Guest Post: The Governor’s Pardoning Powers – Statutory or Constitutional?

[This is a guest post by Pradhyuman Singh, first posted on the Proof of Guilt blog, and cross-posted here with permission.]


On 18th May 2022, the Supreme Court ordered the release of A.G. Perarivalan, a conspirator in the assassination of former Prime Minister Rajiv Gandhi, by exercising its inherent power under Article 142 of the Constitution. Perarivalan, having been found guilty on a number of charges, the State of Tamil Nadu wished to issue a pardon in his favour. To this effect, the Council of Ministers of Tamil Nadu advised the Governor to issue a pardon. The power of the Governor of a State to issue pardons is traceable to Article 161 of the Constitution. As granting pardons is a function of the Governor, it is also pertinent to read this power/function with Article 163 of the Constitution. Article 163 requires the Governor to exercise her functions on the aid and advice of the Council of Ministers. Therefore, the Governor is mandated to act on this advice, being the nominal Executive Head of the State.

In this case, the advice of the Council was tendered on September 9, 2018 and no action was taken by the Governor. Instead, the Governor submitted the matter to the President for his consideration. The Governor used this reason as a justification for the delay in implementing the mandatory advice of the Council. The Supreme Court held that such abstinence was in violation of the Governor’s duty under the Constitution. Thus, to do complete justice, the Supreme Court itself ordered the grant of pardon in favour of Perarivalan, resulting in his release. 

Predictably there has been considerable publicity on this order of the Supreme Court. Amongst the various opinions on the order, (Retd.) Justice V. Parthiban, an erudite judge of the High Court of Madras has also expressed his criticism on many fronts. One particular criticism was that the Supreme Court did not consider the attraction of Section 435 of the Code of Criminal Procedure, 1973 (“CrPC”) to the case. This specific contention will be the focus of this post.

Sections 432-434 of the CrPC provide a statutory power to the President/Governor to remit, commute or suspend punishment of offences. Section 435 provides that if the State Government (i.e. the Governor, as per Section 3(60), General Clauses Act, 1897) wishes to exercise power under Section 432 or 433, and-

  1. The offence in question has been investigated by the Central Bureau of Investigation; or
  2. The offence involved damage to property of the Central Government; or
  3. The offence was committed by an employee of the Central Government,

– the Governor would be obligated to exercise such powers only after “consultation” with the Central Government. “Consultation” here has been interpreted to mean concurrence by the Supreme Court (Union of India vs V. Sriharan). In the facts of this case, the investigation of the offences was carried out exclusively by the Central Bureau of Investigation, thus attracting Section 435 of the CrPC.

At first sight, it may seem that such powers of the Governor under the Constitution stand on a different footing from the statutory power. In no manner can the provisions of the CrPC control the ambit of Article 161 of the Constitution (which stipulates no such requirement of consulting/concurring with the Central Government). However (Retd.) Justice Parthiban emphasises that the decision rendered by the Supreme Court in Sriharan (by a Constitution Bench nonetheless) requiring mandatory concurrence of the Central Government would then be rendered completely nugatory. All that would be required is for the Governor to claim that her power is being exercised under the Constitution and not the CrPC. The entire scheme of statutory powers of the Governor/President to remit, commute or suspend sentences would become redundant.

Further, the Supreme Court on another occasion (K.M Nanavati vs State of Bombay) has also held that Articles 72 and 161 embody the “prerogative power” of the President/Governor. Since there is no express provision in the Constitution saving these provisions from legislative interference, the English common law as to Prerogatives applies to the pardoning power under our Constitution. As a result of this, these powers may be fettered and controlled by legislation. Thus, what would otherwise be a simple matter of examining the relationship between the statute and the Constitution becomes complex. This is due to the scheme of the CrPC historically having analogous powers of remission, commutation and suspension of sentences ever since its recognition in the Code of Criminal Procedure, 1898 (the predecessor to the present CrPC).

This post will enquire into the context with which parallel powers of remission, commutation and suspension exist in our statutory framework, despite our Constitution exhaustively providing for the same. Accordingly, we may determine the relationship between these laws and the legal implications that follow.

History of the Code of Criminal Procedure

The Code of Criminal Procedure, 1898 (“Code”) first recognised the power of the Governor-General of British India to remit, commute and suspend punishments of offences. This power found expression in the form of Sections 401 and 402 of the said Code. At the time, there existed no grundnorm resembling the Constitution, and so the power of the Governor General was exclusively governed by the aforementioned provisions of the Code.

This changed with the eventual enactment of the Government of India Act, 1935 with Section 295. Section 295 was the template for the drafting of Article 72 and 161 of the Constitution, which vested the power to suspend, remit or commute a sentence with the Governor-General as well. The phraseology adopted by Section 295 is as follows-

“Where any person has been sentenced to death in a Province, the Governor-General in his discretion shall have all such powers of suspension, remission or commutation of sentence as were vested in the Governor-General in Council immediately before the commencement of Part III of this Act…[Emphasis Supplied]

Thus, the Government of India Act, 1935, a legislation passed in the British Parliament, gave deference to ordinary laws to regulate the powers of the Governor General in this respect. The intent behind such a provision in the Government of India Act, 1935 was to merely give passive recognition to the power of the Governor General (which was to continue to operate through the statute, i.e. The Code of Criminal Procedure, 1898) and to make clear that such powers would not interfere with the King’s power to issue pardons or remit, commute or suspend sentences (as provided in Section 295(2)). In this manner, the statutory framework of the Code had a functional purpose in recognising the power of the Governor-General. The provisions of the Code worked in perfect harmony with the Government of India Act, 1935.

Such was the arrangement until the enactment of our Constitution. Section 295 of the erstwhile Government of India Act, 1935 was significantly changed to give us what we have as Articles 72 and 161 today. As is clear from the language of these Articles, the power of the President/Governor to remit, commute or suspend sentences was expressly recognised in the Constitution itself. Marking a clear departure from the previous position, there was no deference given to any statutory framework that would determine the powers of the President/Governor. However, the makers of the Constitution chose to still passively recognise the statutory powers of the Governor to remit, commute or suspend sentences. This was in the form of Article 72(3) which provides-

“Nothing in sub-clause (c) of clause (1) shall affect the power to suspend, remit or commute a sentence of death exercisable by the Governor of a State under any law for the time being in force.” [Emphasis Supplied]

The purpose of this clause is to clarify that the power of the President conferred by Article 72(1)(c) should not be construed to exclude the power of the Governor in any manner. However, in stating so, the makers chose to express that the power of the Governor under “any law” was not to be affected. It was equally open for the draftspersons to make a direct reference to Article 161 of the Constitution in this provision. Rather, a conscious decision was made to also recognise any statutory powers vested in the Governor. The implications of this will be examined in the following section of the post.

Moving on, once the Constitution was brought into force, the provisions of the Code were also amended. Article 72(1)(c) of the Constitution recognised the power of the President to remit, commute or suspend the sentence of death. Ordinarily, the President exercises powers with respect to offences in the Union List and the Governor in the State List respectively. An exception is made in the case of death where both the President and Governor exercise concurrent powers. To accommodate this concurrent power, Section 402A was introduced in the Code, vide an amendment. This Section too recognised the concurrent powers of the President and Governor in line with the Constitution.

The next important milestone in the development of law on this issue came in the year 1969. The Law Commission of India was tasked to review the entirety of the Code and recommend comprehensive changes to the law of criminal procedure. This took shape in the form of the 41st Law Commission Report published in September 1969. This report examined Sections 401 and 402 of the Code. It expressly noted how the statutory powers of the Government were ancillary to the Constitutional powers (Refer Page 248, Para 29.1). Interestingly, it also opined that it would be legally impermissible for the Code to be inconsistent with the Constitution (Refer Page 249, Para 29.4). However despite these findings, they expressed a concern with the State Government’s power to remit, commute and suspend sentences. It was felt that certain offences significantly affected the interest of the Central Government (offences involving the employees and property of the Central Government and offences investigated by the CBI). If a free hand was given to the State Government to exercise powers in respect of such offences, it would result in “difficulties of administration” for the Central Government, according to the Report (Refer Page 252, Para 29.13).

This formed the basis for them to recommend the insertion of a draft Section 402B in the Code. This Section would require the State Government to consult the Central Government if it wished to remit, commute or suspend sentences in the offences described above. This recommendation was accepted by the Parliament when the Code was replaced with new CrPC enacted in the year 1973 in the form of Section 435 of the CrPC.

Having a brief idea of the context of this provision, we may now examine the relationship between the CrPC and the Constitution.

Relationship Between the CrPC and the Constitution

The Constitution’s passive recognition of the statutory power of the Governor may have one of two implications-

  1. The Statutory Power is merely a formal recognition of the power of a Constitutional authority. It stands on a different footing from the Sovereign power of the Executive Head. As a consequence, it does not dilute or affect the Constitutional powers of the Governor in any manner. Any statutory provision that would be inconsistent with the Constitution would be void.
  2. The Constitutional power is an expression of the Prerogative power of the Executive. Finding its origin in English Common Law, Article 72 and 161 may still be controlled and subject to legislative provisions that may be made to this effect. (This was the reading adopted by the majority in Nanavati)

The first of these two readings would result in the statutory powers of the Governor being completely redundant. If it would be open to the Governor to exercise the exact same powers to remit, commute or suspend sentences in its Constitutional capacity. Any restrictions or mandatory procedures imposed in the statute (such as Section 435, CrPC) may be bypassed. Despite such redundancy, reading “A” would still be a more suitable reading of the law.

This is because, while it is correct that historically, the power of remission, commutation and suspension of sentences lies with the Executive Head, and that these are Prerogative powers that identify the source of authority from English Common Law – however, these conventions have been crystallised in the form of express Constitutional provisions. There are various other examples within the Constitution where principles of common law find express mention. For instance, Article 129 of the Constitutions declares the Supreme Court as a Court of Record. The powers of a Court of Record in English Common Law would include the power to punish for contempt of itself. Despite this, the framers chose to expressly also recognise the Supreme Courts power to punish for contempt in Article 129.

The consequence of this express recognition of common law norms is that such principles get the status of Constitutional law. It would follow that such principles/law are granted the protection of the highest law of the land which could only be altered by a Constitutional amendment. Therefore, Articles 72 and 161 are by themselves sources of power for the President/Governor to exercise power. It would not be correct to state that they are a reflection of the Prerogative powers which find their authority outside the Constitution.

