Guest Post: A Battle Between Bail and Silence — Comparing Section 45 PMLA with US Pre-Trial Prosecutorial Discovery

(This is a guest post)

The present post analyses the constitutionality of restrictive bail conditions in the Indian anti money laundering law from the perspective of the right against self-incrimination. Section 45 of the Prevention of Money Laundering Act, 2002 (“PMLA”) imposes twin conditions that an accused must meet to secure bail: (i) satisfy a court that there are “reasonable grounds to believe” that the accused is “not guilty of the offence”, and (ii) that the accused is unlikely to commit any further offences while on bail. By way of background, the Indian Supreme Court dismissed a challenge to Section 45, PMLA in Vijay Madanlal Choudhary v. Union of India. Petitions seeking a review of the judgment are pending). However, neither the original judgment nor grounds taken in the review petition sufficiently discuss the interplay between Section 45, PMLA, and the right against self-incrimination.

This post employs a comparative analysis of case law pertaining to pre-trial prosecutorial discovery rules in criminal cases in the United States to explore the intersection between Section 45, PMLA, and the right against self-incrimination.

Hypothetical Scenario: The Conflict Between Self-Incrimination and Disclosure of Material During a Bail Hearing

Imagine an accused is facing charges of money laundering, with one of the key allegations being that they received the proceeds of a crime from person X at location Y on a specific date. Aware that they must provide prima facie evidence of innocence to secure bail, the accused submits an affidavit to the court asserting that their mobile phone’s GPS data places them far from location Y at the relevant time. In response to this assertion, the prosecution requests the accused’s phone password to verify the accuracy of the GPS data. Apprehensive that failure to comply with this request could result in their bail application being denied, the accused agrees to provide the password to unlock the phone voluntarily.

Upon inspection, the prosecution confirms accuracy of the GPS data, but in the process, examines other files and data on the phone, discovering additional incriminating material that they were not aware of. The prosecution subsequently submits a counter-affidavit, which includes this newly discovered material. In response, the accused argues that this material should not be taken into consideration, asserting that it was obtained through their compelled disclosure and is therefore subject to the protections afforded by the right against self-incrimination. The prosecution argues that the password was provided voluntarily without any court order, and it does not amount to compelled disclosure.

I’ll build on this example and examine the constitutionality of pre-trial disclosures made by an accused during the bail stage.

US Case Law

The Indian guarantee against compelled self-incrimination avowedly borrows from the self-incrimination clause of the Fifth Amendment to the U.S. Constitution. Historically, the values of the American accusatorial system required the state to shoulder the entire burden in a criminal prosecution without any compelled help from the accused and the accused is not required to disclose their defence until the prosecution has presented its case. However, this privilege has been eroded through legislative provisions that require the accused to disclose certain portions of their evidence before trial.

In Williams v. Florida (“Williams”), the rule under challenge required the accused to disclose a list of alibi witnesses before trial. The accused was convicted and appealed against conviction on the ground that compelling him to reveal the name of his alibi witnesses violated his right against self-incrimination. However, the challenge was dismissed by the U.S. Supreme Court. To reach its conclusion, the Court first recognised that the privilege against self-incrimination consists of three elements: (i) compelled disclosure (ii) related to testimonial matter (iii) incriminating nature of the communication. The Court held that two elements, i.e., testimonial communication and ‘tendency to incriminate were present in the rule requiring disclosure of alibi witnesses. However, it concluded that the relevant rule did not compel the accused to disclose alibi witnesses. It reasoned that the defence had intended to present the alibi witnesses at trial regardless, and by providing the prosecution with this information prior to trial, the defence was simply ‘accelerating the timing of the disclosure’, thereby facilitating a more efficient trial process. It was held that even during the trial, the accused would still face the same choice: (i) to remain silent, or (ii) to offer a defence that could potentially lead to self-incrimination through cross-examination.

‘Acceleration of timing’ is an established principle today based on which pre-trial discovery by prosecution has been expanded vastly over the years. It is widely believed that prosecutorial discovery enhances the state’s ability to rebut the accused’s case and effectively cross-examine their witnesses. In light of the analysis of Williams, the issue arises as to whether the prosecution is entitled to seek any and all forms of disclosure under the guise of the ‘acceleration of timing’ principle. The subsequent cases suggest a different interpretation.

There are two ways in which pre-trial discovery from an accused can assist the prosecution. It can either (i) help prosecution in investigating the grounds of defence taken by the accused and consequently, help in preparing rebuttal evidence to the defence of the accused; and (ii) provide leads to the prosecution that can help in building a case-in-chief against the accused.

Various commentators have argued that Williams was only concerned with and envisaged the preparation of rebuttal evidence using pre-trial discovery by an accused. The Court's reasoning supports this argument. Firstly, the court equated the situation of pre-trial discovery with the stage of defence evidence at a trial, where the accused must reveal the identity of their witnesses and submit them to cross-examination, which may furnish the State with leads to incriminating rebuttal evidence. Therefore, the court only intended to confine itself to rebuttal evidence while testing constitutionality of the discovery rule requiring disclosure of alibi defence. 

In this regard, it is further pertinent to recall that in Williams, the Court held that the alibi discovery rule did not amount to a compelled disclosure since the choice between (i) remaining silent and (ii) presenting a defence that might incriminate the accused through cross-examination or provide leads to the state to impeach their witness is ‘inherent in the trial process’. The natural corollary of this reasoning is that choices compelling an accused to make choices not inherent in a trial process amount to a compelled disclosure and, hence, violate the right against self-incrimination. In a case where the accused’s disclosure can be used to build the prosecution’s case-in-chief, the accused’s decision whether or not to disclose requires them to choose between defending by affirmative evidence or defending by failure of government proof. 

In his article on prosecutorial discovery, Prof. Eric Blumenson argues that such a choice is not inherent in the trial process, since at a trial, the accused leads evidence only after the government proves a prima facie case, and hence, the adversarial criminal process protects an accused from aiding the prosecution in building its case in chief. Therefore, if the prosecutorial discovery rule, as approved in Williams, is used by the prosecution to build its case in chief, the said discovery would not pass the ‘acceleration of timing’ principle in Williams and, accordingly, would violate the right against self-incrimination.

Secondly, the fact that the majority opinion’s focus was only on rebuttal evidence can be gauged from the examples it cites in the opinion. The Court gave an example to the effect that if the disclosure rule was not in place, the way alibi testimony would have played out is that the prosecutor would have asked for a continuance / adjournment after the alibi witness’s examination-in-chief and would have found rebuttal testimony to impeach the alibi witness before cross-examining them. The Court concluded that since the said sequence of events would not violate the accused’s right against self-incrimination, neither would the pre-trial discovery rule at issue in Williams. Therefore, the said example makes it evident that the scope of Williams was limited to rebuttal evidence and the judgment did not envisage use of prosecutorial discovery as a tool to build a case-in-chief.

