim vs. Turkey,  acknowledging that content creators are not the only ones affected; website blocking interferes with the public’s right to receive information.

The Court also held that the participation of the ISP as a designated defendant was not enough in the case of court proceedings concerning blocking requests, because the ISP has no vested interest in the proceedings. Therefore, in the absence of a targeted website’s owner, blocking proceedings in court would lose their adversarial nature, and would not provide a forum for interested parties to be heard.

Implications for India

The online censorship regime in India is similar to Russian terms of legal procedure, but perhaps worse when it comes to  the architecture of the law’s implementation. Note that for this discussion, we will restrict ourselves to government-directed blocking and not consider court orders for content takedown (the latter may also include intellectual property infringement and defamatory content).

Section 69A of the Information Technology (IT) Act permits the Central Government to order intermediaries, including ISPs, to block online content on several grounds when it thinks it is “necessary or expedient” to do so. Amongst others, these grounds include national security, public order and prevention of cognisable offences.

In 2009, the Information Technology (Procedure and Safeguards for Blocking for Access of Information by Public) Rules, 2009 (‘blocking rules’) were issued under the Act. They lay out an entirely executive-driven process: a committee (consisting entirely of secretaries from various Ministries) examines blocking requests from various government departments, and finally orders intermediaries to block such content.

As per Rule 8, the chairperson of this committee is required to “make all reasonable efforts identify the person or intermediary who has hosted the information” (emphasis ours) and send them a notice and give them an opportunity for a hearing. A plain reading suggests that the content creator can then not be involved in the blocking proceedings. Even this safeguard can be circumvented in “emergency” situations as described in Rule 9, under which blocking orders can be issued immediately. The rules ask for such orders to be examined by the committee in the next two days, where they can decide to continue or rescind the block.

The rules also task a separate committee, appointed under the Telegraph Act, to meet every two months to review all blocking orders. Pertinently, only ministerial secretaries comprise that committee as well.

These are the limited safeguards prescribed in the rules. Public accountability in the law is further severely limited by a requirement of strict confidentiality (Rule 16) of blocking orders. With no judicial, parliamentary or public oversight, it is easy to see how online censorship in India operates in complete secrecy, making it susceptible to wide abuse.

When the constitutionality of provision and the blocking rules was challenged in Shreya Singhal v. Union of India, the Supreme Court was satisfied with these minimal safeguards. However, it saved the rules only because of two reasons. First, it noted that an opportunity of a hearing is given “to the originator and intermediary” (emphasis ours: notice how this is different from the ‘or’ in the blocking rules). It also specifically noted that the law required reasoned orders that could be challenged through writ petitions.

On this blog, Gautam Bhatia has earlier argued that the judgment then should be read as obligating the government to mandatorily notify the content creator before issuing blocking orders. Unfortunately, the reality of the implementation of the law has not lived up to this optimism. While intermediaries (ISPs when it comes to website blocking) may be getting a chance to respond, content creators are also almost never given a hearing. As we saw in the European Court’s judgment, ISPs do not have any incentive to challenge the government’s directions.

Additionally, although the law states that “reasons [for blocking content are] to be recorded in writing”, leaked blocking orders suggest that even ISPs are not given this information. Apart from the opacity around the rationale for blocking, RTI requests to uncover even the list of blocked websites have been repeatedly rejected (for comparison, Roskomnadzor at least maintains a public registry of websites blocked in Russia). This lack of transparency and fair proceedings also means that entire websites may be getting blocked when there are only specific web pages on that website that serve content related to unlawful acts.

When it comes to the technical methods of blocking, the rules are silent, leaving this decision to the ISPs. While a recent study by the Centre for Internet and Society showed that popular ISPs are using methods that target specific websites, there are some recent reports that suggest ISPs may be blocking IP addresses too. The latter can have the effect of blocking access to other websites that are hosted on the same address.

There are two challenges to the rules in the Delhi High Court, serving as opportunities for reform of website blocking and content takedown in India. The first was filed in December 2019 by Tanul Thakur, whose website (a satirical take on the practice of dowry) was blocked without any notice or hearing. Tanul Thakur was not reached out to by the committee responsible for passing blocking orders despite the fact that Thakur has publicly claimed its ownership multiple times, and has been interviewed by the media about the website. When Thakur filed a RTI asking why was blocked, the Ministry of Electronics cited the confidentiality rule to refuse sharing such information!

