Guest Post: Beyond the 'Pin Drop': Rethinking Tech-Enabled Bail Conditions

(This is a guest post by Sarthak Gupta and Tarun Agarwal)

On July 8th, 2024, the Supreme Court delivered its decision in Frank Vitus v. Narcotics Control Bureau, (2024). It held that the bail condition directing the accused to drop a PIN on ‘Google Maps’ to ensure that their location is available to the investigating officer is unconstitutional as it violates the right to privacy under Article 21 of the Constitution. The decision implicitly endorses the proportionality standard for bail conditions and critiques the court’s past rulings that imposed similar onerous conditions, marking a significant step in protecting individual freedoms. At the same time, the Court’s reasoning falls somewhat short of delivering a transformative interpretation of the law. As I argue here, it could have adopted a different line of reasoning to reach its conclusion, potentially leading to a binding precedent and more robust development of privacy jurisprudence.

Courtroom Kaleidoscope, Bail Bedrock

The trend of imposing electronic monitoring and GPS tracking as bail conditions is evident across various courts in India, demonstrating a consistent yet potentially arbitrary approach to bail jurisprudence. The SC in Vernon v. Union of India, while granting bail to Vernon Gonsalves and Arun Ferreira in the Bhima Koregaon case, mandated 24-hour GPS monitoring and phone pairing with the investigating officer. A similar condition was imposed on another Bhima Koregaon accused, Shoma Sen. The SC has followed such arbitrary conditions in a bail order in the case of State of Uttar Pradesh v. Maulana Kaleem Siddiqui, allowing the accused to use only one mobile phone with location setup open so that his location could be traced at any point in time by the investigating agency.

This pattern extends to HCs. The Delhi HC in Neha v. State (NCT of Delhi) imposed a condition that the Applicant was to ‘drop a pin’ on Google Maps so that the SHO/IO can verify the presence and location. In State of NCT of Delhi v. Sanjeev Kumar Chawla it issued similar conditions noting that ‘[d]igital and electronic equipment, as presently used in America, ought to be introduced in India, so that a tracking system similar to the GPS Tracking System, can be used to monitor the movement of the accused released on bail, allowing the authorities to gather information all the time while permitting the accused to undertake the usual and ordinary activities of normal life.’ From what I could gather, the Delhi HC has regularly imposed such a condition in 2024: more than 60 cases  came up during my research.

The use of such conditions by pointing to foreign jurisdictions fails to consider that in other jurisdictions, there are specific statutes that allow such conditions to be imposed. For instance, in the UK, electronic monitoring is permitted under the Terrorism Prevention and Investigation Measures Act, 2011. In the USA, electronic monitoring is permitted under the Adam Walsh Child Protection and Safety Act, 2006, which allows for electronic monitoring as a condition of supervised release or probation for certain sex offenders. In Malaysia, legal frameworks on electronic monitoring have been developed by amending existing legislation and enacting new laws, such as the Prevention of Crime Act of 1959, the Security Offences Act, 2012, the Dangerous Drugs (Special Prevention Measures) Act,1985, and the Criminal Procedure Code. In Canada, the Criminal Code of Canada allows for the use of electronic monitoring as a condition of parole, conditional release, or probation for certain offenders. In Australia, the Crimes Act 1914 and various state-level laws permit the use of electronic monitoring for certain offenders. In India, there is no similar provision anywhere, and nor do the new criminal laws provide for it.

The Punjab and Haryana HC has also been particularly stringent, as seen in cases like Samar Gakhar v. State of Punjab, and Hussain Abbas alias Tippu v. State of Haryana wherein it not only mandated GPS tracking but detailed smartphone usage rules, including restrictions on clearing history and formatting devices. Similar conditions were imposed in many other bail matters including Gursharanjit Singh @ Sunny v. State of Punjab and Umed Singh v. State of Haryana (all by the same bench). The arbitrary nature of these impositions was escalated by the HC of Jammu and Kashmir when in a UAPA case it became the first to mandate the wearing of a GPS-enabled tracking device as a bail condition.