The Scarcity of Jurisprudence on the Use of Pre-Trial Discovery from the Accused to Construct the Prosecution’s Case-in-Chief

There appears to be a dearth of case law addressing whether pre-trial discovery from the accused can be used by prosecution to build its case-in-chief. Before Williams, the California Supreme Court in Prudhomme v. Superior Ct (“Prudhomme”) delineated two important points concerning the right against self-incrimination: (i) the right against self-incrimination is compromised when the disclosure could potentially ease the prosecution’s burden of proving its case in chief and the accused’s disclosure cannot be used for the said purpose; (ii) any disclosure that could serve as a “link in a chain” of evidence establishing the accused’s guilt is prohibited under the principle of right against self-incrimination.

Another Californian case took the same view as Prudhomme. In re Misener (“Misener”), a broader interpretation than Prudhomme was taken when the majority held that Fifth Amendment privilege does not end upon the establishment of a prima facie case by the prosecution and that it barred any prosecutorial discovery that prevented the prosecution from carrying the entire burden of proving the accused guilty beyond a reasonable doubt, whether compelled disclosure might serve to make the prosecution’s case or to unmake the accused’s.

Therefore, in Prudhomme and Misener, the California Supreme Court explicitly recognised that any pre-trial disclosure that lessens the burden of the prosecution in proving its case-in-chief is unconstitutional. However, none of these cases dealt with a scenario in which the prosecution used pre-trial disclosure to build its case-in-chief.

Both Prudhomme and Misener were subsequently overruled in Izazaga v. Superior Court. Here, the California Supreme Court sustained a broad reciprocal disclosure requirement, rejecting the Prudhomme analysis. The California Supreme Court held that under the rationale of Williams, pre-trial discovery does not constitute compelled self-incrimination. Therefore, Prudhomme and Misener were overruled based on the reasoning in Williams, which was itself confined to envisioning pre-trial discovery for the purpose of collecting rebuttal evidence, rather than for building the prosecution’s case-in-chief.

It is pertinent to note that the issue of whether the prosecution can use pre-trial discovery to build its case-in-chief was not answered and has remained unanswered. There is no denying that prosecutorial pre-trial discovery over the years has become broader. For example, trial court rulings, based on the ‘acceleration of timing of disclosure’ laid down in Williams, have accepted a requirement of advance notification of defences, of expert witnesses, and of all other witnesses. Nevertheless, it can be fairly argued that these pre-trial disclosures should serve solely to counter defence witnesses and cannot be utilised by the prosecution to establish its case-in-chief. 

An example of this principle can be found in a trial court judgment in State v. McClaren in which while allowing pre-trial discovery, the court held that “any concerns defendant has concerning the disclosure [itself] potentially being used by the prosecutor in the case-in-chief could be addressed by an in camera review by the circuit court.” Further, even in Izazaga, which overruled Prudhomme and Misener, the concurring opinion by Justice Kennard mentions that there is a limitation on the manner in which the prosecution may use the discovery, and that it would be “impermissible for the prosecution to use, as part of its case-in-chief, evidence gained through discovery of statements of potential defense witnesses.” Therefore, while courts have been liberal in allowing prosecutorial pre-trial discovery based on Williams' acceleration principle, they are cognizant of the possibility of violation of the Fifth Amendment Principle if the said discovery is used for building prosecution’s case-in-chief.

Applying U.S. Pre-Trial Discovery Principles to the Hypothetical Scenario in Bail Proceedings under Section 45, PMLA

The principles discussed in the preceding section regarding pre-trial prosecutorial discovery should be applied in bail proceedings under PMLA. The justification for this is evident: Section 45, PMLA forces the accused to reveal their defence at a pre-trial stage. Consequently, bail proceedings under PMLA mirror the function of pre-trial prosecutorial discovery, insofar as they afford the state early access to key aspects of the accused’s defence before the commencement of trial.

Accordingly, applying these principles in the Indian context, the argument remains that if the prosecution uses the information disclosed by the accused to build the ‘foundational facts’ in its counter-affidavit (or the chargesheet later), the same would violate Article 20 (3) of the Constitution. Currently, the prosecution is not restricted from doing this, and in practice, the prosecution routinely files supplementary reports / complaints based on arguments made by the accused at the bail and charge stage to plug loopholes in its case-in-chief. Accordingly, even as per a broad interpretation of discovery rules under U.S. jurisprudence, the current use of Section 45, PMLA is violative of Article 20 (3) of the Constitution.

In light of the discussion of jurisprudence surrounding pre-trial prosecutorial discovery in the U.S., it can be concluded that Section 45, PMLA directly impacts the right against self-incrimination. The compelled disclosure of the accused’s defence at this stage undermines the privilege against self-incrimination. The argument that such disclosure is voluntary overlooks the coercive nature of the choice between liberty and silence. Courts must adopt a restrictive interpretation of Section 45, PMLA to prevent its misuse by the prosecution as a tool to build its case-in-chief through the accused’s pre-trial disclosures at the bail stage. This approach ensures that bail proceedings do not become a prosecution strategy session.

Guest Post: Trial in Absentia under BNSS — A Procedural Quagmire?

(This is a guest post by Acharaj Kaur Tuteja)

The Bhartiya Nagrik Suraksha Sanhita, 2023 has made several conspicuous departures from the Criminal Procedural Code, 1973 in an attempt to streamline criminal procedures in the country. Among these is Section 356 which permits the trial of proclaimed offenders in their absence and points towards the evolving relationship between state convenience and procedural fairness. At first glance, the provision seeks to solve a practical problem: the stagnation that follows in courtrooms when accused persons abscond. But beneath the pragmatic veneer lies a set of unanswered questions about what is willing to be traded in the name of efficiency.

Trials in absentia are not a novel idea. Jurisdictions across the world have flirted with the idea out of political necessity or emergency. Indian law, however, has resisted this temptation by tethering the right to a fair trial to the physical and participatory presence of the accused. Section 356 marks a departure from that tradition. What does it mean to conduct a criminal trial without the accused? Can the adversarial process survive the absence of one of its central participants? And how, if at all, does this square with the constitutional promise of due process under Article 21?

This post explores Section 356’s structure and underlying presumptions. It demarcates the dangers it presents to a system already overburdened by inequitable procedural protections. The BNSS’s solution leans towards the kind of innovation that is prioritising closure before justice. A hasty conclusion may reconstruct Kafka’s courtroom in modern Indian criminal law, despite the validity of the state’s interest in avoiding delay.

Dissecting Section 356: Text and Structure

To break it down, Section 356(1) empowers the Court to proceed with trial if the accused, after being declared a proclaimed offender under Section 84, continues to evade appearance. This proclamation arises only after issuance and non-compliance with a warrant, followed by a public notice. The trial may not only begin, but conclude, and a judgement may be rendered in the absence of the accused. The Court merely has to record the reasons in writing and ensure representation via a pleader. A notice has to be published additionally in a national or local newspaper – a formality that may neither inform or alert the accused in rural or otherwise marginalised contexts.