Secondly, Article 245 of the Constitution confers power on the Legislature to make laws. The provision begins with a subject clause stating, “Subject to the provisions of this Constitution…”. This makes it abundantly clear that any ordinary law cannot be in contravention of any Constitutional provision. Further, there is nothing in the context of Articles 72 or 161 that would suggest that this straightforward interpretation of Article 245 should not be followed in reading them.

Thirdly. the language of Articles 72 and 161 marks a clear departure from their predecessor in the form of Section 295 of the Government of India Act, 1935. The regime before the Constitution in express terms gave deference to the statutory regime of the Code, which dictated the extent of powers exercisable by the Governor-General. This position no longer stands after independence, where the prerogative power has been given Constitutional status. If ordinary legislation may be allowed to control Articles 72 and 161, it would result in the said provisions themselves becoming redundant.

As mentioned earlier, Article 74/163 requires the President/Governor to exercise their functions only on the binding aid and advice of the Council of Ministers. The powers under Article 72/161 being one such function, would also have to be performed only on the aid and advice of the Council.

If reading “B” of the law is adopted and ordinary laws could independently guide the power of the President/Governor, it may not be necessary to act on the aid and advice of the Council of Ministers. The statutory framework under which the President/Governor would function, would then allow them to act in their personal discretion, free from their Constitutional obligations. This would be in clear violation of the mandate of Articles 74 and 163 of the Constitution which are a reflection of the principles of Executive accountability. The violation of these provisions in this manner would be contrary to the role of the President/Governor envisaged by the framers as nominal Executive Heads of State. 

For these reasons, the decision of the Supreme Court in Nanavati is incorrect. The passive recognition of the Governors statutory powers in Article 72(1)(c) is only indicative of a parallel legal framework within which the Governor may operate. It cannot be construed to mean that it would be open to the Legislature to dilute and control Articles 72 and 161 by an ordinary legislation. Indeed, at the time of commencement of the Constitution, the Parliament made efforts to ensure that no provisions in the Code conflicted with the Constitution. To this effect, Section 402A recognising concurrent powers of the President and Governor to remit, commute or suspend sentences of death was introduced.

Despite the view of the Court in Nanavati, the Supreme Court has subsequently moved away from this approach. In the case of Maru Ram vs Union of India, a Constitution Bench considered the validity of Section 433A of the CrPC, 1973. This provision disallowed the President/Governor to exercise powers under Section 432 and 433 (for convicts sentenced to a certain class of sentences) if the convict had not served at least fourteen years of imprisonment. The Court held that Section 433A would not be attracted when the Governor exercised powers under Article 161. Both provisions stood on a different footing and it would not be open for a statute to control the Executive Power conferred by the Constitution in absolute and unqualified terms.

This position was reiterated by the Supreme Court in the case of State of Punjab vs Joginder Singh, which involved the interpretation of paragraphs 516-B and 631 of the Manual for the Superintendence and Management of Jails in Punjab. These provisions required Jail Superintendents to submit applications for remission to State Governments after the convict served fourteen years of imprisonment. In the course of interpretation, the Court enquired into the relationship between the said Rules and Section 433A of the CrPC. By relying on Maru Ram, it too came to the conclusion that no legislative scheme (in this case, the Prison Manual) may be inconsistent with the Governors powers under Article 161.

Accordingly, Section 435 of the CrPC too cannot be construed such that it controls the absolute powers of the Governor under Article 161. The Governor would not be required to concur (or even consult) the Central Government for the purposes of issuing any pardons in furtherance of Article 161. In this light, the Supreme Court was correct to not consider the involvement of any provisions of the CrPC in the case of A.G Perarivalan.

The ConCast: Episode 2 (27 June 2022)

In this second episode of the ConCast, Abhinav Sekhri and I look back at the month that was: judgments on the right of the accused to a lawyer, voting rights, the Supreme Court’s judgment on the Gujarat riots, and judicial power.

The podcast is embedded below. It is also available on Podbean, Apple Podcasts, Google Podcasts, and (soon!) Spotify.

The ConCast: Episode 2 (27 June 2022) The ConCast

A podcast about the Indian Constitution, the Supreme Court, and beyond, hosted by Gautam Bhatia and the Indian Constitutional Law and Philosophy Blog. Episode 2, feat. Abhinav Sekhri, on 27 June 2022: the right of an accused person to a lawyer, voting rights, the Supreme Court's latest judgment on the 2002 Gujarat riots, and judicial power. 
  1. The ConCast: Episode 2 (27 June 2022)
  2. The ConCast: Episode 1 (27 May 2022)

Readings discussed:

Directorate of Enforcement vs Satyendar Kumar Jain.

Poolpandi vs Superintendent, Central Excise.

Indo-Steam China Navigation vs Jasjit Singh.

Anukul Chandra Pradhan vs Union of India.

Zakia Jafri vs State of Gujarat.

Mirjan Damaska, The Faces of Justice and State Authority.

Responding to Illegal Home Demolitions: The Doctrine of An Unconstitutional State of Affairs

Previously on this blog, we have discussed the recent spate of home demolitions that have been carried out at the behest of various state governments (see here and here). These home demolitions follow a familiar pattern. A protest takes place in a locality or neighbourhood, which turns violent. Soon after, the police declares that a certain individual, or set of individuals, have been identified as the “masterminds” behind the violence. Immediately after that, the municipality declares that these individuals are residing in unauthorised buildings (often – as in the most recent case – with backdated notices of doubtful authenticity). The buildings (homes) are then demolished. In the normal course of things, the time period between the police declaring that it has identified the masterminds behind the violence, the municipality declaring that the buildings are illegal, and the actual demolition, is under twenty-four hours.

On this blog, it has been pointed out that on its own terms – that is, even assuming that the rationale for the demolitions is illegal constructions, and not collective punishment – this modus operandi violates both local and municipal laws, the rule of law, and the Constitution. What, then, is the remedy? As the case around demolitions at Jahangirpuri in New Delhi showed, the speed at which demolitions are undertaken ensures that even where there is judicial intervention, it is often too late to accomplish anything meaningful. Things are even worse when demolitions happen far away from Delhi, or in places – and to communities – where immediate access to courts is substantially more difficult.

This post proceeds upon the important assumption that, at present, the Supreme Court has the will and the desire to address the serious challenge of the executive flouting the rule of law through the pattern of home demolitions across the country. If that assumption is true, then the question is – given the facts laid out in the above paragraph – how the constitutional violation ought to be framed, and how a remedy may be crafted.

I suggest that at present, our constitutional jurisprudence may not have the precise vocabulary to address the issue: as the Jahangirpuri demolitions showed, the Court was minded to treat the demolition as an individual State act, which it would scrutinise for compliance with the law and the Constitution. As the copycat actions across different states have shown, however, a specific instance of home demolition is not an individualised act, but is part of an evolving pattern of collective punishment by the State. To capture this, I suggest, new vocabulary might be needed. One place where this can be found is the Latin American doctrine of an unconstitutional state of affairs.

The doctrine of an unconstitutional state of affairs originated in Colombia, and was later adopted in Brazil. As the term suggests, an unconstitutional state of affairs is specifically meant for a situation where the violation of rights is not individualised, but structural. According to a definition:

The unconstitutional state of affairs is a legal ruling that allows the Constitutional Court to acknowledge the failure of both the Legislative and Executive branches of government to enforce public policies against widespread and systemic violation of fundamental rights, thus justifying a judicial intervention in order to combat the structural causes of the violations and to put everything back in order with the Constitution. 

The importance of the unconstitutional state of affairs doctrine, therefore, lies in its focus upon the widespread and systematic violation of fundamental rights. In the present situation, not only are the criteria for the application of the doctrine met, but it is the only truly effective doctrinal tool that the Supreme Court can apply to deal with the situation. There are two reasons for this. The first is that as long as the Court considers individualised cases of demolitions, it needs to take at face value the entirely implausible assertion that the police action in “identifying” the alleged rioters is separate from the municipality’s action in demolishing the home. In true Orwellian fashion, the Court has to ignore the “evidence of its own eyes and ears” – i.e., the chronology of events that is police identification – municipality declaration – demolition, all within the space of a day. This, in turn, means that the Court is forced to examine what is essentially a case of collective punishment, within the framework of municipality laws.

The unconstitutional state of affairs doctrine, however, expands what the Court can consider, both in time and in space. It allows the Court to take into account the “systematic nature” of this practice, both in the recent past, and in its spread across the country. This is crucial, because it is only from the pattern of home demolitions that the Court can extrapolate the fact that collective punishment has become an informal part of State policy. After all, it might plausibly be argued that a single occasion happens to be a coincidence; when, however, the same thing – police identification, municipality declaration, demolition – repeats itself over time and space, it becomes evident that it is State policy.

Now, once an unconstitutional state of affairs has been identified (on the basis, of course, of evidence collected over time and space), what remedy follows? Here, we are back in familiar territory: the Brazilian and Colombian Courts have developed the remedy of a structural injunction, or – as we know it in India – the continuing mandamus. The continuing mandamus allows the Supreme Court to take cognisance of the situation, issue interim orders, and monitor for compliance – which, crucially, will not be limited to single cases, but will extend to the unconstitutional state of affairs at large. What might those interim orders look like in this situation? That, I think, would depend upon how lawyers and justices might want to craft the relief, but at a pinch – for example – mandatory judicial sanction before demolition as an interim measure is one possibility; the crucial thing to note, however, is that to be effective, the remedy must be: (a) preventive, and (b) extend to all cases. Under present circumstances, where demolition cases are being treated on an individual basis, neither is possible. The unconstitutional state of affairs doctrine, however, provides the Supreme Court with a vocabulary to do both.

Guest Post: The Delhi High Court’s Interim Order on the Right to Counsel

[This is a guest post by Abhinav Sekhri, cross-posted from the Proof of Guilt blog.]

It appears that by an interim order passed on 03.06.2022, the Delhi High Court has stayed operation of a portion of an order passed by the Trial Court whereby it had directed that Satyendar Kumar Jain, who has been arrested by officers of the Enforcement Directorate and remanded to their custody for ten days, was entitled to have counsel present within visual range while he was being interrogated. The reasoning adopted by the High Court to grant interim relief is straightforward: the directions in the Trial Court’s order are seemingly contrary to judgments of the Supreme Court and the Delhi High Court, and so it was appropriate to stay their operation. 

The High Court was swayed by the decision of the Supreme Court in Poolpandi & Ors. v. Superintendent, Central Excise & Ors. [(1992) 3 SCC 259 (“Poolpandi“)] and an order of a Division Bench of the Delhi High Court in Sandeep Jain v. Addl. Director, DRI [Rev. Petition 387 of 2019 (Order dated 10.12.2019) (“Sandeep Jain“)], pressed into service by the Enforcement Directorate. All the Supreme Court judgments relied upon by the Respondent Satyendar Jain were distinguished simply because they, apparently, do not consider Poolpandi; the Bombay High Court judgment cited by his counsel was not relied upon due to the existence of Sandeep Jain which according to the High Court held a contrary view.

