Guest Post: A Conspectus of the Law Regarding Notice Prior to Arrest

(This is a guest post by Sudhanva S. Bedekar)

Whether the Courts are empowered to issue directions to the investigating agencies to give to the accused, a reasonable notice (example 72 hours) prior to arrest is a question which is sought to be addressed in this post. The issue becomes relevant due to a recent pronouncement of the Supreme Court in Vijaykumar Ramchandani v. Amar Sadhuram Mulchandani where the Court has frowned upon such a practice. 

The power to issue such directions could be traced to two provisions of the Code of Criminal Procedure, namely Section 438 and Section 482. While Section 438 grants to the High Court and Court of Sessions a power to direct release of a person on bail in the event of arrest, Section 482 saves the inherent powers of the High Court. 

In the context of Section 438, the said question seems to have been covered by a judgement of the Supreme Court in Union of India v. Padam Narain Aggarwal. It is argued that Padam Narain Aggarwal, insofar as its reasoning and reliance on Gurbaksh Singh Sibbia is concerned, is not decided correctly and needs a re-look.

Revisiting Padam Narain Aggarwal

It is therefore necessary to examine Padam Narain Aggarwal. The case arose from an order of the High Court of Rajasthan which directed Customs authorities to not arrest the Petitioners without ten days prior notice in case any non bailable offence was found to have been committed by them, while at the same time also holding that the Applications of Petitioners were premature. A bare perusal of the said order would reveal that the same was cryptic and did not contain any reasoning whatsoever which could justify the ten days notice period. 

While setting aside this order, the Supreme Court examined whether such an order could have been passed by the High Court in the first place. The facts of the said case reveal that notices under S. 108 of the Customs Act were issued to the Petitioners and due to their brazen non-cooperation, complaints were filed under Ss. 174 and 175 of the Indian Penal Code. Thereafter, the Application for Anticipatory Bail was disposed by the High Court with the direction as aforesaid.

In Padam Narain Aggarwal, the Supreme Court discussed the development of the law relating to grant of Anticipatory Bail in some detail and extensively relied on Sibbia's case. Paragraph 40 of the judgement in Sibbia, which is reproduced in Padam Narain Aggarwal, reads as follows;

“40... We agree that a 'blanket order' of anticipatory bail should not generally be passed. This flows from the very language of the section which, as discussed above, requires the applicant to show that he has "reason to believe" that he may be arrested. A belief can be said to be founded on reasonable grounds only if there is something tangible to go by on the basis of which it can be said that the applicant's apprehension that he may be arrested is genuine. That is why, normally, a direction should not issue under Section 438(1) to the effect that the applicant shall be released on bail "whenever arrested for whichever offence whatsoever." That is what is meant by a 'blanket order' of anticipatory bail, an order which serves as a blanket to cover or protect any and every kind of allegedly unlawful activity, in fact any eventuality, likely or unlikely regarding which, no concrete information can possibly be had. The rationale of a direction under Section 438(1) is the belief of the applicant founded on reasonable grounds that he may be arrested for a non-bailable offence….”

The Court in Sibbia ultimately came to the conclusion that no blanket order of bail can be passed while exercising power under S. 438 of the Code. Similarly, the Court in Padam Narain Aggarwal relied upon Adri Dharan Das, where the Supreme Court had held that normally direction should not be issued that the Applicant should be released on bail whenever arrested for whichever offence. 

In Das, it was also held that an interim order restraining arrest passed in an Application under S. 438 will amount to interference in investigation and will not be permissible under S. 438. After a conspectus of the case law in this regard, the Court in Padam Narain Aggarwal arrived at its final conclusions which are stated in paragraph 45:

“45. In our judgment, on the facts and in the circumstances of the present case, neither of the above directions can be said to be legal, valid or in consonance with law. Firstly, the order passed by the High Court is a blanket one as held by the Constitution Bench of this Court in Gurbaksh Singh and seeks to grant protection to respondents in respect of any non-bailable offence. Secondly, it illegally obstructs, interferes and curtails the authority of Custom Officers from exercising statutory power of arrest a person said to have committed a non-bailable offence by imposing a condition of giving ten days prior notice, a condition not warranted by law...”

The Court clearly arrived at a conclusion that the order directing grant of 10 days’ notice is a blanket one as held in Sibbia. It is submitted that the view taken by the Court is erroneous. 

Firstly, the meaning of the term blanket order has been explained by the Constitution bench in Sibbia and the same was reproduced by the Court in Padam Narain Aggarwal. The Sibbia dictum is that a direction under S. 438(1) that the accused should be released on bail whenever arrested in whichever offence would amount to a blanket order (see para 40 above). It court did not have occasion to examine whether reasonable notice prior to arrest is permissible under S. 438. The order of the High Court which was impugned in Padam Narain Aggarwal did not direct that the accused be released on bail whenever arrested for whichever offence. It merely directed that notice be given prior to arrest. Whether such a course of action would be permissible under S. 438 was the question which the Court sought to examine. 

The Court’s conclusion that such a course of action is not permissible since the same would be a blanket order granting protection in non bailable offence is clearly erroneous, as the direction impugned was not one which directed release on bail “whenever arrested for whatever offence.” What the Court in the impugned order had directed was that the authorities must give ten days’ notice prior to arrest. Such a direction, according to the Court, was necessitated since at the stage when the Application was filed, apprehension of arrest was not present. 

In Padam Narain Aggarwal, the Court failed to draw a distinction between a blanket order which directed release on bail for whatever offence and one which merely directed the prosecution machinery to grant reasonable notice prior to arrest. The settled position of law, insofar as Applications under S. 438 are concerned is that the apprehension of arrest needs to be established. 

When the Court is dealing with Applications where the apprehension is based on notices issued by authorities under special legislations such as the Customs Act or the Prevention of Money Laundering Act, it is not always possible for the accused to establish whether the agency concerned is even treating him as an accused. Thus, such Applications may be easily disposed off on the basis that the same are premature. However, if such is the position, then the provisions of S. 438 would be rendered nugatory. 

To prevent such a predicament, Courts some times direct a reasonable notice prior to arrest, which very often is for a period of 72 hours. Once such a notice is issued, the accused would be at liberty to approach the Court with an Application under S. 438 which can be decided on its own merits. Such orders are by no means anticipatory bail orders or blanket orders as contemplated in Sibbia for the reason that there is no direction contained therein to release the accused on bail.