This diverse yet consistent application of technologically invasive bail conditions across different courts and jurisdictions underscores a growing trend in Indian Tech-Enabled Bail Jurisprudence. While the conditions aim to ensure compliance with bail terms, their widespread and often indiscriminate application raises significant concerns about the balance between judicial discretion and fundamental rights, particularly in the context of privacy and the presumption of innocence.

The Limits of Judicial Review and Unaddressed Legal Questions

 

The judgment in Frank Vitus noted that a GPS tracking condition had been imposed by the Supreme Court in bail orders earlier, but refrained from commenting further, noting that it was ‘not called upon to decide the issue of the effect and legality of such a condition’. This rather cautious approach introduces a problematic ambiguity into the jurisprudence on such bail conditions. By acknowledging that similar conditions may have been imposed in other cases without going further, the Court implicitly admits to a pattern of potentially unconstitutional practices that have gone unchallenged. This underscores a systemic issue where rights-infringing conditions may be, and arguably are, routinely imposed without proper legal examination, allowing unconstitutional practices to persist.

 

Perhaps most problematically, Frank Vitus does not definitively settle the issue either even after raising it and specifically dealing with GPS tracking conditions. By focusing on the technical ineffectiveness of the bail condition in this particular case and the potential impossibility of obtaining an embassy certificate, rather than providing a comprehensive constitutional analysis, the Supreme Court leaves the door open for similar conditions to be imposed in future cases under different circumstances. The Court's failure to establish clear, binding principles regarding the use of surveillance technology in bail conditions is a missed opportunity. Instead of setting a precedent that the rest of the judicial pyramid can follow, the judgment effectively delegates decision-making to a case-by-case basis. This approach will only perpetuate the currently arbitrary use of such conditions (and arguably other conditions as well). Further, by not definitively ruling on constitutionality of such surveillance-based bail conditions in general, the Court has potentially prolonged legal uncertainty. This ambiguity may encourage law enforcement and prosecutors to continue pushing for such conditions, leading to further litigation and potential rights violations before the issue is conclusively settled.

Road Not Taken

Building on the limitations of the Court's approach, a more decisive and comprehensive ruling on the legality of surveillance-based bail conditions would have strengthened the judgment's impact and provided clearer guidance for future cases. The Court could have established clear principles by explicitly addressing the constitutional dimensions of such conditions to allow for more consistent assessment rather than courts merely invoking privacy and then being left to do as they please. For example, a detailed proportionality analysis, combined with an assessment of the 'tripod test' of risks, would be necessary. This integrated framework, drawing on established principles, offers a systematic method for determining whether an infringement on fundamental rights can be justified within the context of bail conditions.

Proportionality analysis involves a four-pronged test: Firstly, the condition must pursue a legitimate aim. While ensuring the accused's appearance in court and preventing flight risk are undoubtedly legitimate objectives of bail conditions, the question to be answered (and left unanswered in Frank Vitus) is how continuous GPS tracking specifically advances these aims beyond traditional methods. Secondly, the suitability prong requires that this tool be rationally connected to the objective. Chandrachud J. (then) in Mohd. Zubair v. State of NCT of Delhi while referring to Parvez Noordin Lokhandwalla v. State of Maharashtra noted that the bail conditions must not only have a nexus to the purpose that they seek to serve but must also be proportional to the purpose of imposing them, ensuring a balance between the liberty of the accused and necessity of a fair trial. Here, the technical limitations of the "drop a PIN" feature, as elucidated by Google's affidavit, cast doubt on whether the condition could effectively achieve its purported aim.

The third prong of necessity demands that the measure must be the least restrictive means available to achieve the objective, which Frank Vitus could have more thoroughly explored alternative, less intrusive measures such as periodic check-ins or travel restrictions, which might adequately serve the same purpose without the pervasive privacy implications of constant location tracking. Finally, the last prong of balancing requires a careful weighing of the degree of rights infringement against importance of the objective. Continuous GPS surveillance represents a significant intrusion into the accused's privacy, liberty, and freedom of movement—rights retained by those on bail. The Court's cursory treatment of this balancing leaves much to be desired, particularly given the presumption of innocence that should inform pre-trial measures.