This admirably straightforward argument, which the High Court has accepted, has one small problem — it is wrong in law. A careful reading of the two main planks of the Enforcement Directorate’s contentions, viz. the judgments in Poolpandi and Sandeep Jain, shows that they were entirely inapplicable to the facts before the High Court in this case of Satyendar Jain. Importing Poolpandi and Sandeep Jain to this factual matrix is worse than comparing apples to oranges.

In Poolpandi the pure issue presented to the Supreme Court was this: do the persons summoned for being questioned under the Customs Act 1962 or the erstwhile Foreign Exchange Regulation Act 1947 have a right to presence of counsel while they are questioned. It was contended that the source for this right was Article 20(3) which protected accused persons against compelled self-incrimination, and alternatively in Article 21 as the questioning by agencies was necessarily coercive. The Supreme Court disagreed, primarily on that (incorrect) technical objection that persons being questioned under Customs Act etc. were not ‘accused’ thereby failing to trigger the Article 20(3) guarantee. As for Article 21, the Court refused to accept a presumption that questioning by agencies was necessarily coercive.

In Sandeep Jain, a review petition was filed challenging the dismissal of a writ petition by a single judge on entirely new grounds. This was, patently, outside the scope of review jurisdiction of the High Court, and the Division Bench rightly observed that the petition deserved an outright dismissal. In light of this, whatever followed was obiter dicta, and that is what has been treated as binding by a Single Judge in Satyendar Jain’s case. Be that as it may, what were the facts in Sandeep Jain? Again, it did not concern an arrested person; it involved a person who had avoided summons under Section 50 of the Prevention of Money Laundering Act, but now sought presence of counsel, within visual range, during his questioning. The Court declined to grant this indulgence, citing that no allegation of any threat existed justifying presence of counsel. The contrast with a case in which the person is arrested and remanded to ten days custody of the law enforcement agency could not be starker.

Critically, the issues before the Supreme Court in Poolpandi and the High Court in Sandeep Jain did notextend to considering whether persons who have been arrested and are undergoing custodial detention are also without any right to counsel during questioning. The scenario was purely about a set of people who had only been summoned, which is why none of the petitioners resorted to Article 22(1) of the Constitution which, plainly, guarantees to all persons who have been arrested the right to be defended by counsel of their choice. Since 2008, with the enactment of Section 41-D of the Criminal Procedure Code 1973, this aspect of Article 22 rights have been given clear and unambiguous statutory footing. Nothing in the Prevention of Money Laundering Act 2002 prevents the application of Section 41-D, and in any event, nothing in a statute can ever override the constitutional mandate of Article 22. 

The position of a person at liberty cannot be compared with a person under arrest. In doing so, and by ignoring the letter and spirit of Article 22(1), the High Court has unfortunately erred and has potentially set a rather dangerous precedent, perilously tilting an already skewed balance between rights of arrested persons and the powers of investigators even more in favour of the latter. There is more which can be argued to contest the legal soundness of the view taken by the High Court here, but it warrants a second look on this primary ground itself. 

The ConCast: Episode 1 (27 May 2022)

I am happy to introduce The ConCast, a podcast on the Indian Constitution, the courts, and beyond. The podcast will be available both in the form of posts on the blog, on Podbean (to download), and in due course, on the major podcast platforms.

In this opening episode, Suhrith Parthasarathy and I look back on the month of May at the Supreme Court, talking about the sedition case, the use of Article 142 in tax cases, the GST and federalism, and more. Give it a listen!

Cases discussed:

Kedar Nath Singh vs Union of India.

S.G. Vombatkere vs Union of India.

Union of India vs Mohit Minerals.

Union of India vs Ashish Agarwal.

RIT Foundation vs Union of India.

Book Review: Roberto Gargarella’s “The Law as a Conversation Among Equals”

“Our constitutional legal tradition [has] been reticent, if not unwilling, to grant the general populace a protagonist role in civic matters.” (p.5)

In Latin American Constitutionalism: 1810 – 2010, Roberto Gargarella introduced the concept of the “engine room” of the Constitution: i.e., that part of the Constitution that deals with the organisation of power. Taking the example of Latin American constitutional history, Gargarella argued that these Constitutions’ attempts at political and social transformation through expansive bills of rights and emancipatory rhetoric were often undermined because the “engine room” remained authoritarian and centralised. Bills of rights – Gargarella observed – would not accomplish much unless they were accompanied by genuine democratisation of power. Subsequently, in The Legal Foundations of Inequality, Gargarella specifically considered the relationship between the Constitution’s “engine room” – focusing on core constitutional institutions such as representative democracy, the separation of powers, and judicially enforceable bills of rights – and its engagement with political and social inequality.

The Law as a Conversation Among Equals – Gargarella’s latest book – continues this conversation. It is also, however, a more ambitious work, launching a thoroughgoing critique of many of the core concepts of contemporary constitutionalism, and arguing for a fundamental reorientation in how we understand Constitutions, their purpose, and the act of interpretation.

In The Law as a Conversation Among Equals, the main question Gargarella asks is: what is the relationship between constitutionalism and democracy? This requires, at the outset, a working definition of democracy. Gargarella provides a thick, substantive understanding, defining the concept of democracy as “one that puts special importance on requirements such as equality (equal status for every participant); making sure all groups affected by a decision have been heard; and a prolonged period for debate (which increases the need for information, transparency and public discussions where proposals can be mutually presented, challenged, and revised).” (pp. 13 – 14)

In this context, Gargarella’s core argument in the book is that ever since their birth in the United States in the late-18th century, right up until the present day, the constitutional institutions that we take for granted – such as representative democracy, periodic elections, the separation of powers and checks and balances, judicial review, and so on – have been collectively dedicated to creating, maintaining, and entrenching the “gap” between constitutionalism and democracy. This gap that has widened over the last two centuries, as societies have grown larger, more heterogenous, and more complex.

The first part of Gargarella’s argument is historical, and grounded within the history of the Americas, although parallels – I believe – exist everywhere. He considers the constitutional thought and writing of three figures – James Madison (US), Juan Batista Alberdi (Argentina), and Andres Bello (Venezuela/Chile), each of whom exercised a significant influence upon the process of constitutional founding in their respective countries, and whose ideas traveled far beyond their homelands. Each of these figures, Gargarella argues, was motivated by “democratic distrust”, i.e., a suspicion of the participation of the People in public affairs. This elitism, in turn, was baked into the institutions that undergirded those early Constitutions (and which continue to this day). In short:

The problem is the way we view the elitist discourse that ended up in our constitutions as somehow innocuous. More specifically, the problem is that elitist assumptions and principles, relics of the predominant public discourse among leaders more than two centuries ago, were entrenched, and still are, in our primary institutions. (p.41)

In a rich historical discussion, Gargarella traces this development through the framework of assumption – principle – institution. For example, the assumption that human nature is egotistical and self-interested (drawn from dominant theories of economic rationality at the time) was translated into the principle that “ambition must counteract ambition” in order for freedom to be secured, and that, in turn, was translated into the constitutional institution of separation of powers, where different bodies of government were designed so as to combat each other, rather than engage in dialogue; similarly, the foundation of the institution of judicial review lay in the assumption that a small body of technically-trained people were more likely to arrive at the truth than large deliberative assemblies. However, even if these assumptions are no longer uncontested, the institutional arrangements that they birthed continue to stay with us even today. As Gargarella notes:

The basic structure for the organization of power, above all, remains intact in our current constitutions. In this aspect, no meaningful changes have taken place. More specifically, there have been meaningful amendments to the two main parts of every constitution – the part regarding the organization of power and the part enumerating rights – but, since the beginning of the twentieth century, they have overwhelmingly concentrated, as we will see, on the part regarding rights. The declarations of rights have tended to expand, often much more than might have been expected (in many cases we have gone from a short list of “classic liberal rights” to ones that also include social, economic, cultural, and multicultural rights). Meanwhile, with regard to the organization of power, almost everything remains as it was: three primary branches of government that are separated by a system of checks and balances; a legislature comprised of a Senate that tends to be conservative and a lower house that tends to be relatively diverse; a judiciary that retains the “last word” on constitutional matters; an executive concentrated in a single, very powerful figure; and some form of federalism. (p. 46)

One of the most detailed discussions that Gargarella undertakes in this regard is that of the institutions of representative democracy. He notes that the assumption that large assemblies were prone to falling prey to passions, and oppressing the (at that time, propertied and wealthy) minority led the framers of the US Constitution to reject various forms of direct democracy, and support representative democracy not as a necessary evil that was required to administer large territories, but as the normatively better form of government. Not only this, the framers systematically worked towards eliminating external, popular controls over representatives (such as, for example, a right to instruct and a right to recall), in favour of internal controls (i.e., different branches of government checking and controlling each other):

The counter-majoritarian bias of Madison’s characterization, in my view, is wholly evident in this perspective of the representative system, a perspective that permeates the entire US Constitution. The Constitution was devised to limit citizen participation in politics, to “separate” citizens from their representatives, to prevent them from imposing their opinions on legislators, to make sure electors cannot force their representatives to favor the particular interests of their locality over the general interests of the nation, and so on. From this characterization and line of reasoning began counter-majoritarian constitutionalism. (p. 65)

As a result, various forms of popular engagement with law-making – including, for example, the town hall meeting – were eliminated:

In general terms, the constitutional model that – starting with the 1787 United States Constitution – began to spread to other countries until it became the prevailing model, places strong emphasis on “internal” over “external” controls; presents a certain bias against democratic participation; and contains many restrictions on the power of the majority. These traits developed, in a very special way, in dialogue with and in reaction against an original, alternative model, which seemed to insist on values in tension with those that became dominant (constitutions that placed great importance on majority rule, democratic participation, and “popular” controls) (p. 107).

And, in turn, the sole method of popular control over lawmakers was the periodic election – and this, in turn, placed a burden of expectation upon the vote that – according to Gargarella – it simply could not fulfil.

Through this analysis, we can therefore see how a range of constitutional institutions that we take for granted today – i.e., representative legislatures, periodic elections, internal checks and balances, and so on – were designed with the specific purpose of denying – in Gargarella’s words – “the general populace a protagonist role in civic matters.”