Section 356 fails to make a legal distinction between two conflated categories: the absconding accused and the absentee accused. The former is a legal status granted by Section 84 BNSS after meeting specific requirements, mostly post-proclamation wilful evasion. The latter’s absence may be for a variety of reasons, including illness, non-service of summons, coercion, or logistical inaccessibility. There is a chance that the legislation will dismantle these classifications, establishing a loophole whereby absence is assumed to be inherently related to guilt. Unfortunately, the language of Section 356 does not operationalise the subtlety that not all absentees are absconding.

Section 356 sidelines the requirement for deeper judicial inquiry into whether such evasion is wilful, or whether the absence results from coercive tactics or abuse of power by investigating agencies. This opens the door to strategic absences orchestrated by the police or other law enforcement authorities. The absence of terms such as ‘maliciously’ or ‘deliberately’ in the statutory text does not relieve the Court from conducting such inquiries; rather, this omission renders the provision even more susceptible to abuse, where the bare fact of non-appearance becomes sufficient to greenlight trial.

The Illusion of Remedy: Examining of Witnesses and Making an Appeal

The proviso under clause (4) of Section 356 states that if the proclaimed offender is arrested and produced before the Court during trial, the Court may, in the interest of justice, allow him to examine any evidence which may have been taken in his absence. Firstly, the clause imposes a double burden: the accused must prove both a) lack of notice and b) the absence of a wilful evasion. This reverses the usual presumption of innocence and places the accused in a position where they are trying to recover a sliver of participation in proceedings that have already condemned them.

Secondly, the court is under no obligation to allow cross-examination, nor is there a requirement to re-open earlier stages of the trial. The word ‘may’ vests a discretionary power with the court. This lack of guarantee under the clause goes against the principle of providing the accused with a reasonable opportunity of being heard. The Supreme Court, in the case of State of Kerala v. KT Shaduli Grocery Dealer, allowed the assessee to cross-examine the witnesses on whose evidence the Sales Tax Officer had relied to make the arrest. The Court recognised that under the rule of audi alterem partem, “to prove” means to establish the correctness, or completeness of a fact by any mode permissible under law. The usual mode recognised by law for proving a fact is by production of evidence and evidence includes oral evidence of witnesses. The opportunity to prove the correctness would, therefore, necessarily carry with it the right to cross-examine the witnesses.

The accused, once labelled a proclaimed offender and tried in absentia, may find themselves returned to a process that has long since passed them by, with key evidentiary stages foreclosed. The danger of an unfair trial is further compounded by clause (6) that states that even if the accused is produced before court or arrested at the conclusion of the trial, the proceedings need not be stalled. The judgement can be pronounced in their absence. Moreover, the right to appeal is curtailed under Section 356(7) wherein the appeal is barred unless the offender presents himself before the appellate court within 3 years of the passing of the judgement.

What happens then, is a skeletal notion of a trial, moving from absence to conviction without substantive procedural rights. In Kartar Singh v. State of Punjab, the Supreme Court was unequivocal in its caution: dilution of safeguards in the name of expediency, carries the risk of systemic abuse and wrongful conviction. The purity of the procedure to discover truth shall always remain fair, sensitive to the needs of the society and fairly and justly protect the accused. And these clauses of Section 356 exist not to protect the accused, but to immunise the provision from critique.

Undermining Adversarial Ethos

Under Section 356(1) the Court is empowered to proceed with the trial of a proclaimed offender “in the like manner and with the like effect as if he was present”, thereby eliminating any procedural distinction between a full adversarial trial and a unilateral one. The assertion that the accused’s wilful absence amounts to a waiver of rights, becomes a linchpin to justify lack of procedural parity.

The right to be tried in person, to confront witnesses, and to mount a defence are all implicit in Article 21’s expansive right to a fair trial. The Supreme Court in Zahira Habibullah Sheikh v. State of Gujarat, while affirming the centrality of a fair trial, clarified that it includes not just the right of the prosecution but, more importantly, of the accused to have their case heard fully and fairly. This renders the trial permitted under Section 356 structurally disfigured. Informing the accused’s friend or relative and ensuring that legal aid is appointed does not cure the deeper malaise of negating the accused’s participatory rights.

This transformation of the trial from adversarial to inquisitorial is not accompanied by the institutional safeguards of the latter. In inquisitorial systems, the judge assumes an investigative role, bound by duties of neutrality and evidence-gathering in favour of both sides. Indian courts, in contrast, are not structured for such a role. The result is a hybrid system that borrows the most coercive aspects of the inquisitorial form while discarding the checks and balances that justify its operation.

Conclusion

There is no denying that trials in absentia address a very real institutional challenge. In an overburdened criminal justice system, where accused persons routinely manipulate safeguards to stall proceedings, Section 356 may act as a corrective. It reaffirms that justice will not be held hostage to deliberate non-appearances.

However, the problem does not justify abandonment of the process. The risk lies in the unstructured application of trials in absentia. The provision, in its current form, rests on a troubling presumption: that a proclaimed offender’s absence is necessarily wilful, and that procedural compliance with notice formalities is enough to substitute meaningful participation. A declaration under Section 84 BNSS is formally premised on wilful concealment or absconding but the statute’s structure outsources that finding to procedural triggers (two warrants, public notice, etc.), not to a substantive inquiry into motive or capacity. In other words, the statute assumes wilfulness from procedural non-compliance. But procedural non-compliance can occur in a number of ways that have nothing to do with intent. The ask isn’t to duplicate the finding of wilfulness - it’s to interrogate how shallow the first one really is. If procedural default alone gets elevated into a legal fiction of intentional evasion, then a second judicial inquiry – before triggering the far more serious consequence of trial in absentia – becomes less redundant and more necessary.

To avoid this turning into a tool of convenience, certain safeguards must be built into the provision. Firstly, the judicial determination of whether the accused has actually absconded should be more than just a declaration. It should follow an inquiry to examine whether the accused had a reasonable opportunity to appear, and whether the evasion is deliberate or due to an extraneous hardship. Secondly, recording depositions via audio-video [as mentioned in Section 356(5)] should be the norm, not a technological afterthought. The attempts to contact the accused in any manner must be furnished duly. Thirdly, the accused should have the opportunity to retrospectively participate in the trial and cross-examine the witnesses if the conviction has been based on their evidence. And lastly, use of Section 356 must be accompanied by periodic judicial review to ensure that the waiver of rights was voluntary in nature.

Trial in absentia, then, is not inherently illegitimate, but it is inherently dangerous. It demands a justice system that can distinguish between tactical evasion and structural absence, between the wilfully hidden and the invisibly marginalised. Without such discernment, Section 356 risks becoming a procedural sleight of hand, manufacturing the illusion of fair process while steadily hollowing it out. The State’s interest in proceeding with trials must not eclipse the constitutional guarantee of a fair one. To that end, the legitimacy of this provision will rest not on how often it is used, but on how scrupulously it is constrained.