Moreover, the proportionality test could have been applied in conjunction with the 'triple test' of risks traditionally considered in bail i.e. i) Risk of reoffending; ii) Flight risk, and iii) Risk of tampering with evidence and influencing witnesses. The Supreme Court could have thus established a comprehensive framework for evaluating surveillance-based bail conditions requiring courts to:

1. Identify the specific risk(s) the condition aims to address (from the triple test).

2. Assess whether the surveillance measure is legally sanctioned and pursues a legitimate aim.

3. Evaluate if the measure is suitable for addressing the identified risk(s).

4. Determine if less intrusive alternatives could effectively mitigate the risk(s).

5. Balance the degree of intrusion on the accused's rights against the importance of addressing the identified risk(s).

This integrated framework would provide a more nuanced and rights-respecting approach for imposing bail conditions. It would ensure that surveillance measures are employed when strictly necessary and proportionate to the risks, thereby safeguarding the accused's fundamental rights while addressing legitimate concerns of the criminal justice system. Such a framework could have offered lower courts with a clear, constitutionally grounded approach, to evaluate proposed surveillance-based bail conditions, promoting more consistent and rights-respecting decision-making across the judicial system.

A forward-looking judgment could have addressed not just the specific GPS condition at issue, but also set guidelines for the use of other current and potential future surveillance technologies in bail contexts. Finally, the Court could have emphasized the exceptional nature of surveillance-based conditions, explicitly stating that they should be considered as a last resort when less intrusive measures are demonstrably insufficient. This would have reinforced the presumption of innocence and the principle that bail conditions should be least restrictive, necessary to ensure the accused's appearance and the integrity of the judicial process.

Notes from foreign fields:

Use of electronic monitoring as a pretrial condition has become increasingly prevalent in various jurisdictions, particularly the United States. Courts have imposed electronic monitoring as an alternative to detention, with the number of individuals subject to such monitoring growing significantly over the past decade. For instance, in 2021 alone, 254,700 individuals were subjected to electronic monitoring, including 150,700 under the criminal legal system and 103,900 by U.S. Immigration and Customs Enforcement. Cities like Indianapolis and San Francisco have seen their rates of pretrial electronic monitoring double or triple in recent years.

American Courts have generally relied on two primary rationales for imposing electronic monitoring: ensuring public safety and preventing flight. However, the effectiveness of electronic monitoring in achieving these goals has been questioned. Studies have shown mixed results, with some early research indicating worse pretrial outcomes for those on electronic monitoring. More recent studies, such as those conducted in the Federal District of New Jersey and California's Santa Clara County, have shown limited positive results, but neither demonstrated overall success in keeping individuals out of detention during the pretrial period. 

Several courts have critically examined the appropriateness of electronic monitoring. In Commonwealth v. Norman (2020), the Massachusetts Supreme Judicial Court found that the causal link between GPS monitoring and ensuring court appearances was too attenuated and speculative to justify its use. Similarly, in State v. Grady (2019), the North Carolina SC ruled that a lifetime electronic monitoring condition was unreasonable under the Fourth Amendment due to a lack of evidence of its efficacy. These decisions highlight a growing recognition that electronic monitoring may not be proportional to the risks it aims to address and that less restrictive alternatives should be considered. The European Court of Human Rights has also weighed in, stating that while GPS surveillance may interfere less with privacy than methods like telephone tapping, it still constitutes a significant intrusion that must be justified and proportionate to the alleged offence and assessed risk.

Conclusion

While the Supreme Court's decision in Frank Vitus represents a step forward in protecting privacy rights within bail conditions, it falls short of establishing a comprehensive framework for evaluating surveillance-based measures. The Court correctly deemed GPS tracking conditions unconstitutional, but its narrow focus on technical ineffectiveness meant it missed an opportunity to provide broader guidance to courts. A more robust approach, incorporating the Puttaswamy proportionality test and the traditional 'triple test' of bail risks, could have offered clearer direction for lower courts and addressed the wider implications of surveillance technology in criminal justice.