Gargarella thus argues that the exclusion of the People is baked into the design of modern constitutionalism, a design that is at odds with his own vision of substantive democracy, and the role that law must play in facilitating “a conversation between equals.” However, what would a constitutionalism founded upon a “conversation between equals” look like? In the second half of the book, Gargarella develops this in some detail. The “idea of a conversation between equals” is quite close to various versions of dialogic democracy, which envisage a process of collective decision-making through a structured dialogue between parties in positions of equality. For Gargarella, this involves reconsidering several core elements of constitutionalism. Consider, for example, judicial review. In a previous chapter, Gargarella critiques the Marbury v Madison version of judicial review as born from an assumption that the best way of protecting rights is by giving groups of technically-trained judges the “final word” over constitutional interpretation. Under the “conversation among equals” model, however, this is no longer quite so obvious. Instead, borrowing from – and deepening – John Hart Ely’s idea of the court playing a “representation-reinforcing” role, Gargarella argues that the purpose of judicial review should be to deepen democracy:

In my view, and guided by a deliberative conception of democracy, judges should set aside cases that they deal with today, and at the same time, begin paying attention to other cases, which today they tend to set aside. Thus, and for example, they should start examining with strict scrutiny (I mean, with a high presumption of invalidity) those laws passed in Congress with the necessary majorities, but without public debate, or unsupported by public. reasons, or just expressing the demands of interest groups. (pp. 228 – 229)


My own approach would rather be oriented to the democratic reconstruction of an already seriously undermined scheme of controls. This would require judges, for instance, to work with a presumption contrary to the concentration of powers. Second, while Ely’s scheme aims to “keep open the channels of political change,” the one I suggest would be oriented (for said reasons) to promote the democratic debate. Third, while Ely’s scheme requires judges to provide special protection to discrete and insular minorities, my proposal would (also) urge a similar concern for the fate of impoverished majorities. In this sense, and in order to honor the values of public debate, judges could make a special contribution to public debate by providing special protection to the voices of socially marginalized groups (i.e., in cases related to social protest). (pp. 230 – 231)

Gargarella sees seeds of this approach towards constitutionalism already present in certain jurisdictions: in the Canadian Charter’s “notwithstanding” clause, for example, and in the South African Constitutional Court’s famous Grootboom decision (although Gargarella concedes that both these examples have flattered to deceive). And he finds particular succour in various citizens’ assemblies and deliberative initiatives that have sprung up in recent years, from public debates over abortion in Argentina, to the same-sex marriage referendum in Ireland, to the popular constitution-making process in Iceland. This book was written before recent developments in Chile, otherwise – no doubt – that too would have featured prominently.

The Law as a Conversation Among Equals combines a nuanced critique of constitutionalism’s “story so far”, and a powerful blueprint for a journey forward. It is also part of an ongoing conversation that acknowledges the historical exclusion of “the People” from constitutionalism, and attempts to bring them back in, in a manner that promotes the idea of a “conversation among equals” (and not simply top-down, superficial popular involvement such as formal referenda). For example, I have no doubt that Gargarella would be fascinated by the combination of direct and representative democracy that is deployed under the Kenyan Constitution’s “popular initiative” process for constitutional amendment (Article 257), and the interpretation of that provision in the famous “BBI Cases” that have previously been discussed on this blog.

One of the most urgent questions facing contemporary constitutionalism is, indeed, about how to alter and transform constitutional institutions that have, thus far, been designed to exclude the People – but how to do it in a way that does not replace one system of domination with that of another. The Law as a Conversation Between Equals is a valuable contribution to that ongoing, global conversation.

Guest Post: Fiscal Federalism and the Centralising Drift – The Supreme Court’s GST Judgment

[This is a guest post by Suhrith Parthasarathy.]

In July 2017, the Government of India heralded a goods and services tax regime by bringing forth the 101st amendment to the Constitution through an unusual midnight session of Parliament. The government claimed the new tax would unify the Indian market. Any fears that the tax would disrupt Indian federalism were sought to be eased by describing the effort as an example in cooperation between the states and the Union. The then chief economic adviser to the government of India, Arvind Subramanian, described the move as a “voluntary pooling of sovereignty”.

The phrase “pooled sovereignty” owes its origins to the creation of the European Union, where member states agreed to delegate some of their decision-making powers to the council. In the case of the GST, the idea was that both Parliament and the state legislatures would delegate some of their power to a newly formed GST Council that would help make a unified law for the nation. All along, it was believed that the Council’s decision would be binding on each of the states and that any dispute that a state wanted to raise would have to be resolved within the structures built into the system. 

This idea, seen as foundational to the functioning of the GST, has come under threat from a judgment of the Supreme Court, in Union of India v. Mohit Minerals, where a 3-judge bench, presided by Justice DY Chandrachud, has held that the GST Council’s decisions are not binding on legislative bodies and that both Parliament and the state legislatures possess plenary powers to make laws as they deem fit. How states react to this finding and to what extent they choose to amend their respective GST legislations might come to have a deep bearing on the future of Indian federalism.

In framing the Constitution, the Constituent Assembly was conscious of making careful divisions of power between the Union and the state governments. Although there were certain areas in which the Union was accorded pre-eminence, when it came to taxation, the framers were keen to vest in the States substantial responsibility. The compartments that were drawn out ensured that the powers of taxation were not mutually exclusive. Income tax (excepting tax on agricultural income) was offered to the Union, along with some indirect taxes such as customs and excise duties. State governments, on the other hand, were given the exclusive authority to tax both sale of goods and the entry of goods into a state. This division was made by inserting these subjects respectively into Lists I and II of Schedule VII of the Constitution. Critically, the concurrent list, that is List III, did not contain any taxing subject. Therefore, on a reading of Articles 245 and 246 and the entries in Lists I and II of Schedule VII, a clear division of power between the Union and the states could be gleaned out.

The 101st amendment toppled this arrangement. It removed from List II a slew of subjects over which hitherto the state government had enjoyed absolute power. These included, for example, entry 52, which was “taxes on the entry of goods into a local area for consumption, use or sale therein”; and entry 55, which dealt with taxes on advertisements. Entry 54 was substituted by a new entry, which had the effect of removing the power to tax on sale or purchase of goods excepting certain categories of products, such as fuel, natural gas, and liquor. The amendment also introduced a new provision, Article 246A, which would, as a stand-alone clause, provide a power to tax goods and services. Article 246A reads as follows:

“(1) Notwithstanding anything contained in articles 246 and 254, Parliament, and, subject to clause (2), the Legislature of every State, have power to make laws with respect to goods and services tax imposed by the Union or by such State.

(2) Parliament has exclusive power to make laws with respect to goods and services tax where the supply of goods, or of services, or both takes place in the course of inter-State trade or commerce.

Explanation.—The provisions of this article, shall, in respect of goods and services tax referred to in clause (5) of article 279A, take effect from the date recommended by the Goods and Services Tax Council.]”

As we can see, Article 246A begins with a non-obstante clause. It overrides the general power to legislate vested in Parliament and the state legislatures through Article 246. It also overrides Article 254, which deals with inconsistencies—including repugnancy—between laws made by Parliament and a state legislature. To give effect to the proposed exercise of unification the 101st amendment introduced a Goods and Services Tax Council through Article 279A. The GST Council comprises the Union Finance Minister (who shall act as a chairperson), the Union Minister of State in charge of Revenue or Finance and the Minister in charge of Finance or Taxation or any other Minister nominated by each State Government.

Article 279A(4) stipulates that the GST Council “shall make recommendations to the Union and the States on (a) The taxes, cesses and surcharges levied by the Union, the states and the local bodies which may be subsumed in the Goods and Services tax; (b) The goods and services that may be subjected to, or exempted from the GST; (c) Model GST Laws, principles of levy, apportionment of Goods and Services tax levied on supplies in the course of inter-state trade or commerce under Article 269A and the principles that govern the place of supply; (d) The threshold limit of turnover below which goods and services may be exempted from GST; (e) The rates including floor rates for specified period, to raise additional resources during any natural calamities or disaster; (f) Special provision with respect to the states of Arunachal Pradesh, Assam, Jammu and Kashmir, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim, Tripura, Uttarakhand and Himachal Pradesh (Referred as Special Category States (g) Any other matter relating to the goods and services tax, as the council may decide.” [Emphasis is mine].

Article 279A(9) states that “Every decision of the Goods and Services Tax Council shall be taken at a meeting, by a majority of not less than three-fourths of the weighted votes of the members present and voting, in accordance with the following principles, namely:— (a) the vote of the Central Government shall have a weightage of one-third of the total votes cast, and (b) the votes of all the State Governments taken together shall have a weightage of two-thirds of the total votes cast, in that meeting.” [Emphasis supplied].

Article 279A(10) and (11) read as follows:

“(10) No act or proceedings of the Goods and Services Tax Council shall be invalid merely by reason of— (a) any vacancy in, or any defect in, the constitution of the Council; or (b) any defect in the appointment of a person as a Member of the Council; or (c) any procedural irregularity of the Council not affecting the merits of the case. (11) The Goods and Services Tax Council shall establish a mechanism to adjudicate any dispute— (a) between the Government of India and one or more States; or (b) between the Government of India and any State or States on one side and one or more other States on the other side; or (c) between two or more States, arising out of the recommendations of the Council or implementation thereof.]”

The language used in Article 279A is somewhat puzzling. The provision uses the word “recommendation” to refer to the GST Council’s advice (which includes advice on a model law) but it also terms the results of the Council’s deliberations as decisions. What is more, it establishes a mechanism to adjudicate disputes that might arise between governments on any decision taken by the council. If one thinks of the GST as a unitary tax, and if the GST has to work in the manner in which it was conceived by the Union government, any advice proffered by the GST Council would necessarily have to be binding on the states—there is no question of allowing a variation in laws, as that would defeat the idea of having a single tax and a single market. This would, of course, mean that the 101st amendment will have to be seen as dismantling the nature of fiscal federalism that the Constitution in its original form established.

But perhaps because there was no other way the government’s idea could be workable, almost right from its inception, the GST Council’s decisions were viewed as binding on the states. Even when states seriously contested a piece of advice that had been offered, they nonetheless acted on such advice. And for this reason, many states also believed that the Council and its workings impinged on powers vested in the states by the Constitution. In other words, it was thought that the 101st amendment violated the federal arrangement in a manner that had the effect of effacing one of the basic features of the Constitution.       

The judgment now delivered in Mohit Minerals does not concern itself directly with any of these questions. The case arose out of an appeal filed by the Union government against a judgment of the Gujarat High Court. The High Court had declared a levy of Integrated Goods and Services Tax [IGST] imposed on importers, on ocean freight charges paid by foreign sellers to foreign shipping lines. The court had found that the importers were already paying a tax on the composite supply that was made and that to impose an additional levy on just the ocean freight would be tantamount to a form of double taxation. Moreover, in the case of CIF contracts (Cost, Insurance and Freight contracts), both the service provider and the service recipient were outside the territory of India, and the tax itself was being cast on an importer, who, in the first place, was not the recipient of the service.