Two Sides of the Same Coin

Outside of the legal fraternity, it did not matter much that some judges of the Supreme Court were retiring in these past weeks. Within the legal fraternity, though, the retirement of Justice A.S. Oka led to a genuine outpouring of affection and emotion from across the aisles; no doubt made more stark by the near total absence of such reaction a few days prior in respect of Justice Trivedi. 

Revisiting this juxtaposition, and reading several judgments authored by the two retiring Justices (limited, largely, to the field of criminal law and procedure), I could not escape thinking how Justices Oka and Trivedi are but two sides of the same coin of judicial discretion. When it came to matters of criminal law, the former wielded discretion to actively question exercise of state power, while the latter wielded judicial discretion to shield the state from questions; of course, subject to exceptions. 

We might like one approach over another, sure, and there are no doubts about which approach this blog might prefer. But the unfortunate truth is that our preferences do not count for much as a legal principle. I say this, because at the end of the day, in our frustratingly polyvocal judicial system with its constitutional mandate of rights riddled with exceptions, there is no single correct approach to issues that these judges were dealing with. There is no fundamental principle, no categorical imperative in these parts.

If anything, the closest thing that Indian constitutional jurisprudence in action appears to have in the name of 'principle', is a loose concept of balance. A jurisprudence of the 'Force' in Star Wars with its constant tussle between dark and light, if you like. In this jurisprudence of balance, every story has two sides, every right is tethered to its exceptions, and so even the most egregious instance of personal liberty being violated is not an absolute but can have some justification that requires we form a special investigation team to find out. 

This is our fickle rule of law, where consistency in outcomes is an honourable exception and even the simplest case appears to be cast in jeopardy. A few senior lawyers I know stress the jeopardy aspect by telling their clients that going to court, especially superior courts, is quite like playing high-stakes poker. Are they wrong? Administration of such a 'rule of law' where every scenario brooks an exception ultimately turns entirely upon how every singular actor wields their discretionary power to advance their personal notions of law, justice, and whatnot. One judge can view that discretionary power as being guided by a vision to uphold liberty, the other can view that same power as guided by a mission to secure law and order. Both views are correct, as much as you might not like one of them. 

The problem is that both views should not coexist in all scenarios. Some situations do not justify two views. Some cases only have one correct view. And these truths should not change whether we are dealing with petty disputes between family members or dealing with dissidents who disagree with the State and its policies. In ours, however, they do. A jurisprudence of balance is not one in which the rule of law retains an inconvenient and unwavering consistency in times of war as in times of peace, but one where law must have the flexibility to bend and contort itself into whatever is the convenient shape for the day. Is this jurisprudence of convenient balance something to be scornful about? At times. But if you think about the high chance that any revisit of the status quo will make things intolerably and inexorably worse — by, say, stacking all the decks in favour of the State — then a constant tussle between impulses is a happy compromise to live with. 

We can eulogise one judge and demonise another all we like, but no one person or their vision can straighten out the many contradictory impulses within the Indian legal system, all of which can be traced back to the Constitution itself. In the grand scheme of things, all that they manage is tilt the balance, by wielding their awesome power of judicial discretion. Depending on where you stand, that tilting of balance is something will either be something to be very grateful for, or extremely despondent about. 

Untangling the Many Threads of the Sarla Gupta Judgment

A Three Justices' Bench of the Supreme Court rendered its judgment in Sarla Gupta & Anr. v. Directorate of Enforcement [2025 INSC 645 (Sarla Gupta)]. The Court was hearing different appeals together, which presented somewhat similar issues that could loosely be grouped under the category of obligations on prosecuting agencies to disclose the material gathered during an investigation to the accused, in context of cases under the Prevention of Money Laundering Act 2002 ['PMLA']. 

While the judgment itself does not categorise issues, I think it is helpful if we are to try and make sense of its conclusions [summarised at Paragraph 55]. An attempt at the issues is as follows:

1. Whether the accused has a right of disclosure to seek copies of documents seized during searches done under the PMLA by the Enforcement Directorate? 

1A. If so, at what stage can the right be enforced and the documents disclosed / shared with the accused?

2. Whether the accused has a right of disclosure to seek copies of the documents filed by the prosecuting agency in support of its complaint under the PMLA ('relied upon documents')?

2A. If so, at what stage can the right be enforced and the relied upon documents disclosed / shared with the accused?

3. Whether the accused has a right of disclosure to seek copies of the documents seized but not placed in support of its complaint under PMLA by the prosecuting agency ('un-relied upon documents')?

3A. If so, at what stage can the right be enforced and the un-relied upon documents disclosed / shared with the accused?

Issues 1 and 1A

The first set of issues — 1 and 1A — were the easiest to answer for the Court. The PMLA itself provides that in event of a search and seizure, a list of seized items must be supplied to the person concerned, and the person can apply to get a copy of the documents when the authority concerned has decided to retain them [Sections 17 to 21]. So, the answer to issue 1 was 'yes', and the answer to 1A was that the person can apply to get copies after a decision is made to retain them for longer than thirty days.

Issues 2 and 2A

The next set of issues — 2 and 2A — are slightly more complicated because of how the PMLA and the general procedural law under the Criminal Procedure Code 1973 ['Cr.P.C.'] — now  the Bharatiya Nagarik Suraksha Sanhita 2023 ['BNSS'] — operate. The Cr.P.C. applies to PMLA prosecutions, but where there is a conflict between the two laws, the PMLA governs. In this realm of supplying complaints etc., there is no specific clause within the PMLA, taking us back to the Cr.P.C. [clarified within Sarla Gupta as well]. 

Indian criminal law envisages prosecutions can start either by a state agency investigating facts and bringing a case, or by an aggrieved party filing a case basis material it has in its possession to prove wrongdoing. Ordinarily, the state agency tasked with investigating cases is the police, and police files what is called a 'police report'. The disclosure rules for material filed with a police report are fairly clear: material which the agency is relying upon to support its case, has to be supplied [Sections 173(5), 207]. 

When it comes to a 'complaint', the Cr.P.C. stipulates that if a court finds that a person ought to face trial upon the allegations in a complaint, it must ensure a copy of the complaint is given to an accused. But, there is no corresponding right of disclosure availing to an accused here, as it is with cases launched pursuant to a police report. The limited exception, textually speaking, is a requirement for a magistrate's court to ensure that materials filed with a complaint are furnished to the accused in cases where the prosecution is one which must be conducted before a sessions court [Sections 204, 208]. I say textually, because courts have read in a similar disclosure requirement in cases launched on complaints for some time now.