The Court in Padam Narain Aggarwal could yet have set aside the impugned order directing 10 days notice prior to arrest since on facts, it is evident that the order did not contain sufficient reasons guiding the discretion of the Court. Moreover, the notice period of 10 days was perhaps otherwise excessive. However, the reliance on Sibbia was misplaced. Recently, a Constitution bench in Sushila Aggarwal has yet again relied upon the very same paragraphs of Sibbia and while summarising the principles, observed as follows;

“52.14. A blanket order under Section 438, directing the police to not arrest the Applicant, “wherever arrested and for whatever offense” should not be issued. An order based on reasonable apprehension relating to specific facts (though not spelt out with exactness) can be made. A blanket order would seriously interfere with the duties of the police to enforce the law and prevent commission of offenses in the future. (Paras 40-41, Sibbia)”

The Court has repeated the error of not appreciating the distinction between an order preventing arrest for a limited period and a blanket order to release on bail. The reliance on paragraphs 40 and 41 in Sibbia is clearly erroneous. A perusal of the above paragraph from Sushila Aggarwal reveals an inherent absurdity. The Court has stated that blanket orders to not arrest 'wherever arrested' cannot be passed. The absurdity lies in the fact that no direction to not arrest can be passed in cases where the accused is already arrested. It is only an order to release on bail that can be passed. It is thus clear that what is meant by the aforesaid direction is that the courts cannot grant orders directing blanket release wherever arrested. Furthermore, interestingly, the Court had observed, apparently in the context of blanket orders, that an order based on reasonable apprehension relating to specific facts though not spelt out with exactness can be passed.

Establishing the Case for Allowing Protective Orders under S. 438

The case of Dr. Sameer Narayanrao Paltewar v. State of Maharashtra decided by a single judge of the Bombay High Court settled a unique question relating to the fate of the accused persons who were directed to remain present for final hearing of their Applications under S. 438 and the Applications were dismissed. Section 438 of the Code of Criminal Procedure had been amended in its application to the State of Maharashtra and sub-section (4) thereof stipulates that the Sessions Court can direct the presence of the Applicant at the time of final hearing of the Application for Anticipatory Bail. 

The Bombay High Court was called upon to issue appropriate directions in order to ensure that liberties of the Applicants are protected by courts when Applications were dismissed while the Applicants are present in Court for final hearing. The High Court noticed that if the Applications under S. 438 are dismissed, the accused runs the risk of immediate arrest, thus frustrating his right to approach the High Court under S. 438 seeking the same relief. 

In this backdrop, the Court directed that the Sessions Courts must extend interim protection for a period of three to four days if Applications are rejected while the Applicant is present in Court. The said direction was issued in order to ensure that the remedy to approach the High Court is not rendered nugatory. 

It is pertinent to note that though it would appear that the order in question was passed in an entirely different context and was not any blanket order like the one impugned in Padam Narain Aggarwal, the principle of law that is applied by the High Court in this case and the principle of law applied in the other cases where prior notice before arrest was directed is one and the same. 

The principle is that by ensuring prior notice or some other protection to the accused, the right under Section 438 is not allowed to get frustrated. It is submitted that on a literal reading of S. 438 of the Code, it is possible to argue that the ruling in Paltewar is erroneous since the provision does not contemplate such a course of action. 

However, such a direction is borne out of a pragmatic and purposive interpretation of S. 438 and is therefore not erroneous. So also, directions of notice of reasonable time before arrest passed under S. 438 of the Code cannot be faulted simply on the premise that the same is not permissible under S. 438 or on the premise that such orders run contrary to the statutory powers of arrest. Such orders are necessary since it is not possible on every occasion for the Applicant to meet the threshold requirement of apprehension of arrest, particularly in cases where the investigating agency is acting under a special legislation such as the Customs Act or PMLA.

Exploring Inherent Powers to Support Notice prior to Arrest Orders

A conspectus of case law on the subject reveal that the discussion regarding orders directing notice prior to arrest has largely remained confined to the exercise of jurisdiction under S. 438 of the Code. However, whether such orders can be passed by the High Court in exercise of its inherent powers under S. 482 of the Code is a question that remains to be considered. 

Padam Narain Aggarwal reveals that the Court restricted itself to the interpretation of S. 438 of the Code and does not extend to exercise of inherent powers by the High Court under S. 482 of the Code. Further, the jurisprudence on the expanse of the powers of the High Court under S. 482 indicates that orders in the nature of notice prior to arrest could be passed in exercise of powers under S. 482. In the context of S. 438, it was possible to argue that the language of the provision does not permit any restriction on the power of arrest but only permits grant of bail in the event of arrest and that fetters cannot be imposed on statutory powers. However, when the High Court exercises its inherent powers, it is not possible to curtail its jurisdiction. The language of S. 482 is also indicative of this fact.

In Petitions for quashing of FIRs, the High Courts pass interim orders in the nature of not to take coercive steps or stay on investigation. Given the expanse of the inherent power of the High Court, an argument that curtailing statutory powers of the investigating agencies by way of interim relief is not permissible, has not been made. However, whether such relief can be granted as final relief is a question which would require some consideration. S. 482 is most commonly invoked in order to quash FIRs, chargesheets or criminal proceedings. 

It is also invoked in situations where the Code does not otherwise provide any remedies to approach the High Court. It remains to be seen whether the High Court’s jurisdiction under S. 482 of the Code can be invoked seeking relief of notice prior to arrest. There has never been an occasion for any accused to apply to the High Court seeking such a relief since such Applications were usually filed under S. 438. However, if it is to be held that the language of S. 438 does not permit such a relief, in the future it is likely that the jurisdiction of the High Court under S. 482 may be invoked.

In Petitions filed for quashing of FIRs, High Courts in fit cases pass interim orders staying investigations or directing the investigating agencies to not take any coercive action. However, such orders are interim orders where the final relief claimed is that of quashing of the FIR or proceedings concerned. In the case of Neeharika Infrastructure Private Ltd. v. State of Maharashtra the Supreme Court held that such interim orders directing no coercive steps cannot be passed if the High Court is dismissing the Petition for quashing of the FIR. However, no fetters have been imposed on the High Court’s power to direct no coercive steps as an interim order, provided that the High Court gives adequate reasons for arriving at the conclusion that such an interim order needs to be passed. 

The judgement of the Calcutta High Court in Ram Chandra Panda and Anr. v. State of West Bengal decided in January, 2023, presents yet another peculiar circumstance where the police were issuing notices under S. 160 of the Code to persons who were not named as accused in the FIR and were thereafter arresting these persons. Thus, a Petition was filed challenging few notices issued under S. 160 of the Code. S. 160 grants to the police making an investigation, the power to require the attendance of any person who may be acquainted with the facts of a case. 