Furthermore, this case highlights the urgent need for legislative reform. The Court cannot continue to rely solely on purposive interpretation in the absence of specific laws, especially as other jurisdictions have already enacted such legislation. Parliament should consider introducing similar enactments to provide a clearer legal framework for use of surveillance technologies. As these technologies become increasingly prevalent globally, this case underscores the ongoing challenge of balancing public safety with fundamental rights in the digital age.

The Prem Prakash Bail Order

The verdict is out on what is to be made of the judgment by the Supreme Court in Prem Prakash v. Union of India [SLP (Crl.) 5416 / 2024, decided on 28.08.2024 ("Prem Prakash")]. The reason it is attracting so much attention is because of the Court's views on how to treat statements given by persons in custody in a PMLA case, which it expressed while granting bail to the petitioner. These views extend to statements made by the person concerned (Paras 21 to 34) as well as a statement of a co-accused person that is being pressed into reliance (Paras 35 to 37). 

The Holding

The facts are largely irrelevant for this post, and a quick summary would do. The petitioner was alleged to have been the puppeteer pulling the strings behind a number of persons, all towards orchestrating transfer of properties and monies by illegal means. The material used to substantiate these allegations consisted almost entirely of statements recorded by the Enforcement Directorate officials from the Petitioner whilst he was in custody, and statements of other co-accused persons and witnesses. 

In respect of statements by persons concerned, the court appears to have held that: (1) where a person was arrested by the Enforcement Directorate in one case and remanded to judicial custody by a court, the Directorate cannot record statements of that person under Section 50 of the PMLA in a different case without obtaining permission from a court which remanded the person to judicial custody (Para 33); (2) if a person is in custody under PMLA irrespective of the case for which he is under custody, statements under Section 50 PMLA of such a person shall be inadmissible against the maker as such statements will be hit by Section 25 of the Indian Evidence Act 1872 (Paras 27, 32). 

In respect of statements by co-accused persons, the Court has held that such statements would be hit by the rule under Section 30 of the Evidence Act which means that such statements cannot be treated as a piece of substantive evidence against an accused. This rule would apply even for the bail hearing, and will require that the agency fields other material since the statement of a co-accused can, at best, be used to lend assurance to other material (Paras 36, 37).

The Controversy

The only real controversy, I would argue, stems from the second part of the holding in respect of how to treat statements of persons in custody under PMLA. This set of observations lends itself to different readings, ranging from narrowest to broadest. 

The narrowest reading of Prem Prakash, which would be the storm in a teacup view, would be that the judgment emerged from unique facts and its conclusions must remain confined to those facts. Here, The petitioner was already in custody in one PMLA case, his statements were recorded and used in a different case. All that the Court has said is that only in such a situation would it be impermissible to use the statements against their maker. In all other cases i.e., where a person is arrested and questioned in the same case under PMLA, or where a person is in custody because of a non-PMLA case, there would be no such bar. 

The broadest view would argue that Prem Prakash has fundamentally altered the landscape by rendering any Section 50 PMLA statement given in custody after arrest, as a statement which cannot be used against the maker. Such a conclusion relies upon Para 32 of the judgment which begins: "We have no hesitation in holding that when an accused is in custody under PMLA irrespective of the case for which he is under custody, any statement under Section 50 PMLA to the same Investigating Agency is inadmissible against the maker. The reason being that the person in custody pursuant to the proceeding investigated by the same Investigating Agency is not a person who can be considered as one operating with a free mind." A person is not under some unique sense of duress merely because he is already in custody in one case and is now questioned in another, which is why Para 32 is phrased as it is. 

The problem, of course, is that Para 29 of the judgment seems to suggest precisely that a person is under a unique sense of threat: "Is a reasonable inference legitimately possible that, due to the vulnerable position in which the appellant was placed and the dominating position in which the Investigating Agency was situated, in view of the arrest in the other proceeding that, there obtained a conducive atmosphere to obtain a confession?" With due respect, this does not make sense, for the reasons already stated above. It is not the quantity of cases in which I am in custody, but the quality of the custodial experience which is what sets it apart. That is the true import of Justice Krishna Iyer's opinion in Nandini Satpathy, which the bench also invokes in Prem Prakash.