This background is necessary to understand issues 2 and 2A. Even though prosecutions under PMLA are lodged after investigations by a state agency — the Enforcement Directorate — the document launching the prosecution is styled as a 'complaint' and not a police report [Section 45 PMLA]. The Cr.P.C., as we have seen, confers clear rights in cases launched on a police report, but is not so strongly worded when it comes to complaints. What this reflects is that, unlike issues 1 and 1A, here the Court in Sarla Gupta had some options on how to proceed. 

How has it answered the issues framed above? On issue 2, the Court has taken two steps. First, it clarifies that when a person is brought to court to face trial in a PMLA case, they must be provided with not just the complaint but also the materials being relied upon in support thereof [Paragraph 25]. This was held necessary owing to the general Cr.P.C. rule on supplying copies of complaints to accused persons. In other words, the Court recognised an obligation on part of the agency to supply a copy of the material it proposes to rely upon in support of its case. 

What about the right of the accused within the statute to demand for such relied upon material? Here, recall, that the rights-language is far clearer for police reports than for complaints. But since the PMLA involves no assignment of a case from a magistrate to a sessions court, the Court in Sarla Gupta noted that the Cr.P.C. complaints' clause on disclosure would not strictly apply to the PMLA [Paragraph 30]. 

It could have deemed the 'complaint' filed by the Enforcement Directorate as a 'police report' or adopted some other means to hold that the rules of Section 207 Cr.P.C. applied to PMLA complaints. If it would have done so, the Court would have thrown open the doors to topple the investigative scheme of the PMLA, because it hinges upon treating these complaints by state agencies as not being akin to police reports. There are many investigative powers that flow to the Enforcement Directorate (and other agencies) from simply not being labelled 'police'. Any ruling that dents this scheme would be extremely troubling for central investigative agencies which file 'complaints'. 

The Court was certainly aware of this position, which had been restated a few months ago by a bench comprising the Chief Justice. As a result, rather than apply either Section 207 or Section 208 by twisting their language, the Supreme Court in Sarla Gupta conveniently holds that the 'principles' of both clauses apply to the PMLA context, and all relied upon materials should be supplied [Paragraph 30]. 

Thus, the answer to issue 2 is a 'yes', but it is more an obligation on part of the agency and court rather than an issue that the accused must chase by enforcing their rights. There is a right too, just to be clear, and issue 2A stands answered by telling us that this right can be enforced from the time that an accused is first summoned to court to face a trial under the PMLA. If any deficiency exists in terms of material that the agency relied upon, the accused can seek copies, and arguments on the aspect of charge should not start till this supply of material is complete.

Issues 3 and 3A

Which then brings us to the last issues of the lot — issue 3 and 3A. That an accused has a right to seek the copy of un-relied upon documents has been an unequivocally clear position now for quite some time, and the Court merely reiterated precedent to stress upon this aspect [Paragraphs 31-33]. Issue 3A is where the controversy existed — at what stage would this right to get un-relied upon documents trigger? This section of the judgment is couched with qualifiers and it is safer to first set out the general position that the Court has identified, before delving into the devilish details:

• There is a clear right to be supplied with a list of the un-relied upon material, and this must be given along with the entire copy of the PMLA complaint i.e. prior to arguments on charge [Paragraph 41].
• There is no clear right to seek copies of un-relied upon material at the time of arguments on charge [Paragraphs 33-41].
• There is a clear right to seek copies of un-relied upon material during defence evidence by moving an appropriate application for summoning such records [Paragraphs 42-51].
• There is a clear right to seek copies of un-relied upon material during bail hearings by moving an appropriate application, and the extent of the right depends upon the stage at which the investigation is [Paragraphs 52-54].

Out of these four conclusions, the first brooks no exception nor generates any controversy. The Supreme Court has endorsed this approach of directing the agency to prepare a list of un-relied upon materials that must be supplied as part of a set of guidelines for better functioning of trials issued in 2021. Sarla Gupta has extended that logic to the PMLA as well. 

The 2021 judgment did not specifically say anything about the stage at which the accused can seek copies of the un-relied materials cited by an agency. This aspect was touched upon in two subsequent decisions — Manoj [2022 INSC 606] and Ponnusamy [2022 INSC 1177]. Out of the two, Manoj did not specifically mention a stage either and generally observed that disclosure of the un-relied upon materials was necessary to secure a fair trial. It was only Ponnusamy which specifically dealt with this exact issue — the two opinions from the bench diverged on precisely this aspect of a right to un-relied upon documents and the stage at which it would trigger. 

Curiously though, Ponnusamy is not mentioned anywhere within Sarla Gupta. This is a problem because it would seem that Sarla Gupta has diverged from the view taken in Ponnusamy. Both Ponnusamy and Sarla Gupta hold that there is no right to seek copies of un-relied upon material at the stage of charge. But where Ponnusamy held that the right would trigger at trial and that a court could consider the relevance of the material sought to be summoned while considering the request, Sarla Gupta has specifically linked the right to defence evidence, not trial generally, and has limited the scope of discretion vesting with a judge to refuse a request for summoning the materials (permitting it only on grounds of delay). 

It is unclear why Sarla Gupta specifically linked the prospect of seeking un-relied upon materials to defence evidence, and not adopted the view taken in Ponnusamy. The Supreme Court's logic in stressing upon the existence of a right for accused persons is that there is a reverse burden operating after charge is framed under the PMLA. In that case, why specifically link the right to only defence evidence? No reasons are offered to explain why an accused should remain quiet throughout prosecution evidence when having this material would enable a more effective cross-examination of witnesses at the relevant stage. Instead, the Court proposes a tortuous route, suggesting that an accused after applying for this material and obtaining it can then apply for recalling of witnesses to cross-examine them on this material. 

This course of action does not commend itself on logical grounds as mentioned above, nor does it commend itself on grounds of efficiency, economy, or strategy. Accused persons will have to adopt an artificial posture and work out two potential cross-examinations of the same witness. If a second cross-examination yields new incriminatory evidence from the witness, that would necessitate a trial court to put these statements to an accused and record a fresh statement under Section 313 Cr.P.C. These are only surface-level problems, mind you, and I am sure that more emerge the moment that lawyers and courts begin to try and apply the rules of Sarla Gupta to the facts of actual cases.    

We are left with the second and fourth conclusions then — no general right to copies of un-relied material at charge, but a right to apply for such material during bail. Although Ponnusamy is not cited anywhere within Sarla Gupta, it has followed the holding in Ponnusamy that there is no right to seek un-relied materials at the stage of charge by moving applications under Section 91 Cr.P.C. It also relies upon another earlier Three Justices' Bench decision in this context, Debendra Nath Padhi, which held that the hearing at charge must be confined to the record of the case, and this record consisted only of material that the prosecution brought forth. But while doing so Sarla Gupta has not slammed the door shut upon potential requests for un-relied upon material at this stage since it qualifies its position by repeatedly stating that 'ordinarily' an accused would not have the right to apply under Section before trial for copies of such materials [Paragraphs 33, 37, 49]. In the right facts, then, one may still have a chance to get un-relied materials prior to framing charge.