During arguments in the Supreme Court, the Union government argued that this decision to levy tax on ocean freight had been made by the GST Council and that the government was, as such, bound by it. In examining this question, the Supreme Court considered the purport of Article 279A and made a series of findings, which could potentially dismantle the idea of GST, as understood through the 101st amendment. The court did not explicitly premise these findings on the Constitution’s basic structure—its analysis was largely predicated on the text of the Constitution and on what it regarded as Parliament’s intention behind the 101st amendment. But the underlying philosophy of the basic structure doctrine was nonetheless at play. The court found that to hold the GST Council’s recommendations as binding would have the effect of impinging on legislative powers granted by the Constitution, and would, in the process, alter “fiscal federalism”.

The court recognised that there were two chief arguments in favour of seeing the GST Council’s recommendations as binding. First, if one sees the recommendations as just that, the “GST will collapse as each State would then levy a conflict tax and collection mechanism”; second, if the recommendations are non-binding there would be no dispute to resolve under Article 279(11) as the States would be free to disregard the recommendations. In the same vein, the court also recognised the two chief arguments against seeing the GST Council’s recommendations as binding. First, it would violate the supremacy of Parliament and the state legislatures since both have been afforded “simultaneous” legislative power on GST; second, the virtual veto given to the Union in the GST Council would lead to a violation of fiscal federalism.

Having recognised these arguments, the court proceeded to analyse and lay down the broad contours of India’s federal structure. It cited, among others, HM Seervai who in arguing that India was a federal state, pointed to important powers that had been vested exclusively in the state governments. “The view that unimportant matters were assigned to the States cannot be sustained in face of the very important subjects assigned to the States in List II, and the same applies to taxing powers of the States which are made mutually exclusive of the taxing powers of the Union so that ordinarily the States have independent source of revenue of their own,” wrote Seervai. “The legislative entries relating to taxes in List II show that the sources of revenue available to the States are substantial and would increasingly become more substantial. In addition to the exclusive taxing powers of the States, the States become entitled either to appropriate taxes collected by the Union or to a share in the taxes collected by the Union.”

The Court then proceeded to examine the language used in Article 246A. It saw that the provision granted to both Parliament and the state legislature co-equal power. That is, the legislative bodies were given simultaneous authority to legislate on GST within their respective jurisdictions. This was, the court said, a sui generis provision, containing “unique features” of federalism. “Article 246A treats the Centre and States as equal units by conferring a simultaneous power of enacting law on GST. Article 279A in constituting the GST Council envisions that neither the Centre nor the States can act independent of the other.” The court said that it was aware that there are certain areas in the Constitution where the division of power is lopsided, where the arrangement provides for a centralising drift. But that such provisions exist cannot take away from the fact that there might be other areas where the central and state governments are given equal power. The court also noted that the Constitution, after the 101st amendment, does not provide for a mechanism to resolve any repugnancy or inconsistency between a parliamentary and state law. This would mean that the GST Council would have to strive to work in a harmonised manner, but at the same time this cannot mean that the GST Council’s decisions will override the basic legislative power vested in the state governments.

Contrary to the much vaunted idea of cooperative federalism that was seen as underlying the GST, the court held—cited scholarship by Jessica Bulman-Pozen and Heather K. Gerken—that what the regime in fact promotes is a form of “uncooperative federalism.” This contestation, Justice Chandrachud wrote, is “valuable since ‘it is desirable to have some level of friction, some amount of state contestation, some deliberation-generating froth in our democratic system.’ Therefore, the States can use various forms of contestation if they disagree with the decision of the Centre. Such forms of contestation are also within the framework of Indian federalism. The GST Council is not merely a constitutional body restricted to the indirect tax system in India but is also an important focal point to foster federalism and democracy.” If the GST Council works not merely through cooperation but also through contestation, and arrives at decisions in a democratic manner, there will be, the court believed, no need for any fears that the taxing regime will crumble as a whole.

To augment this holding the court referred to the bare language used in Article 279A. The word “recommendation”, the judgment found, is used in an array of different constitutional provisions. And the meaning ascribed to it is invariably contextual. In this case, it was impossible to see the word “recommendation” as meaning “binding recommendation” because if that was Parliament’s intention behind the 101st amendment, a qualification to that express account would have been included in Articles 246A or 279A. “Neither does Article 279A begin with a non-obstante clause nor does Article 246A provide that the legislative power is ‘subject to’ Article 279A,” the court held. But the Union also claimed that the legislatures, both Parliament and the states’, had effectively ceded their power. An analysis of many of the provisions of the Central Goods and Services Tax Act, and, for that matter, the state GST legislations, expressly stipulate that the rule-making power delegated to the Government will be exercised on the recommendations of the GST Council. As examples, the Supreme Court cited Section 5 of the IGST Act, which provides that the taxable event, taxable rate, and taxable value shall be notified by the government on the “recommendations of the Council”. Similarly, the power of the Central Government to exempt goods or services or both from levy of tax shall be exercised on the recommendations of the GST Council under Section 6 of the IGST Act. Section 22 provides that the Government may exercise its rule making power on the recommendations of the GST Council. The CGST Act also provides for similar provisions in Sections 9, 11 and 164. Apart from these, a look at the state GST legislations also showed that similar mandates were made by state legislatures. For instance, Section 9(1) of the Tamil Nadu GST Act reads as follows:  “Subject to the provisions of sub-section (2), there shall be levied a tax called the Tamil Nadu goods and services tax on all intra-State supplies of goods or services or both, except on the supply of alcoholic liquor for human consumption, on the value determined under section 15 and at such rates, not exceeding twenty per cent., as may be notified by the Government on the recommendations of the Council and collected in such manner as may be prescribed and shall be paid by the taxable person.”

A reading of this clause would indicate that the state government is bound by the advice of the Council. For that matter, this clause and each of the provisions of the State GST Act was incorporated on the basis of the model law prescribed by the Council. But what if the legislature amends Section 9(1) and allows the government to deviate from the recommendation made by the council? According to the Supreme Court, the scheme of the 101st amendment does indeed allow for such amendments to be made, because the GST Council’s recommendations can never constrain the basic legislative power prescribed in Article 246A. A reading of Paragraph 59 of the Supreme Court’s judgment is instructive:

“59. The provisions of the IGST Act and CGST Act which provide that the Union Government is to act on the recommendations of the GST Council must be interpreted with reference to the purpose of the enactment, which is to create a uniform taxation system. The GST was introduced since different States could earlier provide different tax slabs and different exemptions. The recommendations of the GST Council are made binding on the Government when it exercises its power to notify secondary legislation to give effect to the uniform taxation system. The Council under Article 279A has wide recommendatory powers on matters related to GST where it has the power to make recommendations on subject matters that fall outside the purview of the rule-making power under the provisions of the IGST and CGST Act. Merely because a few of the recommendations of the GST Council are binding on the Government under the provisions of the CGST Act and IGST Act, it cannot be argued that all of the GST Council’s recommendations are binding. As a matter of first principle, the provisions of the Constitution, which is the grundnorm of the nation, cannot be interpreted based on the provisions of a primary legislation. It is only the provisions of a primary legislation that can be interpreted with reference to the Constitution. The legislature amends the Constitution by exercising its constituent power and legislates by exercising its legislative power. The constituent power of the legislature is of a higher constitutional order as compared to its legislative power. Even if it is Parliament that has enacted laws making the recommendations of the GST Council binding on the Central Government for the purpose of notifying secondary legislations, it would not mean that all the recommendations of the Council made by virtue of its power under Article 279A have a binding force on the legislature.”

It is my submission that this finding by the Supreme Court in paragraph 59, which has a potentially far-reaching effect, is correct and laudable. Critics of the judgment may well point to the fact that in allowing the state legislatures plenary power to legislate beyond the Council’s recommendations the court has potentially allowed a pathway for a collapse of the GST regime. But the court’s interpretation is predicated on two things: one, the bare text of the Constitution, in that Article 279A uses the word “recommendation” and in that Article 246A does not limit in any manner the equal power granted both to the Parliament and the state legislatures; two, that any other interpretation would allow the federal compact, as originally conceived by the framers, to collapse. In other words, what is at stake here is the Constitution’s basic structure. Upholding that structure requires us to see both the Union and the state governments as equal partners.

The judgment in Mohit Minerals is not a comment on fiscal policy or on the merits of a unified taxing regime. If such a regime is desirable, then it is up to the Union and the state governments to arrive at a consensus through democratic deliberation, whether within the confines of the GST Council or outside of it. But if a state government believes that it must disregard some decision or the other of the Council, its basic legislative power to do so cannot be arrogated. To ascribe any other meaning to the 101st amendment would only render it a nullity.

A Question of Consent: The Delhi High Court’s Split Verdict on the Marital Rape Exception

Today, a division bench of the High Court handed down its judgment on the constitutional challenge to the marital rape exception [“the MRE”]. Put simply, the marital rape exception states that “sexual intercourse by a man with his own wife … is not rape.” Petitioners – supported by amici – argued that the marital rape exception – which, in effect, immunises married men from being prosecuted for rape – violated Articles 14, 15(1), 19(1)(a), and 21 of the Constitution. The two-judge bench delivered a split judgment: Justice Shakder struck down the MRE as unconstitutional on all of the above grounds, while Justice Hari Shankar upheld its constitutionality.

Previously, on this blog, I have analysed the constitutional issues around the MRE in some detail. In this post, I shall argue, first, that the fundamental point of difference between the two judges is on the question of consent. Justice Shakder believes that whether in a marriage or out of it, sexual consent is paramount and inviolable. Justice Hari Shankar – although he denies it from time to time – believes that within a marriage, a woman’s consent to sex carries less weight. Secondly, I shall note that under existing Indian constitutional law, Justice Shakder is correct, and Justice Hari Shankar is wrong. Consequently, when this split judgment goes for resolution before a Full Court (or to the Supreme Court), Justice Shakder’s views ought to be upheld, and the MRE struck down.