It is not just this obvious sense of internal contradictions that hamper Prem Prakash from achieving its fullest potential as a radical judgment which could force investigative agencies to abandon what appears to be the standard operating procedure to force the accused to make his own noose by recording damning statements which lack any kind of corroboration in material particulars, and then hang the accused with it. A related issue is the judgment's seeming reluctance to cross the rubicon on that extremely familiar issue in Indian criminal law and procedure — how to define 'police officer' for purposes of Section 25 of the Evidence Act. That clause bars evidence of confessions to police officers [discussed at length here] and it has been retained as section 23 in the Sakshya Adhiniyam. 

Prem Prakash refers to the Supreme Court's recent magnum opus on PMLA, the judgment in Vijay Madanlal Choudhary (2022). For a judgment which set out to clarify the law, it did a woeful job, as can be seen from the observations on this issue of whether Enforcement Directorate officers would be 'police officers' for the purposes of Section 25: 

"Ex-consequenti, the statements recorded by the authorities under the 2002 Act, of persons involved in the commission of the offence of money-laundering or the witnesses for the purposes of inquiry/investigation, cannot be hit by the vice of article 20(3) of the Constitution or for that matter, article 21 being procedure established by law. In a given case, whether the protection given to the accused who is being prosecuted for the offence of money-laundering, of section 25 of the Evidence Act is available or not, may have to be considered on case-to-case basis being rule of evidence."

This is remarkably confusing. Either an officer is or isn't a police officer, that is not a 'rule of evidence' that can be left to a case-to-case determination. Nevertheless, Para 24 in Prem Prakash skilfully uses this part to show us that the door was clearly not slammed shut on this issue by Vijay Madanlal, but there were a few cracks left yet. So far, so good. 

What happens next is the problem. Prem Prakash has relied upon a decision by three justices in Raja Ram Jaiswal [AIR 1964 SC 828] to understand 'police officer' under Section 25 in a purposive sense by looking to the kind of powers conferred upon an officer. It failed to mention that after Raja Ram Jaiswal came the Constitution Bench decision in Badku Joti Savant [AIR 1966 SC 1746], which offered a different and arguably narrower view of how to interpret 'police officer' for purposes of Section 25. In this limited view of who is a police officer, Badku Joti Savant argued that it has to be a person who can file a police report at the end of an investigation, something which the Enforcement Directorate officials conspicuously do not exercise under the PMLA. 

In other words, if the Court was taking the plunge on Section 25 in Prem Prakash, it had to set out more clearly how the existing law could be rationalised first with that part of Vijay Madanlal, and then give us a framework for expanding the contours of Section 25 to include Enforcement Directorate officials within it. This is not difficult. The periodic amendments to the PMLA and Section 44 which speaks of 'complaints' filed at the end of investigations, has brought that the PMLA investigative process closer and closer to a traditional police investigation, blurring the distinction between a complaint proceeding and one launched on a police report. For instance, Section 44 today has an explanation which allows 'further investigation' after filing a complaint , allowing 'subsequent complaints' to be filed.  

Thus, even if we cannot close our eyes to the unfortunate holding of Badku Joti Savant (as much as we might want to), it is today not that hard to slot in Enforcement Directorate officials as police officers for Section 25 even with that regressively formalistic framework. Unfortunately, this is left unsaid in Prem Prakash, and it would fall upon subsequent benches confronted with the problem to develop this thought. One would hope that they water the seed and nurture this plant rather than weed it out even before it has had a chance to show the smallest of green shoots. 

Will Prem Prakash prove to be seismic decision, or a mere storm in a teacup? Let us wait and see.