What complicates matters a great deal further is the fourth conclusion — that an accused has a right to apply for un-relied upon material during bail hearings. In this section of the judgment, the Court has moved entirely free of any past precedent and taken a view that fairness demands that the accused must have this right because bail under the PMLA places an onus upon the accused to show that no offence is made out. To ensure a fair opportunity for an accused to discharge this burden, the Court felt it was necessary to have a right to seek un-relied upon material which the agency might have.

This section greatly complicates the task of making sense of Sarla Gupta. Firstly, because it further undermines the already logically precarious position to allow a right to apply for un-relied upon material at the stage of defence evidence alone, since the reverse burden operates at the time the trial begins. Secondly, because if the reverse burden on an accused is such a critical hit to fairness, then why should an accused have to suffer framing charge on the basis of potentially unfair prosecutorial conduct and then get a valuable right to expose this during trial? Thirdly, and finally, because ordinarily bail will be a stage that comes before charge is framed, and if a court ends up viewing the un-relied material as being relevant to deciding the bail application it will undoubtedly end up thinking it is relevant to decide the point of charge as well. This creates a weird and inconsistent position where persons in custody may have greater opportunities to fair hearings at the stage of charge than those who are never arrested during investigation.

Sarla Gupta cites many past decisions of the Supreme Court on this issue of disclosing un-relied upon materials by prosecuting agencies. A unifying theme through all those past decisions was clearsighted acknowledgment that there was a need for a right of disclosure to mitigate against the very real possibility of unfair suppression by prosecuting agencies of material gathered in an investigation but of an exculpatory nature. But in Sarla Gupta the issue of suppression is nothing but an elephant in the room. Throughout its 86 pages, not once does Sarla Gupta mention this aspect of possible suppression and the unfair prejudice it causes, even though it is clearly the driving force behind the Court conferring a right to un-relied upon material at the bail stage. A full and frontal acknowledgment of this issue being one of the guiding reasons to confer a right of disclosure may well have yielded different configurations of how the right operates.

Conclusion

To conclude, then, One is left feeling that Sarla Gupta is yet another one of those judgments in which a bench of the Supreme Court is torn between adopting fairness as a yardstick for reading rules of criminal procedure, but is unable to adopt this yardstick consistently when faced with the ghost past judicial precedent that has taken a restrictive and unfairly textual view of the statute. Even so, the judgment is bound to be immensely relevant not just for PMLA cases, but criminal trials generally, and criminal trials with reverse onus clauses more specifically. One cannot help escape the feeling that this is not the end of the matter, but only the start of new complexities in the law on disclosure obligations and rights of accused persons. 

The Promise of Imran Pratapgadhi

Somnath Lahiri, the Communist Party representative in the Constituent Assembly of India, is not the name that comes first to mind when one thinks of the fundamental rights guaranteed by the Constitution which that august house was formulating. Yet, it is his canny observation in 1947 which many return to when assessing the life of the fundamental rights clauses in the Indian Republic: That the fundamental rights chapter appeared to have been drafted from the point of view of a police constable. Each right guaranteed by the Constitution came with a qualifier, which meant that while the supreme 'people' of India had many rights, their nascent Indian State decided whether or not the exercise of such rights was lawful or not in any given set of circumstances. Creating a system where the State could contest the legality of how people exercised their rights had the in-built potential for rendering them a dead letter. Parliament could outlaw certain brands of activity, the police could swoop in and forcefully stop persons from engaging in not only such activity but anything that was remotely prejudicial, and by the time the aggrieved persons secured a vindication from courts it would be rendered meaningless by the efflux of time.   

If Imran Pratapgadhi was not a member of the Rajya Sabha (Council of States), perhaps his struggle in exercising his fundamental right to freedom of speech would have played out across the template sketched above. But he is, and thus was able to successfully challenge a case registered by the police against him in December 2024 before the Supreme Court, which in March 2025 set aside the case and in doing so has delivered a judgment of great significance. The Court in Imran Pratapgadhi v. State of Gujarat & Anr. [2025 INSC 410] begins by bleakly noting that "even after 75 ears of the existence of our Constitution, the law enforcement machinery of the State is either ignorant about this [the fundamental right to freedom of speech and expression] or does not care for this fundamental right." An honest effort at implementing the approach outlined in this judgment could bring about some change in this undesirable status quo.

But I get ahead of myself. What had happened in Imran Pratapgadhi? The petitioner had shared a video of a mass wedding program on social media, and the background audio to this video included a poem recited by him. This poem allegedly incited people of different communities against one another and created enmity and hatred between the groups. An FIR (first information report) was registered for the alleged commission of offences under Sections 196, 197(1), 302, 299, 57, 3(5) of the Bharatiya Nyaya Sanhita 2023 [BNS], i.e., doing acts to cause enmity or hatred between groups; making imputations prejudicial to national integration by words or signs; insulting religious beliefs; wounding religious feelings, and so on. The Petitioner challenged this FIR before the High Court, which refused to entertain his plea stating that the investigation was at an initial stage. The Supreme Court, as we already know, had no such qualms in setting aside what it viewed as a patently illegal exercise of power: 

"The poem does not refer to any religion, caste or language. It does not refer to persons belonging to any religion. By no stretch of imagination, does it promote enmity between different groups. We fail to understand how the statements tehrein are detrimental to national unity and how the statements will affect national unity."

The Court could have stopped here — at page 16 of 54 — but it ploughed further to deal with procedural law issues surrounding the registration of criminal cases. The issue arose thus: If police is obligated to file a case upon receiving complaints with allegations of cognizable offences being committed, then how can it be argued that the registration of this case was improper? To determine whether the allegations are made out or not requires an investigation which can happen after registration of a case. Throwing out cases at the initial stage by analysing the merits of the allegations would place the cart before the horse. This is a rather overstated position, mind you, because the Supreme Court itself adopted an equivocal position on whether instant registration of a case is in fact mandatory upon receiving complaints disclosing cognizable offences. This judicial equivocation in the context of the old Criminal Procedure Code 1973 [CrPC] has been crafted upon the statute itself by Section 173(3) of the Bharatiya Nagarik Suraksha Sanhita 2023 [BNSS], which permits a preliminary inquiry into the allegations before registering cases of a comparably lesser gravity. 

Where Imran Pratapgadhi advances the law on this aspect is, first, in its realistic treatment of the process of registering a case. When we speak about police receiving complaints alleging commission of offences, we may imagine simple scenarios such as X hitting Y with a stick. However, allegations revolving around appreciation and interpretation of speech acts is qualitatively different from this relatively straightforward case of bodily harm. Here, as the Court explains, the officer "will have to read or hear the words ... the officer must consider the meaning of the spoken or written words." This, the Court clarifies, is not even a preliminary inquiry, but a process inherent to the very determination of whether a cognizable offence has been committed or not. This intervention appears simplistic but it is far from it. Rarely does the Court deal so candidly with the processes by which police exercise power. 