This month, an American company providing mobile notifications services, One Signal Inc., has alleged that ISPs are blocking its IP address, and petitioned the court to set aside any government order to that effect because they did not receive a hearing. Interestingly, the IP address belongs to a popular hosting service provider, which serves multiple websites. Considering this fact and the lack of transparency in blocking orders, one may question whether One Signal was the intended target at all! The European Court’s judgment in Vladimir Kharitonov is quite relevant here: ISPs should not be blocking IP addresses that are shared amongst multiple websites, because such a measure can cause collateral damage, and make other legitimate expression inaccessible.

Given the broad similarities between the Indian and Russian website blocking regimes, the four judgements by the European Court of Human Rights will be instructive to the Delhi High Court. Note that section 69A is used for content takedown in general, i.e. censoring posts on Twitter,  not just blocking websites): the right to hearing must extend to all such content creators. The principles applied by the European Court can thus provide for a more rights respecting foundation for content blocking in India for the judiciary to uphold, or for the legislature to amend.

Notes from a Foreign Field: The South African Constitutional Court on the Freedom of Speech [Guest Post]

[This is a guest post by Sarada Mahesh.]

The Constitutional Court of South Africa, on 27 November, 2020, delivered a landmark judgement that gave precedence to the right to freedom of speech (Economic Freedom Fighters v Minister of Justice).   

Facts of the Case

Charges were levied against Mr. Julius Sello Malema, the president of a political party known as the Economic Freedom Fighters. The judgement highlighted three instances where the accused gave speeches where he allegedly made statements that incited citizens to occupy land that the State had failed to give them. He asked them to do so unapologetically, and without the fear of going to jail. The Prosecution charged him with Section 18 (2) (b) of the Riotous Assemblies Act, 1956, along with section 1(1) of the Trespass Act, 1959.

For the purposes of this article, we will focus on one of the two questions that the Court was required to answer – on the constitutionality of section 18(2)(b) of the Riotous Assemblies Act, 1956. The section reads as follows:

(2) Any person who (a) conspires with any other person to aid or procure the commission of or to commit; or (b) incites, instigates, commands, or procures any other person to commit, any offence, whether at common law or against a statute or statutory regulation, shall be guilty of an offence and liable on conviction to the punishment to which a person convicted of actually committing that offence would be liable.

History of the Riotous Assemblies Act, 1956

Before going into the reasoning provided by the Court for both the questions, it is important to understand the history of the Act. The Act came into existence during the apartheid regime as a response to the landmark adoption of the Freedom Charter in 1955. The Defiance Campaign of 1952 was at its peak at that point with South Africans actively taking part in civil disobedience movements against the reigning power. The Riotous Assemblies Act was a re-enactment of the previous statute. Over the years, however, many provisions of this archaic law were repealed. Now, all that is left of the Act are its Preamble and sections 16, 17 and 18.  

The Preamble of the Act, which the Court unanimously declared to be ‘symbolic of the iniquitous apartheid regime and utterly indefensible in (the) constitutional dispensation’, reads as follows:

To consolidate the laws relating to riotous assemblies and the prohibition of the engendering of feelings of hostility between the European and non-European inhabitants of the Union and matters incidental thereto, and the laws relating to certain offences.

It was argued by the applicants that the section generally, and more particularly the term ‘any offence’ was constitutionally invalid because it constituted an unreasonable restriction of the fundamental right to free expression.

Decision by the High Court

The High Court declared the section to be unconstitutional and invalid on two grounds. First, the section actively criminalised conduct that was protected under section 16 (1) of the Constitution. Section 16 (1) lays down the forms of expression that are protected by the Constitution, while section 16 (2) provides for the forms that fall outside this scope (propaganda for war, incitement of imminent violence, or certain kinds of advocacy of hatred).

While the objective of the section is similar to Article 19 of the Indian Constitution, the structure is different. Article 19 (1) (a) to (g) provide for the protection of certain rights regarding freedom of speech. Article 19 (2) to (6) provide for the reasonable restrictions on these rights. Section 16 (1) and (2) however are two different circles altogether. They provide for the types of expressions that can and cannot be protected under the Constitution. The equivalent of article 19 (2) is section 36 (1) of the Constitution. It lays down five tests that have to be fulfilled to justify limitations that are imposed on the right: the nature of the right, the importance of the purpose of the limitation, the nature and extent of the limitation, the relation between the limitation and its purpose, and whether there exist less restrictive means to achieve the purpose.