The peculiar facts of the case before the High Court prompted it to ask the question as to why the police were resorting to this indirect method of arrest when they had the power to arrest at any time. The High Court held that such conduct was only to ensure that the accused does not get the chance of approaching the Court by filing an Anticipatory Bail Application or seek other protective orders. It further held that if this be the intention of the police, the situation required the court to invoke its inherent jurisdiction. It held that if a criminal case is to be initiated against the Petitioner, he shall not be arrested for a period of ten days so as to enable him to avail his remedies. 

It is apparent that the said order too has not been passed in exercise of powers under S. 438 but in exercise of inherent powers under S. 482. However, it appears that insofar as grant of relief of pre-arrest bail is concerned, there appears to be a fundamental flaw in the reasoning of the Court where it holds that the police were trying to ensure that the accused does not get a chance to avail anticipatory bail. It is noteworthy that in order to seek pre-arrest bail, it is not material as to under which provision of the Code a notice has been served. What is material for the said purpose is as to whether there is any apprehension of arrest.

Conclusion

To conclude with, it is submitted that a purposive interpretation of S. 438 would entail that the no fetters can be imposed on the power of the High Court or Court of Sessions from directing notice prior to arrest. Such orders are issued in situations where it is not possible to highlight the apprehension of arrest. In cases involving legislations such as the PMLA, such orders would be necessary so as to not render futile the mandate of S. 438. As an illustration, such orders could be necessary in situations where multiple FIRs are being registered on the basis of the same set of allegations. 

Further, the term “blanket order” as interpreted in Gurbaksh Singh’s case refers to orders directing release on bail in whatever offence and does not refer to orders directing notice prior to arrest. Hence, the rulings in Padam Narain Aggarwal and Vijaykumar Ramchandani require reconsideration.

A Closer Look at the 1976 Deletion of Section 438 CrPC in Uttar Pradesh

(Special thanks to Ashna D and Aniket Singh for their help with research on some aspects.)

In 2019, the State of Uttar Pradesh amended the Criminal Procedure Code of 1973 insofar as it applied to the state, to re-insert Section 438 (its at the end of the linked file) which provides for anticipatory bail. This came forty-three years after the original amendment passed in 1976 by which Section 438 had been deleted, by way of Section 9 of the Code of Criminal Procedure (Uttar Pradesh Amendment) Act of 1976 ["1976 Act"]. 

This came almost nine years after a government-appointed commission had recommended the move. The Commission observed that the objects accompanying the 1976 Act had claimed deletion of anticipatory bail was necessary because it was creating "practical difficulties"; which, the Commission reasoned, was a thinly-veiled justification provided to reduce the scope of personal liberty during the state of Emergency prevailing at the time (1975-77). Since the Emergency was long gone, it no longer made sense to continue without anticipatory bail in the state. 

The deletion of anticipatory bail in Uttar Pradesh had always piqued my curiosity. The justification offered and accepted in 2009 — that it was a reaction to the declaration of Emergency — neither seemed complete nor accurate. That the Supreme Court did not fully explore the issue while upholding the constitutional validity of the amendment in Kartar Singh [yes, the TADA judgment] was also unhelpful — all that the judgment carried was the submission of the state counsel that the 1976 amendment was passed to meet the "deteriorating situation". 

This set me off on a course of digging to try and get better answers. This post is the result of that exercise. 

Setting the Scene

After two decades of various courts coming at mostly the same conclusion—that anticipatory bail could not be granted under the Criminal Procedure Code of 1898—it was mildly surprising for anticipatory bail to be codified as a possible remedy under Section 438 of the new Criminal Procedure Code in 1973. When this relief was debated in Parliament, one general critique was its potential to upend criminal investigations by denying police the chance of arrest and interrogation. On top of which was the more vocal criticism about this relief being designed to only favour the rich litigants who would get word of potential cases and then rush with their lawyers to sessions courts and high courts. 

A combination of the two contributes to the kind of scandal standard left-leaning politics of the time loved (and still does), and this led anticipatory bail to earn mentions in parliament during question hour even before its codification — usually, in relation with alleged smugglers (For instance, during Question Hour in the Lok Sabha on 01.09.1972). The discontent, perceived or otherwise, with anticipatory bail was such that, within a year of the new Code codifying the relief, a Committee had been setup in 1974 at the central level to look into the question of its "misuse". On the basis of parliamentary questions in 1977 (see, Lok Sabha Debates of 14.12.1977), it appears that this Committee looked at data for all of 1974 and concluded that there were many cases in which investigations were delayed because of anticipatory bail, and that it was being taken "advantage" of by persons "concerned in criminal activities".

Most of the other changes brought about by the 1973 Code did not affect the existing scheme of rights but affected administration of criminal law by separating the functions between executive and judiciary — a pre-independence demand, and later a constitutional promise enshrined in the Directive Principles of State Policy. The new Criminal Procedure Code separated the police from the prosecutorial wing, took away judicial functions from executive magistrates, and reduced the ability of governments to handpick courts for dealing with specific classes of cases. Many states had already taken steps towards realising the separation between executive and judiciary prior to 1973, but very few states had gone to the lengths that the new Code asked them to travel. 

Uttar Pradesh was no exception. It did not take active steps to rectify inconsistent practices with the new regime, and so it only a matter of time before the existing practices were challenged in court. By two separate judgments passed in 1975, the Allahabad High Court set aside government circulars (i) placing prosecutors under control of the police [August 1975], and (ii) placing all cases prosecuted by the CBI within the state of U.P. before one single court [November 1975]. The result of the former would be to disrupt how trials took place across the state. The latter would potentially result in a transfer of thousands of cases, as well as the establishment of several new courts across the state to then handle them.

Such unplanned financial outlay usually triggers quick government response. The same happened here, and the government decided to take immediate action by promulgating an ordinance on November 28, 1975. It was this Ordinance which then became the 1976 Act, in which anticipatory bail met its demise. 

Legislative Debates

The previous section demonstrated that, it wasn't the Emergency which led the U.P. government to delete the provision on anticipatory bail — there were rumblings at a central level prior to 1975. Further, the events leading up to the 1976 Act suggest that the deletion of anticipatory bail was one small part in what was a broader set of changes which the government pushed through to circumvent a need to change the existing administration of justice in the state. This helps understand the contents of the 1976 Act better as well. 