Thus, in Imran Pratapgadhi, the Court unwinds this most basic and simultaneously most awesome of powers wielded by police in our system — to transform a narrative complaint into a criminal case. Sheer candidness alone is not what makes Imran Pratapgadhi an important decision. After having unwound the fabric of police power, the Court proceeds to clarify the standards which the Constitution demands of the police in how it exercises its power. It holds that wherever fundamental rights under Article 19 are at stake and the alleged offences meet the contours of Section 173(3) BNSS the police must conduct a preliminary inquiry. In this inquiry, or even in an investigation, police must be mindful that the "effect of the spoken or written words will have to be considered based on standards of reasonable, strong-minded, firm and courageous individuals and not based on the standards of people with weak and oscillating minds. The effect of the spoken or written words cannot be judged on the basis of the standards of people who always have a sense of insecurity or of those who always perceive criticism as a threat to their power or position."

The third prong of its significance is that Imran Pratapgadhi also offers guidance on how constitutional courts must address such cases. It notes that it is the "bounden duty" of courts to ensure that the "ideals of the Constitution are not trampled upon. ... The courts must not be seen to regulate or stifle the freedom of speech and expression. As a matter of fact, the Courts must remain ever vigilant to thwart any attempt to undermine the Constitution and the constitutional values, including the freedom of speech and expression." In this vein, the Court clarifies that there is no barrier to quashing cases where no offence is made out, even at the earliest stages after registration of an FIR.

To conclude, then. 75 years on from the adoption of the Constitution, the Supreme Court of India appears aghast at how the police seemingly dealt with a citizen's exercise of his fundamental right to freedom of speech and expression. A more cynical observer might remind the Court of Somnath Lahiri's remarks and ask that if we gave to ourselves a police constable's Constitution, is it any surprise that the constables have run riot with it? That there is still some life left in the fundamental right to freedom of speech and expression for citizens of India is perhaps in spite of, and not because of, the text of the Constitution and its police constable's perspective of rights. Upturning the police constable's constitution will take much more than pious observations from the bench, especially at a time when national tempers are rising. It will take constant vigil from all corners, including from all courts, to ensure that the ideals which the Supreme Court hails as cherished by the Indian Republic are not emblems for securing convenient speech of desirable citizens, but the constants equally protecting the inconvenient speech of supposedly undesirable citizens as well.

IBC Interim Moratoriums and Cheque Bounce Cases

On 01.04.2025, the Supreme Court rendered its judgment in a batch of cases [decision reported as Rakesh Bhanot v. M/s Gurdas Agro Pvt. Ltd., 2025 INSC 445 ("Rakesh Bhanot")]. The common issue across the cases concerned the interplay between cheque bounce cases under the Negotiable Instruments Act 1881 [NI Act] and the Insolvency and Bankruptcy Code 2016 [IBC] provisions on moratoriums pending the resolution of the insolvency process for persons and firms. 

The problems begin with how the issue is framed at paragraph 4 of the judgment: 

"The common legal question that arises for consideration herein is, whether the proceedings initiated against the appellants / petitioners under Section 138 read with Section 141 of the N.I. Act, 1881 should be stayed in view of the interim moratorium under Section 96 IBC having come into effect upon the appellants / petitioners' filing applications under Section 94 IBC. In view of the commonality of issues involved in all the cases, we need not necessarily review the facts of each case individually"

If we read this paragraph without an inkling of the facts, it is reasonable to think that Rakesh Bhanot dealt with general applicability of the moratoriums for persons and firms under Sections 96 / 101 IBC to proceedings for cheque bounce cases. This impression, however, is entirely wrong. Rakesh Bhanot was dealing with a specific and niche issue that the petitions presented. It was not at all a question of deciding the general applicability of the IBC moratorium clauses for personal insolvency to cheque bounce cases. Rather, I would frame the issue as follows:

Company X is arraigned in a cheque bounce case under Section 138 NI Act, and Y is arraigned in his capacity as its director. Company X moves insolvency proceedings for corporate debtors under IBC, and separately, Director Y moves personal insolvency proceedings in his capacity as a personal guarantor of the company. Can Director Y use the interim moratorium triggered by his filing an application of personal insolvency, to stall proceedings of the cheque bounce case filed against Company X and him?

This issue is an offshoot of the Supreme Court's earlier decision in P. Mohanraj [(2021) 6 SCC 258 (Three Justices' Bench)] where it had to decide whether a cheque bounce filed against a company would be hit by the IBC moratorium provisions. P. Mohanraj held that the moratorium provisions for corporations covered cheque bounce cases. The key element of the Court's reasoning was that a cheque bounce case is, essentially, a proceeding for recovering debts owed to a person. Even though it may potentially involve a jail sentence, cheque bounce cases could be seen as 'quasi-criminal' at best. 

Crucially, though, the Court specifically held that the moratorium would not apply to the persons arrayed in their capacity as directors / officers of the corporate debtor itself. In other words, while proceedings in cheque bounce cases against a company undergoing insolvency would be halted against the company during the insolvency process, they could continue against its directors. The findings on this aspect in P. Mohanraj are, unreasoned, to say the least. Nevertheless, they have not been challenged. Instead, it would seem that litigants across the country devised a new approach to try and use IBC moratorium clauses to stall cheque bounce cases in which they were arraigned as directors of companies. 

This approach was to file for personal insolvency under Section 96 IBC which has a separate moratorium process. A look at the clause makes it clear that this route could not be availed so simply. Section 96 IBC halts any pending proceedings in respect of a debt of the person filing for insolvency. In a cheque bounce case against the company, the 'debt' would be that of the company which issued the cheque, and not the director. The liability of the director is purely vicarious, flowing from his position in the company as an officer responsible for the company's affairs. 

This is where the guarantor arrangement is relevant. Section 126 and Section 128 of the Indian Contract Act 1872 when read together explain that the liability of a surety (the person giving the guarantee) is coextensive with that of the principal debtor (here, the company). So in situations where a director stands as a guarantor for debts of the company that the company pays by way of a cheque which is dishonoured, one cannot separate the debt in respect of which a cheque is issued as being that of the company alone, and it must be seen as also being a debt of the director who stands as guarantor in the contract. If the debt is equally that of the guarantor, then it would bring us back within the fold of Section 96 IBC, and the moratorium would have to apply.