Retrospective Application of BNSS and the Supreme Court Order in 1382 Prisons

Since 2013, the Supreme Court has been seized of a public interest litigation concerning prison conditions [W.P. (Civil) 406 / 2013, In Re Inhuman Conditions in 1382 Prisons]. Over the course of a decade, it has passed a slew of orders which, it hopes, would ameliorate prison conditions. A key priority has been to try and decongest prisons, which are notoriously overpopulated.

In this vein and with this intent, on 13.08.2024, the Amicus Curiae assisting the Court had directed the judges to Section 479 of the Bharatiya Nagarik Suraksha Sanhita 2023 [BNSS]. More specifically, the first proviso therein, which stated that first-time offenders would be entitled to release from custody if they had spent up-to one-third of the maximum possible sentence. This clause, the Amicus submitted, "needs to be implemented at the earliest and it will help in addressing over-crowding in prisons."

The Court asked the Union Government to get instructions, and it heard the case next on 23.08.2024. The Union Government Counsel stated that "instructions" had been obtained from the relevant department to the effect that Section 479 of the BNSS "would apply to all undertrials in pending cases irrespective of whether the case was registered against them before 01st July 2024, the date when the newly minted legislation has come into effect." 

This led the Court to observe that:

"... Having regard to the fact that the substituted provision under the BNSS is more beneficial vis-a-vis Section 436A of the Code of Criminal Procedure, 1973, wherein the period undergone by the first time offender was prescribed as up to half of the maximum period of imprisonment specified for such an offence, this Court had called upon the learned Additional Solicitor General to obtain instructions from the Department and submit a clarification regarding application of the said provision to all undertrials across the country. ... 

In that view of the matter, it is deemed appropriate to direct immediate implementation of Section 479 of the BNSS by calling upon Superintendents of Jails across the country wherever accused persons are detained as undertrials, to process their applications to the concerned Courts upon their completion of one-half/one-third, as the case may be, of the period mentioned in sub-section (1) of the said provision, for their release on bail. This step will go a long way in easing overcrowding in jails which is the primary focus of this Court in the present petition." [Emphasis supplied]

Considering how the issue of BNSS provisions applying to pending cases has been vexing many judges, lawyers, and laypersons, the Supreme Court's intervention is an important moment in the brief history of the new law, and one which should be studied carefully. In this post, I will try to explain why I think the order is arguably a misstep, and ought to be clarified by the Court going forward.

Court Fails to Explain its Reasoning

A previous post offered certain thumb rules on how to navigate the retrospective application of BNSS. The one, basic rule, when it came to applying procedural laws it argued, was that procedural laws by default are applicable retrospectively, unless they disturb vested rights of parties to their detriment. When it came to the BNSS, the legislature had decided to carve out a larger category of exceptions through Section 531 (following the formula adopted in 1973), which specified a swathe of situations where BNSS would not apply in existing cases. Nevertheless, these were exceptions, and show that the BNSS is intended to be made applicable even to pending criminal cases, albeit not at a particular stage of a case and gradually. 

It is important to keep this thumb rule in mind when we consider retrospective application of Section 479 BNSS. The Supreme Court correctly noted that this clause closely corresponded to Section 436A of the erstwhile Criminal Procedure Code of 1973. Both speak about a right of default bail conferred to persons in custody for up to one-half of the maximum possible sentence they can serve. But it appears that neither on 13.08.2024 nor on 23.08.2024 was the next, logical question raised i.e., whether Section 436A was an ordinary procedural law or had it conferred vested rights upon persons which could not be disturbed to their detriment? Surprisingly, both the orders did not refer to Section 531 BNSS either, to try and use the statutory formula for applying its provisions retrospectively. 

On 23.08.2024, as extracted above, the Court noted that Section 479 BNSS appeared to be "more beneficial" than its predecessor, Section 436A CrPC. Thus, even though it did not provide any framework to explain its thinking, one would argue that the Court was following the same approach i.e. considering whether or not the new provision is more beneficial to determine whether it can be made retrospectively applicable. What went unsaid is the Court's appreciation that 436A conferred a vested right, which I would argue is the correct appreciation of that clause. Section 436A conferred an enforceable right upon all persons in custody which began to operate from the first moment in time a person was placed in custody. It was Parliament's belated attempt to reconcile lengthy undertrial detention with the letter of Article 21 of the Constitution, two decades after the Supreme Court had paved the way through Hussainara Khatoon and other cases.  