The Opinion of Shakder J

The core of Justice Shakder’s argument can be found in paragraph 135.2 of his opinion. Examining s. 375 of the Indian Penal Code in some detail, which sets out the seven circumstances under which a sexual act counts as rape, he observes that:

A close reading of the circumstances would reveal that except for the sixth circumstance (which concerns a girl-child under 18 years of age), willingness (as in the first circumstance) and consent (as in the second to fifth and seventh circumstance)- form the basis of separating acts which are lawful from those which are construed as unlawful. The circumstances are clearly agnostic to the relationship between the offender and the woman victim. [Emphasis supplied]

As Shakder J notes, therefore, the core of the offence of rape is non-consensual sexual intercourse. The MRE creates a “firewall” that protects one class of putative perpetrators – married men – from being prosecuted for this offence, even though the ingredients of the offence are exactly the same. The question then follows: is this distinction constitutional? Shakder J holds that is not, as in essence what it conveys is that “forced sex outside marriage is ‘real rape’ and the same act within marriage is anything else but rape.” (paragraph 137.1) Thus, the MRE “with one stroke deprives nearly one-half of the population of the equal protection of laws.” (paragraph 137.1) This is because:

The immediate deleterious impact of the provisions of MRE is that while an unmarried woman who is the victim of the offence of rape stands protected and/or can take succour by taking recourse to various provisions of the IPC and/ the Code, the same regime does not kick-in if the complainant is a married woman. In this context, one may have regard to the following provisions of the IPC and the Code : Section 228A of the IPC prevents disclosure of the identity of a rape victim except in certain circumstances set out therein. Likewise, Section 26 of the Code provides that the offences concerning rape/aggravated rape shall be tried as far as practicable by a court presided by a woman. Section 53A empowers a medical practitioner to examine, a person charged with committing an offence of rape if he has reasonable grounds for believing that such examination will furnish evidence with regard to the commission of the offence. (paragraph 141)

For these reasons, Shakder J holds that the MRE fails the reasonable classification test of Article 14. He then addresses two counter-arguments: the idea of a “conjugal expectation to sex” and the “preservation of the institution of marriage.” On both issues, his response is grounded in the right to individual autonomy and consent. On the first, he notes that whatever the expectation might be (i.e., “unreasonable” denial of sex counts as a ground for divorce under Indian family law), it does not extend to an “unfettered right to sex” without consent (paragraph 146); on the second, he notes that the marital bond is itself based on the idea of choice, and mutual respect for “physical and mental autonomy” (paragraph 148); once again, therefore, a legal provision predicated upon the denial of consent cannot be saved by appeals to the institution of marriage.

This focus on choice, autonomy, and equality also leads Shakder J to hold that the MRE violates Articles 21, 15(1), and 19(1)(a) of the Constitution. In paragraph 163, he holds that “modern-day marriage is a relationship of equals. The woman by entering into matrimony does not subjugate or subordinate herself to her spouse or give irrevocable consent to sexual intercourse in all circumstances. Consensual sex is at the heart of a healthy and joyful marital relationship.” For this reason, denial to married women the right to trigger prosecution for the violation of sexual consent infringes Articles 21; it also infringes Article 15(1), as it is discrimination based solely on marital status; and it infringes Article 19(1)(a), as “the guarantee of freedom of expression includes a woman’s right to assert her sexual agency and autonomy.” (paragraph 166)

The Opinion of Hari Shankar J

How does the opinion of Hari Shankar J respond to these contentions? This opinion is based on two prongs. First, Hari Shankar J identifies what he believes to be a fundamental flaw in the petitioners’ logic: i.e., that all non-consensual sex is, by default, rape, and that the MRE is an impermissible departure from this default; and secondly, that when it comes to sex, the marital relationship is distinct from all other relationships, in that it carries with it a “legitimate expectation of sex.” This – according to Hari Shankar J – provides the “intelligible differentia” under Article 14, that justifies the legislative decision of treating non-consensual sexual intercourse within marriage as “not rape.”

Let us examine both steps of the argument. On the first step, Hari Shankar J tries to drive his point home by drawing an analogy with the crime of murder. Just like not every instance of taking a life is not deemed under criminal law to be “murder”, therefore – it follows – that not every act of non-consensual sex is deemed “rape”; rather, it is the legislature that decides which kind of non-consensual act is to be deemed “rape”, just as it defines when the taking of life is deemed murder. (paragraph 103)

In this context, Hari Shankar J repeatedly – and rather intemperately – accuses petitioners’ counsel, and the amici, of making arguments devoid of logic, and attempting to substitute the legal definition of “rape” for “what they feel should be the definition of rape.” If there is anything that demonstrates a complete lack of logic, however, it is Hari Shankar J’s choice of analogy. The relationship between the MRE and the offence of rape is not equivalent to the legislature defining the circumstances under which the taking of a life amounts to murder. The correct analogy – as should be immediately evident – is that of the legislature defining the offence of murder in full detail, and then adding – for example – an “MP exception” that goes “the killing of a human being by a member of parliament is not murder.” This is because – and this is the point of Shakder J’s judgment that Hari Shankar J fails to deal with in any sense – s. 375 exhaustively defines the ingredients of the offence of rape (which – as Shakder J correctly notes – involve non-consensual sex in various forms), and then exempt a class of perpetrators from prosecution on no other ground than that they belong to that class.

It is this simple elision that thus allows Hari Shankar J to dodge the issue of consent entirely, and repeatedly insist throughout his judgment that he supports consent, and indeed – incredibly – that this case is not about consent at all. As is immediately obvious, however, this case is all about consent: the entire scheme of s. 375 is designed to define non-consensual sex as rape, and then shield married men from the consequences of that legislative design.

Hari Shankar J then notes that there are a range of provisions in the IPC where the relationship between the parties matters (in a somewhat disturbingly violent analogy, he argues that a father slapping his child is not an offence, but a stranger slapping the same child is (paragraph 134)). This brings us to the second prong of his argument, which is the intelligible differentia. Hari Shankar J argues that the intelligible differential is founded upon the “unique demographics” (paragraph 104). What are these unique demographics? This comes in paragraph 113:

Equally plain, and real, is the fact that the primary distinction, which distinguishes the relationship of wife and husband, from all other relationships of woman and man, is the carrying, with the relationship, as one of its inexorable incidents, of a legitimate expectation of sex.

This idea of a “legitimate expectation of sex” comes in repeatedly through the judgment, and is the basis of Hari Shankar J’s finding that the MRE is constitutional. In paragraph 116, he notes that marriage “is the most pristine institution of mankind”, and that the “sexual aspect is but one of the many aspects” upon which the marital bond rests; in paragraph 119, he says that “sex between a wife and a husband is, whether the petitioners seek to acknowledge it or not, sacred.” In paragraph 120, he says that “introducing, into the marital relationship, the possibility of the husband being regarded as the wife’s rapist, if he has, on one or more occasion, sex with her without her consent would, in my view, be completely antithetical to the very institution of marriage, as understood in this country, both in fact and in law.” In paragraph 127, he says that unlike live-in relationships, “the expectation of sex of the husband, with his wife is, therefore, a legitimate expectation, a healthy sexual relationship being integral to the marital bond”; in paragraph 130, he says that “any assumption that a wife, who is forced to have sex with her husband on a particular occasion when she does not want to, feels the same degree of outrage as a woman raped by a stranger, in my view, is not only unjustified, but is ex facie unrealistic.” He then adds that “it cannot even be assumed, in my view, that the perceptions of the petitioners reflect the views of the majority of Indian women.”

It is important to extract these observations in some detail, because they are characteristic of the muddled legal thinking that runs through Hari Shankar J’s opinion as a whole. Even if you take all these observations and assertions to be true (and there are many who would contest them!), what they demonstrate – at their highest – is that sex within marriage is somehow qualitatively different from sex outside marriage, because it forms an integral part of a set of reciprocal rights and obligations that constitute the valuable social institution of marriage.

But even if true, this is entirely besides the point. The only evidence that Hari Shankar J can muster up as evidence in his support is that unreasonable denial of sex can serve as grounds for divorce. That is true, as Shakder J also recognises. But there is a chasm of difference between saying – on the one hand – that the reciprocal social rights and obligations in a marriage create a ground for dissolution of that marriage if they are not discharged by either party, and saying – on the other – that they justify immunising the violation of sexual consent from being prosecuted as it normally is, outside of marriage – i.e., as rape.

Indeed, when you strip away the verbiage, what Hari Shankar J is effectively saying is that marriage not only gives the husband a legitimate expectation of sex, but the further right to violently enforce that expectation without suffering the same consequences as other people suffer. This not only flouts the rule of law, but also flouts basic logic, which appears to be particularly dear to Hari Shankar J.

A quick note on paragraph 130, which I found particularly disturbing. First, there is the assertion that a married woman who is subjected to non-consensual sex (since Hari Shankar J objects to using the word “rape”) will not feel as “outraged” as woman who is raped by a non-married person (whether that person is a stranger, a friend, or an intimate partner). This assertion has no business being in a judicial opinion. Secondly, there is the assertion that “the majority of Indian women do not share the views of the petitioners.” Whether true or not, this is entirely irrelevant, and indeed, a return of the infamous “minuscule minority” view that appeared in Koushal v Naz, was seemingly buried in Navtej Johar, but appears to have infinite lives in the halls of the Court.

It is this extraordinary reasoning that allows Hari Shankar J, to hold in paragraph 165, that:

Plainly read, it is clear that there is nothing in the impugned Exception which obligates a wife to consent to having sex with her husband, wherever he so requests. All that it says is that sexual acts by a husband with his wife are not rape. It does not even obliquely refer to consent, or want of consent.

Once again, we see the absence of logic. It is nobody’s case that the Exception itself “obligates” a wife to consent to sex at all times. The case is that the Exception devalues a wife’s consent purely by virtue of her marital status. Hari Shankar J sets up this straw-man to knock it down in the second sentence – and then, in the third sentence, he comes up with a non-sequitur, noting that not only does the MRE not force a wife into non-consensual sex, but that it has nothing to do with consent at all! It is almost trite at this stage to point out the absence of logic: when s. 375 says that non-consensual sex is rape, and the MRE says that “except where it is a married man”, what the section – read as a whole – says is that non-consensual sex between a married man and a wife is not rape. Repeatedly – and belligerently – stating that all this has nothing to do with consent does not make it true.

The intellectual dodge at the heart of the judgment is finally laid bare in paragraph 169, where Hari Shankar J notes, by way of conclusion, that:

…the legitimate conjugal expectations of the man, as the husband of the woman and the reciprocal obligations of the wife, the peculiar demographics and incidents of marriage, vis-à-vis all other relationships between man and woman, and all other legitimate considerations to which I have already referred, and which justify extending, to sexual intercourse and sexual acts within marriage a treatment different from such acts committed outside the marital sphere.