It would have been great to have the 1975 Ordinance and then compare it with the 1976 amendment to see if there were new insertions proposed in the Bill from what was earlier pushed through hurriedly, but I was not able to get my hands on it. What I could get access to was the legislative debates of 30.03.1976 leading up to the 1976 Act replacing the ordinance, which are available on the government website (can also be seen here for convenience) in Hindi. The debates are fascinating and, at least going by the opposition, would have one think that the High Court's judgments of 1975 were nothing but a convenient excuse to usher in a set of regressive amendments to the criminal law. The real driving force was the Emergency. By way of these changes, the government wanted to halt the separation of the executive from the judiciary, and thus keep its vice-like grip on the administration of justice in the state. 

On the question of anticipatory bail, the opposition members asked what, if any, were the glaring instances of abuse which led the government to propose such a radical change? Anticipatory bail was new, yes, and perhaps viewed with suspicion by everyone — but that had not led any other state to delete it altogether. The reply brought up the classic, haunting image of the white-collar criminals — smugglers, hoarders and black-marketeers — all taking advantage of this new provision. It wasn't only such kinds of offenders, but all notorious criminals seemed to be taking advantage of the law as per the government. Ultimately, as the law minister Prabhu Narayan put it, as everyone agreed that the police had a right to investigate fully, and since it was evident that the provision for anticipatory bail was used to stymie investigations especially by white-collar criminals and notorious persons, deleting the provision was the only possible choice for the government.    

Parallel Developments at the Centre

The 1976 Act was published in the gazette on 01.05.1976. On 26.08.1976, a Bill had been taken up for discussion in the Rajya Sabha proposing amendments to the Criminal Procedure Code, which included a suggested amendment deleting Section 438 altogether. The Deputy Minister for Home Affairs remarked that "Unfortunately, experience shows that this provision has been availed of mostly by rich people, white-collar criminals and others. It has, therefore, been proposed in Clause 31 that this provision should be deleted." 

The proposal did not meet with wholesale opposition; rather, as one member put it, it felt as if "that experiment had not succeeded". The Code of Criminal Procedure (Amendment) Bill, 1976 was passed by the Rajya Sabha and sent to the Lok Sabha. It was only because Parliament was dissolved that the debates were not carried forward on the same. However, when on 14.12.1977, questions were put up to the new Janata Government on amendment of anticipatory bail, and the Deputy Minister for Home Affairs replied that the question of amending some provisions including anticipatory bail was being examined. 

A year later, on 28.11.1978 the new government introduced a bill in the Lok Sabha to amend the 1973 Code. It was based on the same Bill that was passed in the Rajya Sabha back in 1976, but deleted clauses which the government now thought were "unnecessary and detrimental to the interests of the parties". This included the clause deleting anticipatory bail, finally setting the controversy to rest. Of course, these developments at the centre did not compel any change of heart in the State of Uttar Pradesh. Even as the state passed further amendments tweaking the Code in 1977, anticipatory bail remained ousted.    

Connecting the Dots

The brief glimpse into the turbulence that accompanied the codification of anticipatory bail in the Code of Criminal Procedure 1973 gives us a window into the circumstances which led to the forty-three year long absence of Section 438 from the State of Uttar Pradesh (though not an absence of anticipatory bail, which remained alive as an extraordinary relief given by the High Court). Ascribing that move to the Emergency and the discomfort it generated towards personal liberty is a convenient, though ultimately inaccurate, tale. Those circumstances certainly would have played a role behind the 1976 Act, but history suggests that the discontent with Section 438 existed from before, and if anything, that sense of uneasiness has never gone away. It resurfaces each time anticipatory bail is granted in a media-friendly case, at which point talking-heads cry foul over criminals "taking advantage" of the law.   

This uneasiness with "criminals taking advantage" of what the law itself provides is an irrational, emotive outburst. But it has shaped the legal regime on bail and anticipatory bail over the years. Legislatures have excluded anticipatory bail outright in various contexts after Uttar Pradesh first did it, and have also placed much stricter regimes for granting bail for certain offences. These exceptional regimes have later been upheld by constitutional courts as well. All of which proceeded on the same fallacy — labelling persons accused of certain crimes as criminals, which justified lengthy pre-trial custody in spite of the prevailing presumption of innocence. The sentiment is evident in how Uttar Pradesh has reintroduced the provision. It did not simply adopt what the central law states, but modified Section 438 to exclude its application in what might seen as "serious crimes", i.e. offences punishable with death and, of course, those punishable under the UAPA.

If we manage to look past the hullabaloo, we come across a classic struggle at play here. Underneath all that hasty labelling of persons, I would argue, is the longstanding executive mistrust of unhindered judicial independence. It is that same struggle for separating judicial and executive functions within the criminal law, which has been going on since before India's independence. 

Think about it: what the law confers in respect of non-bailable offences is merely a right to apply for bail / anticipatory bail — whether or not you get relief depends on the court. Trusting judicial discretion on matters of bail and anticipatory bail is the defining, albeit problematic, feature of the legal regime in India. In a setting where arrests and custodial interrogation is still viewed as critical for criminal investigations, courts are trusted to make the right decision on matters of bail by ensuring these interests are considered without disregarding personal liberty altogether. It is for this reason that bail is not simply about securing the personal appearance of an accused, but involves courts looking at the allegations as well. 

By taking away anticipatory bail outright or by shackling the right to bail, in cases of certain crimes, what the legislature really does is strike a blow to this belief by sending a clear message that judicial discretion cannot be trusted to make the right decision. The already wide berth given to police interests in matters of bail is expanded even further, reducing the role of courts from independent arbiters to requiring that they defer to the executive's interests and consign a person to jail on the strength of allegations, not proof. The result is a legal system in which, somehow, the judiciary is further weakened in discharging its basic role of securing personal liberty. 

Thus, even as the State of Uttar Pradesh reintroduced Section 438, this is not a break with the past. Rather, it carries forward the legacy of the original deletion by the 1976 Act, a legacy which has been gradually, and almost unquestioningly, embraced across the country with the passage of time.

The Supreme Court and Anticipatory Bail — A Troubled Courtship

This post is not prompted by a recent case or statutory development, but instead the desire to take a step back and look at "bigger picture" concerns in an area of law, navigating which is akin to fortune-telling on the best of days and charting a hidden minefield on most others. This is Anticipatory Bail, that great and peculiar legal tool that lawyers in South Asia are familiar with [yes, it is not only India which has this remedy folks]. For the uninitiated, Anticipatory Bail is a legal remedy which enables a person to approach a court asking for bail "in the event of an arrest". In India, this is statutorily provided under Section 438 of the Criminal Procedure Code 1973 [Cr.P.C.], and requires that an accused show reasons to believe that she is likely to be arrested on accusation of having committed a non-bailable offence. If a court is satisfied in respect of the same, then it can pass an order directing the police to release the individual on bail in the event that it decides to arrest her, where this Anticipatory Bail will come with certain conditions such as ensuring cooperation with investigation and not coercing witnesses, etc. 