Before we can discuss whether the Court was correct in its fleeting assessment of Section 479 BNSS being more beneficial, it is important to flag the issue which arises because of the Court's refusal to spell out its reasoning. Today, the lack of detail in the order dated 23.08.2024 can lend itself to a school of thought where all provisions of the BNSS are now sought to be tested on the anvil of their supposed beneficial nature before we can decide whether they are retrospectively applicable. That, I would argue, is incorrect, and contrary not only to Section 531 of the BNSS but also Section 6 of the General Clauses Act 1897. The general rule of applying procedural laws retrospectively has not been altered to such a great extent in the BNSS, and courts hopefully will remain cautious going forward.

Section 479 BNSS is not "More Beneficial" than 436A CrPC

Now, returning to whether Section 479 BNSS is "more beneficial" than Section 436A. Is that true? 

Section 436A CrPC provided default bail to persons held in custody under any law except for offences in which death was one of the punishments, where such persons had spent up to one-half of the maximum period of imprisonment. It also provided that a court may order continued detention for longer than this period, upon giving reasons in writing, but no court could permit the detention of persons for longer than the maximum period of imprisonment provided. Delays caused by the accused would be excluded from computing the period of detention.    

Compared to its predecessor, Section 479(1) BNSS expands the category of offences in which the default bail right will not accrue, specifying that even offences punishable with life imprisonment are excluded, and retains the earlier language about allowing courts to permit longer detentions. Section 479(1) retains the one-half formula, but ameliorates it through a proviso which says that a first-time offender can be released after serving up to one-third of the maximum possible sentence. This is what prompted the Court to view the new law as more beneficial than the old one. 

However, to close our review of Section 479 BNSS only after reading Section 479(1) would mean wishing away the fact that this provision contains two more sub-clauses, which find no mention anywhere in either the order dated 13.08.2024 or 23.08.2024. While Section 479(3) pertains to the enforcement of the right, Section 479(2) is integral to the scope of the right conferred by Section 479(1). It carves out an entirely new category of exceptions to create new situations where the right would not accrue to persons, and states that "where an investigation, inquiry or trial in more than one offence or in multiple cases are pending" against a person, that person "shall not be released on bail" by the Court.     

I would argue that Section 479(1) BNSS read with its new proviso is certainly more beneficial than what was earlier provided under Section 436A. Ordinary rules of statutory interpretation would mean that even if the Union of India did not give these so-called instructions, the beneficial aspects of Section 479(1) had to inure to the benefit of all persons in custody retrospectively who met the criteria. One wonders why the Court even bothered with that exercise which was sheer gloss.

However, at the same time, the whole of Section 479 BNSS is undoubtedly not more beneficial than its predecessor. It obviously worsens the position of persons who are in custody as on 01.07.2024 if applied retrospectively. Simple examples will do — X was in custody for more than 15 years in a case under Section 307 IPC; Y has spent 5 years in custody as of 01.07.2024 in a case where the chargesheet invoked more than one offence but the maximum imprisonment in the offences was 10 years; Z has spent 5 years in custody as of 01.07.2024 in two cases, in both of which the offences involve a highest possible sentence of 10 years. All of them were just about to file for default bail. But because of Section 479 BNSS applying retrospectively, none of them would be entitled to do so, since Section 479(1) deprives any person accused in a case of life imprisonment with the benefit, and 479(2) denies this right to any person in case involving more than one offence or against whom multiple cases are pending. 

Conclusion

What has the Supreme Court done with its orders in 1382 Prisons? It has not merely clarified (which is all that it needed to do) that Section 479(1) BNSS with its proviso applies to everyone in custody even prior to 01.07.2024. Instead, it has conferred its blessing on the retrospective application of the entirety of Section 479 BNSS, and in doing so has unequivocally worsened the position of all persons, in a manner entirely contrary to law. That too in a public interest litigation where nobody potentially affected by this retrospective application of the law was being heard.