For the reasons I have explained in some detail, the dodge is simple: it is not enough for Hari Shankar J to show that sex within a marriage is in some way “different” from sex outside of marriage. He has to show that it is different in such a way that justifies diluting a married woman’s consent to sex. He does not show this, because he – incorrectly – attempts to argue that the entire case is not about consent in the first place. And the only way he can show that is by ignoring the actual text of s. 375 altogether – the text that is the starting point of Shakder J’s judgment – and which makes clear that consent is baked into the very ingredients of the offence of rape.


Having deconstructed the fundamental flaws of law – and of logic – that constitute Hari Shankar J’s opinion, it should be obvious that the opinion is unsustainable. In 2022, Indian constitutional law does not support the dilution of sexual consent based on marital status. One does not need to look too far for this: the issue is considered squarely in the Puttaswamy judgment, where Chandrachud J’s plurality opinion is explicit on this point, while many the other judgments make it clear the decisional autonomy is a fundamental facet of the right to privacy, and is not lost or in any other way compromised through social institutions such as marriage. Decisional autonomy within the marriage was also the fundamental basis upon which adultery was decriminalised in Joseph Shine; and sexual autonomy was at the core of Navtej Johar. It is – thankfully – too late in the day to go back on this rather fundamental precept.

Three final points. First, I have not in this post analysed all parts of the two opinions. For example, the two judges differ on whether striking down the MRE would lead to the creation of a new offence. I have analysed this issue in some detail in my previous post, and interested readers may refer to that.

Secondly, as this post shows, I believe that Shakder J’s judgment is opinion, and ought to be upheld on appeal. However, I also believe that the appellate forum needs to do more than that. I believe – and I say this with due consideration – that parts of Hari Shankar J’s opinion have no place in a jurisprudence that is formally committed to the basic idea of individual autonomy, dignity, privacy, and equal concern and respect. These include, for example, the frankly repulsive statement – that occurs on more than one occasion – that a married woman who is raped will “feel” less outraged than an unmarried woman who is raped. Examples can be multiplied; and when this judgment goes on appeal, the least that can be done is a formal expunging of these observations from the record.

And finally, this judgment shows – if anything does – the often Janus-faced character of the courts. We have two opinions – delivered in the same case – that, like ships in the night, sail past each other without even the chance of a conversation, because their premises are so very different. One opinion sees the task of constitutionalism to be interrogating power differences and breaking down social hierarchies, in order to achieve genuine substantive equality and freedom. The other opinion takes upon itself the task of defending and entrenching those hierarchies. I think we don’t see the first face of the courts often enough; but when we do – as in Shakder J’s opinion – it’s a powerful reminder of what constitutionalism, at its best, can be – and do.

[Disclaimer: The present writer was involved in the initial drafting and hearing of the petitions challenging the MRE. He has not been involved in the case since 2019.]

Guest Post: The Illegality of the Khargone Demolitions

[This is a guest post by M. Jannani.]

Recently, it was reported that homes and shops were allegedly demolished in Khargone, Madhya Pradesh following the Ram Navami violence, with neither a reasonable notice nor hearing by the district administration. Soon after, the district collector admitted that the demolitions were carried out within 5 kilometers from where the alleged Ram Navami violence took place “in order to teacher rioters a lesson”. Some news reports that reported on this demolitions have alleged that it has disparately affected Muslims in the particular region.  However, the SDO (Revenue) through an RTI reply mentioned that such action was taken against illegal encroachments under the MP Bhu-Rajasva Sanhita 1959 and section 248 of the amended act, 2018.

In Puttaswamy v. Union of India, the opinion authored by Justice Chandrachud laid down the test of proportionality to be followed in the case a state action invades the right to life or personal liberty. The following is the relevant extract of the judgment:

“An invasion of life or personal liberty must meet the threefold requirement of legality, which postulates the existence of law; need, defined in terms of a legitimate State aim; and proportionality which ensures a rational nexus between the objects and the means adopted to achieve them.“ (emphasis supplied)

In this piece, I aim to argue that the Khargone demolitions violate the above mentioned test of proportionality. In the first part of the essay, I justify the use of the proportionality test in the present case. I will then proceed to explain how, in my limited opinion, the different prongs of the proportionality test are violated by the Khargone demolitions.

An infringement of the right to life

In the previous post on this blog, Rishika Sahgal had explained in detail about the procedural safeguards and requirements of adequate notice, reasonable opportunity to be heard and access to rehabilitation facilities that have been interpreted into article 21 by various High Courts and the Supreme Court. The post explained how the judgment in Olga Tellis specifically affirmed that the right to life under article 21 of the Constitution encompassed the right to housing and livelihood, which included the right to hearing and the provision of a notice in the case of evictions. It is also made a very pertinent observation about how the procedural requirements pertaining to demolitions that were laid down in Sudama Singh – notice, hearing, meaningful engagement and rehabilitation – have been crytallized by the Supreme Court through subsequent decisions. Such requirements, the essay argued, have to be met by authorities across the country if they seek to initiate demolitions.

In the case of Khargone, it was alleged by certain persons affected by the demolition that they were neither provided a proper notice nor a reasonable opportunity of being heard before their property was demolished. Thereby, it flies in the face of the precedents discussed above as the demolitions infringe the right to life guaranteed under article 21 of the Constitution of India by violating procedural safeguards. Hence, it justifies the use of the test laid down in Puttaswamy.

Demolitions and section 248 of the Madhya Pradesh Land Revenue Code, 1956

As mentioned earlier, the authorities have justified the demolitions on the ground that the structures violate section 248 of the MP Land Revenue Code. Even though the provision empowers the tahsildar to “summarily eject” in the case of encroachment, it was held in various decisions that such powers contained in the provision are necessarily subject to a reasonable opportunity of being heard offered to the persons against whom the adverse action will be taken against. In Arun Bharti v. Madhya Pradesh, the Madhya Pradesh High Court looked into unauthorized occupations and section 248 of the Code. The court held that section 248 of the Code is a penal provision which inherently contained the “necessity of compliance of the principle of natural justice of audi alteram partem by affording reasonable opportunity of hearing”.

In Turabali v. State of Madhya Pradesh, the Madhya Pradesh HC looked into writ petitions challenging notices issued under section 248 of the Code. In this case, a time period of 3 days was given by the authorities for the removal of an encroachment. The High Court while stating that the time period provided for removal was “absolutely insufficient” also held that:

“Even if they were encroachers, then, it was for the competent authority to give them proper notice, applying proper law and providing them a reasonable time to file reply and also an opportunity of hearing. From the notices it appears that this was not intended.” (Emphasis mine)

It can therefore be observed that section 248 of code inherently allows for a reasonable opportunity of hearing and notice to be provided to persons against whom the coercive action is taken. However, such an opportunity was alleged to have not been provided to the persons against whom the coercive action was taken in the present case. Thereby, the Khargone demolitions are not backed by legality.

Collective punishment

Various ministers of the state cabinet and the district collector justified the demolitions on the ground that it was done in response to the violence that endured during the Ram Navami procession on 10th of April. But later, the action was justified on the ground that the demolitions were done against illegal encroachment and hence the state proceeded under section 248 of the code. However, it is to be noted (as had been discussed above) that the demolitions were alleged to have been done without adherence to natural justice principles. The state action in this case also suffers from over-inclusion since certain properties that did not fall within the category of illegal encroachments were also demolished and persons who owned such property were not afforded an opportunity to present their case just because such the properties were located in a particular area.  

In the case of Chandni Chowk Sarv Vyapar Mandal (Regd.) v. Municipal Corporation of Delhi , the Delhi High Court observed that all shopkeepers and allottees of chabutras could not be characterized as trespassers or encroachers. It also observed that when the state undertakes coercive action which have an effect of causing adverse consequences to persons, there exists a duty for the state to apply its mind with respect to the facts of the particular case. On the aspect of mass action against a class of persons the court held that:

“It is not enough to take precipitate action against a class of persons, with the allegation that all of them are guilty, and tainted. Even if mass action is required, principles of fairness demand that the authority apply its mind to the materials regarding individual cases.” (emphasis supplied)

Thereby, when demolitions are initiated against the properties of a class of persons, it can be observed that by merely asserting the allegation that they are guilty or tainted will not justify the “need” for such an action to be taken by the state.

Means employed is in excess of object that is sought to be achieved

In the context of mass action against a class of persons in violation of the principles of fairness, it is important to note that the Delhi High Court in Chandni Chowk Sarv Vyapar Mandal (Regd.) v. Municipal Corporation of Delhi  referred to the Supreme Court decision in UOI v. Rajesh. The Apex Court in Rajesh held that an adverse action which has an effect of unfairly punishing innocent persons and overlooks contextual considerations, would amount to “throwing to the winds the principle of proportionality in going farther than what was strictly and reasonably to meet the situation”. The Supreme Court further mentioned that such action which is excessive and not in keeping with the gravity of the offence could have the consequence of “virtually rendering such a decision to be irrational”.

Thereby, the means employed in the case of the Khargone demolitions i.e., mass action against persons with disregard to procedural safeguards and factual considerations is in excess of the object the authorities sought to achieve.

For the above mentioned reasons, in my opinion, the Khargone demolitions violate the test of proportionality.

Guest Post: Judicial Review of Preventive Detention Orders – The Supreme Court’s Progressive Judgment in Mallada v State of Telangana

[This is a guest post by Varun Ahuja.]

“…But, Sir, what the Government is after is not the power to take preventive action; what it is after is immunity from the responsibility to prove guilt or intent before a court of law. What they want is to substitute suspicion for evidence, substitute the pleasure of the executive for conviction by the judiciary. The Bill, therefore, is not only an attack on the citizen’s rights, but is also an attack on the judiciary, the power of the judiciary to protect the citizen.”

– Shri Ravinder Varma, Lok Sabha Debates, 15 December, 1980, Debate on the National Security Bill, 1980


On 4th April 2022, a two-judge bench of the Supreme Court delivered a judgment titled Mallada K Sri Ram v. State of Telangana (“Mallada”) quashing a preventive detention order under Section 3(2) of the Telangana Prevention of Dangerous Activities Act, 1986 (“TDA” or “the 1986 Act”). The detenu had filed an appeal against the Telangana High Court judgment which had upheld the order of detention. The order was quashed because it was passed on ‘stale grounds’ and that there was non-application of mind by the detaining authority; specifically, by the time the order of detention was passed, the detenu was already on bail and no incident had happened even after the bail conditions imposed had lapsed.