The Issue

The primary cause for confusion in respect of the legal position on Anticipatory Bail is, as one would expect, the manifold judicial decisions which have expounded on its scope and ambit. After all, merely reading the text of the provision might lead one to think that Anticipatory Bail is there for the asking if a person can show reasons to believe she will be arrested for accusations of committing a non-bailable offence. This is hardly so, as a court "may" grant the relief, and this question of determining the proper exercise of judicial discretion which has been the subject of controversy in all these decisions. The point being made here is not simply to complain about this confusion or berate courts for it. Instead, what I want to focus upon is a choice that the Supreme Court of India made in 1980 in respect of the judicial discretion afforded in matters of Anticipatory Bail in India, when a Constitution Bench of that Court considered the then-recently-added provision in a batch of petitions which was reported as Gurbaksh Singh Sibbia v. State of Punjab & Ors. [AIR 1980 SC 1632 ("Sibbia")]. 

This was a choice to remain faithful to the perceived framers' intent behind adding Anticipatory Bail, or to go beyond these confines by adopting a more purposive interpretive bent. The Constitution Bench opted for the latter, and the consequences have been bittersweet. On the one hand, Anticipatory Bail became an option to secure personal liberty more generally and it also added teeth to the fundamental right secured under Article 21 of the Constitution. On the other hand, this approach militated against what one might imagine as the DNA of the criminal process by denying arrest and thus allowing police investigations and / or the public interest to be "frustrated" as a result. Which is why, in spite of it being backed by a Constitution Bench decision of the Supreme Court, the liberal approach to Anticipatory Bail continues to raise many eyebrows and often meets with disapproval from that very Court till date.

The Interpretive Question in Anticipatory Bail

The framers approach, I argue, was to keep Anticipatory Bail as a remedy against baseless accusations — as a defence against being arrested in a malicious prosecution. This was the context in which such orders had been sought for before various High Courts in the two decades prior to the Cr.P.C. 1973, albeit with limited success. It was this context which led to the question being presented before the Commission. 

However, the 41st Law Commission Report (1969) sponsoring insertion of a provision for Anticipatory Bail did not stipulate this malicious prosecution context as the only purpose for which such a relief of an Anticipatory Bail order may be needed. The Commission noted in Para 39.9 that:   

 

The suggestion for directing the release of a person on bail prior to his arrest (commonly known as "anticipatory bail") was carefully considered by us. ... The necessity for granting anticipatory bail arises mainly because sometimes influential persons try to implicate their rivals in false cases for the purpose of disgracing them or for other purposes by getting them detained in jail for some days. In recent times, with the accentuation of political rivalry, this tendency is showing signs of steady increase. Apart from false cases, where there are reasonable grounds for holding that a person accused of an offence is not likely to abscond, or otherwise misuse his liberty while on bail, there seems no justification to require him first to submit to custody, remain in prison for some days and then apply for bail. [Emphasis mine]

Anticipatory Bail was therefore not only as a relief against malicious prosecution but also as a potential remedy against unjust incarceration caused because of the slow pace at which the legal system worked as people who otherwise be released on bail would have to "remain in prison for some days" and then apply for bail. It would be a stretch to consider this as a general expansion as "some days" come in almost every case. The only sensible way would be to treat this suggestion as offering a remedy in the exceptional case, and not more; doing so would render the distinctions between bail and Anticipatory Bail redundant which was certainly not what the Law Commission wanted to do. This view of treating Anticipatory Bail as the exceptional remedy is only buttressed if one considers the 48th Report which followed the introduction of the Criminal Procedure Code Bill of 1970 — the basis of the Cr.P.C. 1973 — wherein at Para 31 the Law Commission did retain its support for the measure but suggested it be an option only in "very exceptional cases".       

When, in 1976, the Supreme Court took up the limited issue of the exclusion of Anticipatory Bail in the statutory context of the the Defence of India Act 1971 in Balchand Jain [AIR 1977 SC 366], the separate opinions of Justices Bhagwati and Fazal Ali echoed this view of Anticipatory Bail being fit only for the exceptional case. The Punjab & Haryana High Court followed this view in its decision dismissing the petitions filed by Gurbaksh Singh Sibbia and others [1978 Cri LJ 20]. In a bid to settle the confusion that it had seen arise, in some measure due to the Law Commission's views and also due to the charged political context into which Anticipatory Bail was inserted, the High Court went ahead and offered strict guidelines regulating the scope of judicial discretion on the point of Anticipatory Bail. One of the guidelines was to permit Anticipatory Bail as an option only where an issue of mala fides was being alleged in respect of the accusations levelled against the individual seeking relief. The High Court considered limiting the scope of discretion as advisable also because it seriously bought into the argument that an expansive approach to Anticipatory Bail would stymie investigations. It agreed with the government counsel that speed was of the essence in investigations where the initial period of police custody was often critical to solving the case. Liberal pre-arrest bails would thwart investigations and hurt the public interest.    

The strict guidelines authored by the Punjab & Haryana High Court seemed to extinguish the remedy of Anticipatory Bail altogether and met with stern disapproval from the Supreme Court in Sibbia, observing that it did not see "why the provisions of Section 438 should be suspected as containing something volatile or incendiary, which needs to be handled with the greatest care imaginable." The worries about thwarting investigations by depriving police of custody during the initial period were played down by the Supreme Court which was of the view that all investigative needs could be met with placing conditions requiring that a person cooperate with police. There was little discussion about the value of custodial interrogation to an investigation. Rather than see value in police custody, it is fair to say that in this post-Emergency, post-ADM Jabalpur, Maneka Gandhi era of the Supreme Court what weighed more heavily with the Bench was the threat posed by arbitrary arrests to personal liberty. One could see the stark difference in approach when, rather than talk about the importance of custodial interrogation, the Supreme Court spoke about a need to respect the presumption of innocence of the person applying for Anticipatory Bail. As a result, the only acceptable approach was an expansive one and not one which insisted on mala fides to be established for granting Anticipatory Bail. The Court acknowledged that there could be problems caused to investigations by Anticipatory Bail, but it noted that "a wise exercise of judicial power invariably takes care of the evil consequences which are likely to flow out of its intemperate use."   