Any limitation on the right of Article 16(1) must be tested against Article 36(1). The right to freedom of expression however does not extend to the speech under Section 16(2) – there is thus no question of imposing any limitation on these provisions. In the Indian context, it must also be noted that there tests are similar to the test laid down in the Puttuswamy I judgement with regard to restrictions on the right to privacy.

Secondly, the High Court highlighted that the objective of section 18 (2) (b) was to prevent crimes. However, it found the rationale of making the inciter compulsorily liable to the same extent as the actual perpetrator of the violence to be unreasonable.

Decision by the Constitutional Court (written by the Mogoeng CJ)

The aggrieved parties appealed to the Supreme Court, which did not confirm the decision of the High Court as it had made an incorrect interpretation. The reason was that the word ‘liable’ did not always mean ‘compulsory’. In fact, there had been cases where the inciter of the violence was given the same treatment as the actual perpetrator of the violence when it came to imposing sanctions.

However, it was reasoned that the acts have to be criminalised only if the pose a ‘real and substantial risk’ to the constitutional values. The Court also discussed the de minimus rule. It originates from the Latin maxim de minimis non curat lex which translates to the ‘law is not concerned with small things’. The rule does not apply to all ‘minor cases’ and is dependant on the discretion of the prosecutor, who can choose not to investigate into trivial matters if it is overburdened. The Court held that the exercise of the right to free speech should not depend on the likelihood of eventually being let off on the basis of this rule. Instead, the limitations on free speech could be recalibrated so as to respect the right.

The Court also considered the possibility of imposing lenient or appropriate sanctions in criminal cases keeping in mind the number of years it takes for cases to get resolved and the costs involved in litigation. The words ‘any offence’ failed to satisfy both these tests, and was in fact vague, uncertain and open to interpretation, thus posing a risk to the right of free expression. It was inconsistent with Section 16 (1) of the Constitution, and failed to fulfil the tests under Article 36 (1). The Court cautioned that the overbreadth of the legitimate purpose of the Government (crime prevention, in this case) should not be confused with the possibility of using less restrictive means of enquiry. Allowing a broad interpretation of the term ‘any offence’ could be applicable to any crime and have a negative impact on the constitutional values of free speech.

As an interim solution, the Court suggested that the term ‘serious’ be inserted in between the term ‘any’ and ‘offence’. It could not define the term ‘serious’ as it was pressed for time, but trusted that the judicial system would be able to define it on a case by case basis. Serious crimes include “murder, rape, armed robbery, human trafficking and corruption”, all of which are deemed mala fide.

Dissent by Maijedt J. (Jafta J. and Tshiqi J. concurring)  

The three judges dissented on the following grounds:

First, they did not agree with the full judgement’s comparison of the idea of free speech in South Africa with that of the United States. The latter gave primary importance to its First Amendment – it was always given precedence by the Courts. South Africa however, had a different jurisprudence that was closer to countries like Europe and Australia. Judgements from these countries showed that factors such as public health and morality were given due consideration in questions relating to free speech.

Secondly, in relation to the point 1, the Courts in the jurisdictions considered – Europe and Australia – made punishable acts of incitement. This, they did, without resorting to whether the offence was serious or not. The dissenting judgement argued that the word ‘serious’ was vague and could be misused. Instead, prosecutorial decisions on what constituted ‘any offence’ must be trusted. This however, could be an issue, because the decision on the definition would be subject to the prosecutor in power at that point of time. “Incitement of imminent violence” which falls under section 16 (2) is a constitutionally unprotected speech. The provision does not use the word ‘serious’, hence making it incorrect for the Full Judgement to read in the word. Additionally, there was no recorded case of misuse of this provision in the past. (which is a weak argument that shouldn’t be considered).

Thirdly, while the inciting of violence might not show any demonstrable harm, it nevertheless causes unreasonable harm to society. The inciter had the intention to commit the crime and even took steps to realise it. The Courts must not wait for the crime to actually happen in order to punish it.

This is a reasonable argument, provided that the burden is on the State to prove the intention of the inciter beyond reasonable doubt.


While the decision of the Court to find Section 18 (2) (b) unconstitutional and invalid is important, it still left a lot of issues unclear. Going by its reasoning, a person who incites violence could only be punished if it was a ‘serious’ offence. Unlike the examples of serious crimes listed above, crimes relating to free expression are more fluid and open to interpretation.  Consider the situation in countries like India, where leaders whose speech actually incited violence against a minority community, while those who protested against the Government without any actual impact were charged with sedition. It thus becomes important for Courts to handhold free speech cases, delving deeper into the jurisprudence of different countries and providing as much clarity as possible about what can and cannot be allowed.