In India, the Executive enjoys a lot of powers when it comes to preventively detaining individuals to ‘prevent’ a commission of an offence. There are four central laws (Conservation of Foreign Exchange and Prevention of Smuggling Activities Act, 1974, National Security Act, 1980, Prevention of Black Marketing and Maintenance of Supplies of Essential Commodities Act, 1980 and Prevention of Illicit Traffic in Narcotic Drugs and Psychotropic Substances Act, 1988) under which authorities can preventively detain individuals, and the States have their own power to enact laws authorizing preventive detention. The Parliament, under Entry 9 of List I, is authorized to make laws providing for preventive detention on subject of ‘defence’, ‘foreign affairs’ or the ‘security of India’. Simultaneously, the States can make laws under Entry 3 of List III for reasons connected with ‘security of a state’, ‘maintenance of public order’, or the ‘maintenance of supplies and services essential to the community’. The constitutionality of these terms – vague and over-broad in their own right – was upheld by the Supreme Court in A.K. Roy v. Union of India. The Court admitted that the terms might be vague and incapable of a precise definition; but refused to strike down the law (National Security Act, 1980) on the ground of vagueness and uncertainty and ‘hoped’ that they will be applied and interpreted narrowly. Furthermore, judicial review of detention orders is largely limited to only procedural grounds barring few exceptions in certain cases in order to ensure some fairness and uniformity.

Three important aspects set Mallada apart in the Supreme Court’s preventive detention jurisprudence. First, the Chandrachud J rightfully holds that mere apprehension of breach of law and order will not be sufficient to meet the threshold of ‘maintenance of public order’ to justify invoking preventive detention (Para 15). Second, on the judicial reviewability of subjective satisfaction by the detaining authority, the judgment falls into the exception where the court reviews if an order of detention could have been passed on the material placed before the detaining authority. Third, the Court highlights the misuse of the 1986 Act and deviates from the normal procedure by not allowing the State to file a counter affidavit.

In this article, I examine the Court’s jurisprudence on judicial reviewability of the detaining authority’s subjective satisfaction comparing Mallada to an earlier decision of the Court in Union of India v. Dimple Happy Dhakad (“Dhakad”). I also comment upon the procedure adopted by the Court in not allowing the State to file a counter affidavit and how it can benefit the Court in more efficiently deciding habeas corpus petitions against preventive detention.

Judicial Review of Subjective Satisfaction

For a preventive detention order to be passed, the detaining authority has to review the material placed before it by the police or any other agency asking for the detention, apply their mind, and then take a decision whether to authorize it. There are many judgments which state that this subjective satisfaction of the detaining authority (whether a detention order should be passed on the material placed before it) is not to be interfered with. [See Senthamilselvi v. State of Tamil Nadu, (2006) 5 SCC 676].

However, over the years the Courts have consistently pushed the boundaries of this limitation to ensure that there is some kind of fairness to the decision-making power. Therefore, decisions can be classified broadly into two types; one where the Court follows the general rule that it will not place itself in the shoes of the detaining authority to see if, in fact, detentions order could have been passed. And second, where the Court allows review of this subjective satisfaction to prevent arbitrariness and ensure that the procedure remains reasonable. For example, in a situation where a material fact (most commonly, if the detenu was already in judicial custody) is not placed before the magistrate, an order made on such a material will be quashed by the Court [See, Farooq Ahmad Bhat v. UT, J&K, 2021 SCC OnLine J&K 369]. I highlight that the judgement in Mallada, which falls in the second category of decisions, starkly differs from the Court’s approach in Dhakad.

To review or not to review

The decision in Dhakad has been criticized on many counts. Most importantly, that being a two-judge bench, it holds contrary to a three judge bench decision in Rekha v. State of Tamil Nadu [(2011) 5 SCC 244]. For our discussion, its only relevant to state that Dhakad holds: firstly, it is not illegal to pass a detention order against a person who is already in custody and about to be released on bail, and secondly, imminent possibility of release is an issue that depends upon the ‘subjective satisfaction’ of the detaining authority and as such, is beyond the scope of judicial review (Para 46). It has been argued that this pronouncement dilutes the ‘trend of heightened review’ set in cases where the detenu is already being processed under regular criminal law; and furthermore, that the requirement to specifically record satisfaction that states ‘there is an imminent possibility of release’ and ‘a grave likelihood of the person returning to the prejudicial activities’ does not interfere with the subjective satisfaction of the authority but only makes it clear that the authority had considered the issue. It is important to keep in mind that the issue that Dhakad was dealing with was that preventive detention orders were passed against a person who was already in custody and therefore, posed no threat to public order per se. The argument that an order in such case is to prevent a person being released on bail only goes to indicate the normalization of the extraordinary powers of the executive.

The judgment in Mallada does not stop by simply saying judicial review of subjective satisfaction is not possible if the authority records satisfaction to the effect that there is possibility of bail and the detenu might repeat the offence. It adds that the order stated that the detenu ‘may violate the bail conditions’ and that ‘there is an imminent possibility of him committing similar offences’. Therefore, the detaining authority, in its wisdom, had reached the conclusion that there is an apprehension of the detenu violating bail conditions and repeating similar offences. However, the Court analyzed both the grounds and stated that the conditions of bail had concluded as of April 2021 and the order was only passed in May 2021. Moreover, there had been no further incident after the bail conditions expired. It concludes by stating that the case is of clear non-application of mind and competent to be dealt under ordinary criminal law. Interestingly, the State had neither moved to apply for cancellation of bail nor appealed the order granting bail; instead, it used its extraordinary powers to preventively detain the individual for an incident in which a Court had already granted bail.

Reports have shown that the the 1986 Telangana Act is used quite often as a tool to supplant ordinary criminal procedure and deny bail to individual, even in cases where no bail application is pending before the court but there might be a possibility that the detenu might file one. A recent amendment to the Act has  further broadened its scope and the procedure made more efficient in subverting ordinary judicial process. In 2018, to get away from justifying the invocation of the Act by showing that an activity was an actual threat to public order (the only ground on which the Act could be invoked – Section 3), the Title of the Act was amended to explicitly include more categories of offences which could normally be dealt by the ordinary criminal procedure. It now includes – ‘Spurious Seed Offenders’, ‘Insecticide Offenders’, ‘Fertiliser Offenders’, ‘Food Adulteration Offenders’, ‘Fake Document Offenders’, ‘Scheduled Commodities Offenders’, ‘Forest Offenders’, ‘Gaming Offenders’, ‘Sexual Offenders’, ‘Explosive Substances Offenders’, ‘Arms Offenders’, ‘Cyber Crime Offenders’ and ‘White Collar’ and ‘Financial Offenders’. The Court in Mallada has rightly called out the abuse of law by stating the number of orders quashed by it and directed the Government to look at the challenges pending before various courts and the advisory board to ensure ‘fairness’ (Para 17).

‘Live and Proximate Link’

As stated in the beginning, there is another line of decisions where the Court looks at the material on which the detaining authority passes the detention order; for example not considering a material fact that the detenu was in custody. Similarly, the Court in Mallada makes an important point while analyzing the material placed before the detaining authority. Chandrachud J notes that the order was passed nearly seven months after the registration of the first FIR and about five months after the registration of the second FIR and was interestingly executed after one month (Para 11). The Court relies upon its decision in Sama Aruna v. State of Telangana (2018), to iterate that if there is absence of a ‘live and proximate’ link between the incident which is part of the material and the detention order, it will be equivalent to ‘punishment without trial’.

This requirement of a live and proximate link is not a recent safeguard created by the Supreme Court. The first case which acknowledged there has to be a ‘live and proximate link’ between the grounds of detention alleged and the purpose of detention was in Bhawarlal Ganeshmalji v. State of T.N., (1978). Although, the order was not quashed in that case, it stated that this link will be assumed to be ‘snapped’ if ‘there is a long and unexplained delay between the date of the order of detention and the arrest of the detenu.’ Thereafter, relying on Bhawarlal Ganeshmalji, the Supreme Court in Shafiq Ahmad v. D.M. (1989) and P.U. Iqbal v. Union of India (1991) quashed the detention order in the absence of ‘live and proximate link, where the delay was of two and half months and one year respectively. Furthermore, relying on P.U Iqbal, the Court in Sama Aruna had authoritatively stated that a detention without the link is punitive in nature.

The underlying principle for having preventive detention law is not to punish the detenu but ‘prevent’ them from doing a crime. A judicial review, quashing an order, for ‘stale’ grounds, gives an insight behind why the executive chose to exercise the power; not to prevent a crime, but punish the detenu using an extraordinary criminal procedure. The executive is allowed to do this with impunity and an implicit sanction from the Court since it takes too much time in deciding the legality of the detention order; long enough for the Government to achieve its purpose.

A Welcome Deviation

It is important to understand preventive detention as it operates in today’s framework from a punitive aspect, because of the amount of time a person has to spend in custody while the legality the detention order is being decided. Prolonged incarceration aids the State in achieving the objective of keeping a detenu – oftentimes a political dissident – out from interacting in public life. One of the easiest ways to delay the proceeding is to take adjournments to file counter affidavits. Although the problem is not recognized, there is a judgment delivered by Madras High Court, where the Court deprecated the practice of taking multiple adjournments to file a counter affidavit by the detaining authority.

Shrutanjaya Bhardwaj, in a study, analyzed the time taken by the Supreme Court in deciding habeas corpus cases against preventive detention between the years 2000 to 2019. The study concluded that the writ was reduced to a meaningless remedy because of the amount of time taken by the Court was more than the maximum time allowed for detention in various preventive detention acts. At the end of the study, a possible remedy was mooted that the Court should decide a habeas corpus petition without a counter-affidavit from the State and solely on the basis of the documents produced before it; only in cases where the State needs to explain a delay, should a counter affidavit be allowed.

In Mallad the Court did exactly this. It lamented that even though the notice was served no counter affidavit was filed. It proceeded taking into consideration the affidavit filed with the High Court. As a result, the case was decided in a span of 45 days (Supreme Court website shows filling date as 16-02-2022); compare this to the findings of the habeas corpus study with regard to the successful petitions i.e. where the Supreme Court was the first court to grant relief (as was the case in Mallad). The Court on an average took 159 days; the shortest was 34 days and in two cases it took longer than one year to decide the petition (448 and 377 days). Therefore, in cases where the Court can decide matters solely looking at the documents, it should not insist on a counter affidavit and give repeated adjournments for the same.

Admittedly, Courts have often delivered judgments without taking a counter affidavit from the detaining authority, but only after providing multiple opportunities to them (For example, see, here and here). The present case is different because there was only one date of hearing before the judgement was pronounced and thus, only one opportunity of three weeks was given to the detaining authority. Hearteningly, the Court has deployed fierce rhetoric to make its point (‘The liberty of the citizen cannot be left to the lethargy of and the delays on the part of the state’). Of course, it remains to be seen if the Court continues with this practice.