The Fallout: Between a Rock and a Hard Place

In 2020, Sibbia was endorsed as being the correct and authoritative position of law on Anticipatory Bail by another Constitution Bench of the Court in Sushila Agarwal [2020 SCC OnLine SC 98] where, if anything, the Supreme Court went even further than Sibbia in endorsing an expansive and liberal approach to Anticipatory Bail. However, the journey from 1980 to 2020 is miles away from being imagined as a linear progression in the legal position affirming greater protection for personal liberty. Beneath these lofty pronouncements of the Supreme Court about Anticipatory Bail lies a great degree of reservation around this remedy. Just as an example, how can one square the law as laid down in Sibbia with the judgment in P. Chidambaram v. Enforcement Directorate [2019 SCC OnLine SC 1143] rendered by the Supreme Court in September, 2019, denying Anticipatory Bail to the petitioner. 

Yes, arrests can be unnecessary and arbitrary, but at the same time they are critical to ensure the criminal process "works" in the eyes of stakeholders. Arrests and the police custody which may follow within the first fourteen days thereafter are still accepted as being the most important part of any investigation. This is a feature of the system which the Supreme Court itself recognises time and again [see, as an example, CBI v. Anil Sharma, (1997) 7 SCC 187], and in doing so goes against the logic of Sibbia that the interests of an investigation can be sufficiently secured even without police custody. In addition to this, the criminal process endorses a logic that pre-trial arrests and custody, for at least some duration, are justified in cases of serious crime. Bail has never been only about ensuring that a person faces trial, but its desirability has historically been linked to the gravity of accusations. This bent of mind has, naturally, accompanied the exercise of judicial discretion on the point of Anticipatory Bail. A good example of this is the 2019 judgment in P. Chidambaram which reflects the consistent hesitation of courts to allow Anticipatory Bail for economic fraud. If anything, the reservations in granting Anticipatory Bail in serious crime appear to be even more pronounced than those associated with the grant of regular bail in this regard, as is evident from the complete exclusion of Anticipatory Bail as a remedy in certain statutory contexts — for example, anti-terrorism laws.   

Conclusion

In expanding the idea of Anticipatory Bail back in 1980, the intention of the Supreme Court in Sibbia was to make sure that this new tool is not jettisoned due to perceived fears of certain sections. However, the manner in which the Court sought to realise this was to leave everything to judicial discretion. The result is a branch of law that is pockmarked by inconsistency and uncertainty, which are perhaps the two worst things that an accused or potential accused person has to contend with. Anticipatory Bail became exactly like the concept of bail in respect of non-bailable offences — a concession, not a right.

It would be surprising to see Anticipatory Bail being scaled back to a pre-Sibbia position as a remedy only in cases of malicious prosecutions, even though this might be the most desirable way to make sense of the legal position. This perspective still continues to inform the law, in my opinion, and in practice it might still be the most common category of cases in which Anticipatory Bail is granted. It could well be argued that this perspective neatly explains why courts remain hesitant to accept Anticipatory Bail in contexts where the accusation comes not from private parties, but from government officials who are presumed to do their job correctly. However, the recognition of other residual categories of deserving cases for the grant of Anticipatory Bail is seen as too important to give up, and Sushila Aggarwal underscored that the Court still sees Anticipatory Bail as being a potential tool to help reduce arbitrary arrest and detention. Of course, just not a very good tool. As a result, it is fair to imagine Anticipatory Bail continuing to seem like going to play slots in a casino and hoping to get lucky till the system somehow moves past its preference for arrests and pre-trial custody of accused persons.       

The Constitution Bench Reference in Sushila Aggarwal v. State

A Constitution Bench of the Supreme Court decided a reference made in Sushila Aggarwal & Ors. v. State (NCT of Delhi) & Anr. [SLP (Crl.) 7281-82/2017, decided on 29.01.2020], where two questions had been posed for consideration:

(1) Whether the protection granted to a person under Section 438 Cr.P.C. should be limited to a fixed period so as to enable the person to surrender before the Trial Court and seek regular bail.

(2) Whether the life of an anticipatory bail should end at the time and stage when the accused is summoned by the court. [Para 1, Shah, J.]

Two separate opinions, and 133 pages later, we find that the five Justices unanimously answered the Reference (which is, frankly, just one question broken up into two) as follows: There is no statutory time limit on the protection of "Anticipatory Bail" given under Section 438 Cr.P.C., and nor does this protection automatically expire when an accused is summoned to Court. Such a limit can be imposed by the Court if it thinks fit, but should not be normally done. The Bench also went beyond the Reference, and "clarified" the law and practice on Section  438 Cr.P.C., by issuing 12 pointers for all courts to consider [See the section bearing signatures of the entire Bench].     

What led to the Reference?

The Reference made in Sushila Aggarwal seems to have been the result of two conflicting streams of opinion in past precedent. One line of cases, symbolised by the decision in Mhetre [(2011) 1 SCC 694] had held that there can be no limit as to the time duration of an order granting Anticipatory Bail. The second line of cases, symbolised by Salauddin [(1996) 1 SCC 667], had instead held that any order passed under Section 438 Cr.P.C. must be of a limited duration. And, at the heart of it all, was an old decision in Gurbaksh Singh Sibbia [(1980) 2 SCC 565], which had been interpreted and relied upon in both lines of cases [See Para 7.4 (Shah, J.)].

The Reference — Reaffirming Judicial Discretion in S. 438

It is a shame that the concise manner in which the issues were framed for the Reference (Yes, I am taking a dig here), did not contribute to a similarly concise set of opinions. Both opinions, of Shah, J. & Bhat, J., carry lengthy extracts from statutory provisions and prior precedent which unnecessarily burdens the reader. 

On cutting through the prolixity, one finds that both opinions are premised on the same logic — ensuring that judges have near-total room for discretion in considering any requests for Anticipatory Bail under Section 438 Cr.P.C [Para 7.5 (Shah, J.); Paras 68, 76 (Bhat, J.)]. While the opinion of Shah, J. answers both questions on this basis alone, Bhat, J. also offers independent bases to refute arguments to limit anticipatory bail simply because the investigation was over. As he observes, that an investigation stands complete is, by itself, a factor towards granting bail and not taking persons into custody [Paras 70, 76]. 

Both opinions view the conflicting decisions as having incorrectly interpreted Sibbia, insofar as they sought to create mandatory rules for courts in the realm of Anticipatory Bail. This discussion goes beyond rules imposing time limits, to also mention opinions suggesting that entire categories of offences should be outside the ambit of Section 438 [Paras 65, 68 (Bhat, J.)]. Through the opinions, we are repeatedly reminded that the sole basis for limiting judicial discretion in this realm of bail is an express provision in statute — anything less is, at best, a point of guidance for judges to help them exercise the weighty discretion that they wield [Para 56, 59, 62, 64, 68, 80 (Bhat, J.)]. 