How has Rakesh Bhanot resolved the issue? Deeply unsatisfactorily, to say the least. The Court notes at Paragraph 10.1 that the moratorium under Section 14 IBC for corporate debtors "is not available to the surety or in other words, to a personal guarantor" but that is merely stating the obvious since none of the pleas invoked Section 14 to begin with but invoked Section 96 IBC to make their case. On that score, the Court notes in the same paragraph that "[T]he use of the words 'all the debts' and 'in respect of any debt' in sub-section (1) of Section 96 is not without a purpose, as the moratorium is intended to offer protection only against civil claim to recover the debt. Hence, such period of moratorium prescribed under Section 14 or Section 96 is restricted in its applicability only to protection against civil claims which are directed towards recovery and not from criminal action." Through Paragraphs 10 to 13 the Court hammers down this nail of its reasoning, emphasising that a cheque bounce case is not a civil claim for debt recovery but a criminal case, and notes the deterrent effect of this criminal case in Paragraph 17 to hold that a person should not be allowed to evade liability by using IBC moratorium clauses. 

In other words, rather than explain to us whether the debt in question could be that of the surety to extend Section 96 IBC to cheque bounce cases against the director, or offer any argument of public policy that militates against extension of Section 96 IBC, what the Court in Rakesh Bhanot has held is that Section 96 IBC cannot be applied to cheque bounce cases at all. This finding is directly contrary to the judgment of a larger strength in P. Mohanraj which held that moratoriums under both Section 14 and Section 96 IBC would extend to cheque bounce cases because they were "'civil sheep' in a 'criminal wolf's' clothing". In fact, P. Mohanraj observed that the language of Section 96 IBC covered cheque bounce cases against persons / firms, and its used this as the starting point for its conclusions that the Section 14 IBC moratorium for companies also covers cheque bounce cases. 

None of these observations in P. Mohanraj are dealt with or mentioned by the bench in Rakesh Bhanot through paragraphs 10-13. The 2021 decision is only referred to is in paragraph 14 by citing the extract from P. Mohanraj that denies extension of Section 14 IBC moratorium to the directors of a company in cheque bounce cases, which has been followed by subsequent cases. Again, extending Section 14 IBC to directors was not the issue. It was only the starting point for understanding the issue, which concerned the impact of the coextensive nature of debts under guarantee contracts on the moratorium provisions of the IBC. This issue has not been touched, let alone discussed, in Rakesh Bhanot.

Given how difficult it is to seek a review of a verdict, one wonders whether this spells the end for what is an interesting legal issue that could benefit explanations through a reasoned judgment. Even so, this entire branch of litigation appears to be the result of loopholes in the IBC regime on moratoriums in personal insolvency cases. Section 96 IBC triggers an 'interim moratorium' upon the filing of an insolvency plea, and this interim moratorium remains in force till the application is 'admitted' by the authority. Compare Section 96 IBC and its 'interim moratorium' with the actual 'moratorium' governed by Section 101 IBC post admission of the plea, and you can see the loophole. Where Section 101 limits a moratorium to 180 days after the application is admitted, there is no time limit to how long interim moratoriums can subsist, because there is no time limit to how long it can take for the application to be 'admitted'. Since the interim moratorium remains in force till the application is admitted, enterprising litigants will naturally try and delay things at this stage itself. This is a problem with Section 96 / 101 IBC which the legislature really ought to address, which might clean up a lot of issues in pending litigations.

Paper Update: Confronting the Violence Within

At the start of the month, a bench of the Supreme Court dismissed a plea for anticipatory bail filed by a person accused of corruption [Devinder Kumar Bansal v. State of Punjab (2025 INSC 320)]. The person in question was accused of asking for bribes to do audit work, and now sought relief from court. For the uninitiated, 'anticipatory bail' is a remedy where a person goes to court claiming he entertains a reasonable apprehension of arrest in connection with a case, and if the court agrees, then it can pass orders directing that in the event of his arrest the police should release him on bail. 

Even a mildly critical reading reveals the problem. If all we need is reasonable apprehension that there may be an arrest, where does that leave the police's power of arrest and the consequential 'benefits' that arrests and custodial questioning are seen to bring to the pursuit of justice? It can bring it to an absolute standstill. Which is why no court since the introduction of anticipatory bail has read the provision strictly, and the law since 1973 is a study in seesawing judicial opinions on how to exercise this great discretionary power that the criminal procedure code had conferred upon judges. 

Devinder Kumar Bansal falls in the bucket of opinion which warrants extreme circumspection in the grant of anticipatory bail. It notes that the relief "can be granted only in exceptional circumstances where the court is prime facie of the view that the applicant has been falsely enroped in the crime or the allegations are politically motivated or are frivolous." Note that none of this is statutory criteria by the way — something made even more interesting by the fact that in Pakistan, law specifically made testing the falsity of allegations something a court must test while deciding anticipatory bail pleas. 

What makes Devinder Kumar Bansal even more interesting, and allows me to use it as a hook to shamelessly plug a small essay that has come in a volume titled Police Violence in India, is a one-line observation that it made in passing, stating that "if liberty is to be denied to an accused to ensure corruption free society, then the courts should not hesitate in denying such liberty. [Emphasis supplied]" No matter the precepts that we should not turn persons into tools for achieving some unknown larger aims — such as a corruption free society — the observation is deeply revealing of the consistent tendency of courts to emphasise upon custody as a means to secure justice. Especially during investigations. 

In an essay titled 'Confronting the Violence Within', I've taken up this tendency of courts to make interpretive choices that privilege custodial interrogation and questioning in criminal cases, which only entrenches the importance of the custodial setting within the law. And courts conveniently want to only focus our attention on the good parts — custodial interrogation is more 'elicitation oriented' as they say — while turning a blind eye to what makes the custodial setting so unique for the investigator — its inherently violent nature. The paper looks at this tendency through two specific areas — law on the right against compelled self-incrimination, and law on anticipatory bail. While Devinder Kumar Bansal has been published after the paper and is relevant to its arguments from an anticipatory bail perspective, even for the law on self-incrimination there have been interesting developments, as the Supreme Court in granting some reliefs to Ranveer Allahabadia specifically directed him to be questioned without his lawyer present. The direction is not unique and in line with previous decisions by the Supreme Court which have consistently sought to remove any intermediary from the custodial setting to ensure its 'uniqueness' — again, without ever squaring up to face the violence which is what makes the custodial setting unique.

If the courts themselves places such a premium on the need for custodial interrogation, how can we expect custodial violence which is an inherently inseparable part of the process, to vanish? 

If liberty is to be denied to an accused to ensure corruption free society, then the courts should not hesitate in denying such liberty. Where overwhelming considerations in the nature aforesaid require denial of anticipatory bail, it has to be denied. It is altogether a different thing to say that once the investigation is over and charge-sheet is filed, the court may consider to grant regular bail to a public servant - accused of indulging in corruption

Anticipatory bail can be granted only in exceptional circumstances where the Court is prima facie of the view that the applicant has been falsely enroped in the crime or the allegations are politically motivated or are frivolous. So far as the case at hand is concerned, it cannot be said that any exceptional circumstances have been made out by the petitioner accused for grant of anticipatory bail and there is no frivolity in the prosecution.