There is any easy way out. Even though the terms of order dated 23.08.2024 are broadly worded, nothing apart from Section 479(1) and giving the fullest application its beneficial proviso was in the mind of the Court when it passed these orders. The bench could clarify the order to make it clear that nothing ought to be construed as conferring retrospective application to all of Section 479 BNSS. If not, then it would fall upon a different bench to clarify this issue which will surely come up to the Supreme Court sooner rather than later.

Errantry - Jury Trials

If there is one thing that has consistently piqued my interest and curiosity, and I am sure it is true for many others, it is jury trials in India and their phasing out. I blame it on the Nanavati Case, a perfect blend of all that makes criminal trials irresistible (anyone interested in the story may want to read In Hot Blood).

The story is well known: A naval-officer husband dropped his wife and children to the theatre, went and got himself a weapon from his ship, then started looking about for his wife's paramour in South Bombay. The husband, K.M. Nanavati, went to the house of the paramour, Prem Ahuja. Nanavati went straight for Ahuja's room after confirming he was home, locked it behind him, and then three shots were heard. He left as swiftly as he came, and went straight to the police to turn himself in. Well, almost. 

The case was sensational material for news and even more so for the tabloids which lapped it up from all angles. After a quick investigation, Nanavati was tried before a jury for the murder of Prem Ahuja, which acquitted him. The verdict was sent up to the Bombay High Court by the Sessions Judge. A Division Bench of the High Court set aside the acquittal and held that there was enough evidence to conclude that Nanavati was guilty of murder.

According to folklore, Nanavati's legacy to the Indian criminal process was the eventual condemnation of the jury trial. The last jury trial, which convinced the powers-that-be that a fickle public which returned perverse verdicts contrary to available evidence could not be trusted with dispensing justice. Henceforth, the criminal trial would be the provenance of the trained judge, to ensure such errors are not repeated.

Existing scholarship by James Jaffe and Kalyani Ramnath, amongst others, has already debunked some myths around Nanavati being the 'last' jury trial, as well as it being the determinative factor leading to the demise of that system of being judged by one's peers. In this short post, I want to add to this small but very fascinating area of work by positing two separate points which appear to have not been commented upon.

First, the judgment of the High Court convicting Nanavati of murder is misunderstood as having fired a broadside at the idea of jury trials in India. Sure, it held that the verdict of the jury was perverse, but this was not because of some inherent fault that it found in the system. Rather, both Justices Shelat and Naik in their separate opinions held that one big reason for the jury's incorrect conclusions was that the trial judge had misdirected the jury on several key points of fact and law. In other words, the jury was not per se bad or perverse, but it had rendered a perverse verdict being misdirected on how to treat some crucial parts of the evidence. This, then, makes the verdict a condemnation of the judge as much as the jury, dynamically altering the received wisdom about the trial itself. This re-appreciation of what Nanavati actually held, as opposed to what it has come to stand for, contributes to the laregely untapped area of scholarship on how a nascent Indian state treated the jury trial.

Flowing from the first we come to the second point, which is that even if we assume that the Nanavati jury had deliberately returned a verdict contrary to the evidence, it is arguably unidimensional to view this as a singularly bad outcome. A history of the jury trial in England shows us that the duty of the jury was not to return a verdict that was only in agreement with the law, but also return a verdict which was in agreement with its conscience, and it was the bounden duty of the system to respect such a 'perverse verdict'. In other words, a jury returning perverse verdicts was not a glaring error pointing to a breakdown but a part of the process. It should come as no surprise that the colonial government was not-too-happy in securing the jury trial in its fullest glory for India. Contrary to established practice in England the Indian criminal procedure code allowed presiding judges to disagree with a jury and refer its verdict to the High Court for a final decision (section 307 of the 1898 Code). It was this same procedure which was adopted in Nanavati's case by the Sessions Judge, reflecting its ordinariness from a procedural perspective.

So here is an alternative history in Nanavati. Maybe the system did not flounder in the case any more than it does in any sensational case.