Thus, at the end, it is confirmed that there is nothing warranting an outright refusal to consider an application under Section 438 Cr.P.C., except what may be specifically provided for in statute. Everything else, be it the nature of the offence, or the particular needs of the investigation, is part of the factors that courts must consider while deciding the fate of an application.

Appraisal — The Perennial Victory of Standards over Rules 

As discussed, at the heart of the Reference in Sushila Aggarwal is the Supreme Court's affirmation of standards over rules — enabling judicial discretion rather than curtailing it in any manner at all. If we look at the issue with this lens, then the choice appears almost like a foregone conclusion. Take up any branch of criminal procedure law in India, and it will show that similar conflicts between having strict rules versus malleable standards always end in victory for the standards. After all, since having standards implies more power for judges, who have more discretion, it is counterintuitive to expect a ruling that clamps this discretion in favour of bright line rules. 

But the problem is that in criminal procedure, sometimes clear rules are actually more useful than broad standards, and I fear that the Supreme Court in Sushila Aggarwal may just have compounded the pre-existing vagaries that cripple the practice of bail cases across most parts of India. Even though rules curtail wriggle-room for doing justice in exceptional situations, they provide a measure of certainty, consistency, and predictability that remains elusive when dealing with standards. And, at some level, this choice between rules and standards must also be cognizant of the system in which the choice is being made. Enhancing discretion implies a degree of trust in the ability and capability of the officers entrusted with that discretion, while rules ensure that the fate of an individual does not disproportionately hinge upon the peculiar attributes of any judge. 

It is unsurprising to read that the Supreme Court trusts judicial officers with discretion. But, can it be said that the trial courts across India, criminally overburdened as they are, have the capability to sit and consider the umpteen number of factors that the Court in Sushila Aggarwal suggests should be looked at while considering applications under Section 438 Cr.P.C.? I am not so sure.

Anticipatory Bail and Jurisdiction

Suppose you are ordinarily a resident of Delhi, but come to know that a case has been filed (or is about to be filed) against you in Mumbai, alleging you committed non-bailable and cognizable offences such as "Cheating", punishable under Section 420 of the Indian Penal Code 1860 [IPC]. Your law-savvy friends tell you that it is wise to urgently move an Application under Section 438 of the Criminal Procedure Code 1973 [Cr.P.C.] to get "Anticipatory Bail" and diffuse chances of an arrest. 

For those who are unaware of the term, "Anticipatory Bail" refers to a legal remedy where a person can go to court to seek bail in anticipation of her arrest on allegations of having committed a non-bailable offence, which may or may not have yet been registered as a case. If the court finds the apprehensions are genuine, and that there are reasons to protect against arrest, then it can direct the police to release the Applicant on bail in the event of her arrest.  

This, obviously, makes sense in our example, and so you get the papers ready. But here's the problem: where do you move the Application? In Delhi, where you reside, or in Mumbai where the case is filed? This post discusses the issue of jurisdiction in context of seeking Anticipatory Bail. I explain why the issue persists, and then try and make sense of the different opinions prevailing across High Courts. I argue that the wisest course seems to be to chart the middle course, and allow persons to file the Application in both places.

Some Necessary Background 

Section 438 Cr.P.C. was not part of the old Criminal Procedure Codes passed in British India. It was a new feature in the 1973 Code, based on a trend emerging in some states that was endorsed for statutory recognition by the Law Commission of India in its 41st Report (see pages 320-321 of the document), and its 48th Report (see page 10 of the document). Why was this new remedy added? In the 41st Report, the Commission argued that it was needed "mainly because sometimes influential people try to implicate their rivals in false cases for the purposes of disgracing them or for other purposes by getting them detained in jail for some days." It recommended that the power to pass such orders not be given to Magistrates, and be reserved for the Sessions Court or High Court. What the Commission recommended as Section 497-A for the old 1898 Code, has become Section 438 in the current Cr.P.C. 1973. 

As mentioned above, the idea of anticipatory bail was to ensure that persons be released on bail after arrest. Which is why many judicial decisions on the subject state that it is not a protection from arrest per se. However, in practice, anticipatory bail ends up working as a protection from arrest, because rather than arrest a person and release her on bail the police choose to not conduct any arrest at all. As a result, a person ends up requiring to file for bail once again after the police investigation is over. In a later section, I will come back to how this practical aspect of anticipatory bail also contributes to the issues of jurisdiction.       

The Jurisdiction Problem in Anticipatory Bail

Take a look at Chapter XIII of the Cr.P.C., which explains how to determine the jurisdiction of criminal courts. It is striking how this set of provisions regularly confers jurisdiction upon more than one place. This is not accidental, as it is guided by the idea that technicalities should not override the criminal process. So, if the Cr.P.C. itself is happy with multiple jurisdictions for the inquiries and trials, why am I suggesting there is a jurisdiction problem? Why can't both places, i.e. the site of accusations and the site of my fear of arrest, have jurisdiction in cases of anticipatory bail? 

This is because throughout Chapter XIII, the jurisdiction of criminal courts is based on places having connections to the alleged offence, not where the accused lives. This leads some people to argue that jurisdiction, as a concept in Indian criminal procedure, is offence-based.

This becomes important when we compare anticipatory bail with regular bail. In the latter, a person is taken into custody, and then moves for being released. In those situations, there are no qualms about where to file for release: you either go to the relevant police officer, or the court which extends your custody beyond 24 hours. And if you want to challenge your remand to custody, you go before the appellate courts within that area. Thus, custody is the key factor for deciding jurisdiction in regular bail cases, and thus ties in nicely with the Chapter XIII concepts. 

Anticipatory bail, on the other hand, precedes custody. What triggers this process is the apprehension of being taken into custody for committing a non-bailable offence. And since this apprehension is person-specific, it is quite possible for the place of potential custody to be divorced from where I reside and harbour my apprehensions of arrest. So, it appears that there are two key factors to decide jurisdiction: accusations of committing an offence, and apprehension of arrest; and it is not necessary for them to always overlap. And the place of residence for a defendant, as we saw, is not a principle for determining jurisdiction under Chapter XIII.

Besides relying on principles, the argument of particularity in deciding anticipatory bail jurisdiction also gets support from the text of Section 438(1) Cr.P.C. itself:

When any person has reason to believe that he may be arrested on an accusation of having committed a non-bailable offence, he may apply to the High Court or the Court of Session for a direction under this Section; and the Court may, if it thinks fit, direct that in the event of such arrest, he shall be released on bail. [Emphasis mine]

The words in Section 438(1) are clear: you can apply to the High Court or the Sessions Court, not a / any High Court or Sessions Court. This "the" cannot be accidental, and therefore, must be seen as specifying one court as the court of competent jurisdiction. And since one provision of the Cr.P.C. cannot be contradicting the overarching principle on jurisdiction identified above, this particular court must be the place where the alleged offence occurred / case was registered. 

Diverging Judicial Opinion

In State of Assam v. Brojen Gogol / R.K. Krishna Kumar & Ors. [(1998) 1 SCC 397], the Supreme Court had a chance to address the issue. The case involved a batch of petitions where the Bombay High Court had granted anticipatory bail to accused persons in a case registered in Guwahati. When the State of Assam moved the Supreme Court, it transferred the cases to Assam, but while doing so expressly chose to leave the jurisdiction question undecided. 

As a result, today there is still a spectrum of divergent judicial opinion on the issue.* On one end are courts which subscribe to an expansive view on jurisdiction and allow applications to be filed in the place where the accused resides. This set includes some orders of the Delhi High Court. These courts do not bunk offence-based jurisdiction, but consider the expansive view necessary because of another, equally fundamental, principle of criminal procedure: ensuring the widest possible protection for personal liberty. 

At the other end of the spectrum are courts which subscribe to the restrictive view sketched out in the previous section. Further textual support for their view is derived from Section 81 Cr.P.C., which expressly confers jurisdiction in bail cases on a court other than the court within whose territorial jurisdiction the case is filed. Thus, the Patna High Court, Calcutta High Court, and Punjab & Haryana High Court all hold the view that only courts having jurisdiction over the place of alleged offences can entertain anticipatory bail applications.** 

Somewhere in the middle of this spectrum fall the Kerala High Court, Karnataka High Court, Bombay High Court and Gujarat High Court. The Kerala High Court held that an application can be filed where the accused resides. But, the scope of protection is limited to any arrests carried out within the territorial jurisdiction of the state. The other three High Courts don't impose territorial limits on the scope of protection, and instead place a time limit on the duration of protection from arrest, so as to enable to person to subsequently move an application before the other court. Both these views highlight different ways of expressing a judicial compromise between the rule of offence-based jurisdiction, with the other rule of protecting personal liberty from unnecessary restraint.   

Solving the Muddle 

Let me state at the outset that, in my view, the restrictive approach to jurisdiction is incorrect. 

This is because courts have wrongly derived from Chapter XIII that offence-based jurisdiction is an immutable principle in Indian criminal procedure. Chapter XIII applies to "inquiries and trials" in respect of cases. A hearing to decide whether anticipatory bail should be granted is certainly not a trial, but is it an inquiry? A Constitution Bench of the Supreme Court in Hardeep Singh [(2014) 3 SCC 92] said that the "inquiry" stage commences after filing a Charge-sheet. While this decision was in context of explaining the scope of Section 319 of the Code, the views on what is an "inquiry" are not limited to that context, are in fact supported by earlier cases as well.

Anticipatory bail hearings are at the start of the investigation process, very distant from the filing of a charge-sheet. It can be argued, therefore, that the principles of offence-based jurisdiction located in Chapter XIII are thus not applicable to the anticipatory bail context at all. Thus, while the principled opposition to residence-based jurisdiction falls, the support for such jurisdiction definitely remains, as there is nothing contrary to the principle that protecting personal liberty is a paramount consideration in the criminal process. If anything, the legislative history behind Section 438 Cr.P.C. confirms that this protection of personal liberty was the main reason why anticipatory bail was inserted in the new Cr.P.C. of 1973. To adopt a reading of the provision contrary to this express legislative intent is surely to be avoided.  

Therefore, the restrictive approach end of the spectrum is clearly undesirable. For different reasons, the views at the opposite end of the spectrum are equally undesirable — and, frankly, even the Delhi High Court has moved away from this position slowly — and make the middle ground arguably the most viable approach to follow. Even though it involves creating technically incorrect jargon — such as "transit" anticipatory bail, or "interim" anticipatory bail.

These reasons are two sets of practical considerations. The first, is how anticipatory bail orders are treated in practice (discussed a few sections above). Recall that technically anticipatory bail is not a protection from arrest. Thus, technically, the residence-based court would cease to matter after the urgent issue of anticipatory bail is decided. Because if a favourable order is passed, then the police would arrest the person and execute a bail bond, which would cement the offence-based jurisdiction henceforth. Any issues about bail etc. will have to be litigated in the local court now, because in law, that is the forum supervising the liberty of the accused person who is out on bail. 

However, we know that in practice anticipatory bail doesn't work this way. Rather, it becomes a protection from arrest because police refrain from carrying out the formal exercise of arresting the person and consequently releasing her on bail. To allow for an accused to be at liberty on the basis of orders passed by courts in a different state altogether is a problem. For starters, since there is usually no actual arrest, it means that there are, technically, no conditions on the person's liberty yet, which increases the chances of absconding; something already high if the accused lives in another state. Beyond this, I would argue that once we have moved away from the state of urgency that a scenario of anticipatory bail can entail and the principle of personal liberty is duly considered, it makes lesser sense to derogate from the logic of offence-based jurisdiction that the Cr.P.C. does subscribe to.

This is why the middle path makes most sense. It appreciates how anticipatory bail works in practice, and in this practical context gives due consideration to both principles, of protecting personal liberty and also securing local jurisdiction for prosecuting crimes. This view knows that ultimately, an order granting anticipatory bail proper is going to apply for the remainder of the investigation. Which is why it allows residence-based courts to protect against arrest for a sufficiently long period within which a person can approach the offence-based courts for anticipatory bail. One might argue that the view of the Kerala High Court, that protection from arrest cannot go beyond state-boundaries, is one that respects the federal system. But since we don't follow hard-federalism at all, and High Courts are constitutional creatures, one can also argue that the effect of a High Court's orders should certainly be allowed to travel beyond state-boundaries.

Conclusion
The issues regarding anticipatory bail and jurisdiction discussed in this post should not be issues in 2019, when we have had the provision on the statute books for almost half a century. Parliament could solve this muddle with a simple pen-stroke — delete "the" from Section 438(1) Cr.P.C. and it would seriously dilute the force in the restrictive view. But hoping for Parliament to amend criminal procedure is akin to Waiting for  Godot, and so we must wait, for the day when the Supreme Court might change its mind to settle the conflicts.

*Note, that none of these courts clearly tell us whether or not applications can also be filed before a Sessions Court.

**Note, that while the judgments don't say it, this view naturally means that if a case has not yet been filed in a particular place, then a plurality of courts can exist based on the jurisdiction rules of Chapter XIII.

[Post amended on 30.07.2019]