On never-ending criminal investigations, the Supreme Court is its own worst enemy

Since March 2020, the Delhi Police have been investigating the wanton violence and riots which engulfed parts of North-East Delhi. But on 29 February 2024, when, during bail hearings for an accused before the High Court, the bench asked if the investigation was complete, prosecutors remained tight-lipped. 

Central agencies have been investigating alleged illegalities in the formulation and rollout of the excise policy for Delhi since 2022. Yet, six reports and the Delhi Chief Minister’s arrest later, the Enforcement Directorate stated that the investigation was still continuing. In both sets of cases, several accused are in custody and trial is yet to commence. 

These are only two instances where agencies continue with their investigations for years, filing what is termed a ‘supplementary chargesheet’ or ‘supplementary complaint’ as and when the need arises. Why does this practice warrant a conversation? In part, because a bench of the Supreme Court observed in a recent bail hearing that this practice was “bothering” them and suggested that it may take up the issue in some future case.

These remarks require some unpacking, which is what I hope to offer here. I explain the tussles within the procedural law as well as the Supreme Court’s interpretation, which has made these problems intractable. To put it bluntly, the Court is its own worst enemy when it comes to untangling the problem. 

Filing incomplete reports to stop the clock

Traditionally, the criminal process is divided between investigation and trial—the latter begins when the former ends. It would be foolish to conceive of these as rock-solid divisions. Such a view would prevent the courts from looking at material which investigators had no idea about or access to at the time they concluded their probe.  

Ideally then, one imagines the police concluding their investigation and submitting a case for trial, with a standard-issue rider that it may come back if it finds something new. Indeed, this rider is present throughout the trial, for a court can allow entry of material that was unavailable before. Even verdicts can be overturned on this basis of fresh material being brought to light. 

In this ideal sense, some degree of investigation is always pending during trial. Why has this not upended the process in all these years? Probably because in standard cases, the police and prosecutors reasonably believe that they have concluded the probe. 

Our problem arises when police submit a report but in the same breath inform the court that they will most certainly return with more material which they know most certainly exists. This is especially relevant in cases like the Delhi Riots probe and the Liquor Scam, where a sweeping conspiracy is alleged to be at play. 

The police are seemingly constrained to break up the probe into pieces because there is a time-limit to complete investigations under Indian law while an accused is in custody—60 days for most offences and 90 days for more serious ones. If the investigation is not completed within the stipulated time, the accused is entitled to be released on bail. 

When such reports are filed with a view to stop the clock and prevent bail, courts find it difficult to frame charges and start trial. That’s because the case against the accused and the material being relied on is still fluid. In some cases, the prosecution itself may ask for deferral as it is hamstrung from presenting its case on the larger conspiracy when investigation on those aspects is yet to be concluded. In other cases, the accused is likely to request deferral, fearing that commencing arguments will allow the investigation to cover up issues that may be pointed out.  

Confronting the problem

Since 2019, following the decision of the Court in Vinubhai Malaviya, there has been a measure of clarity on how to deal with the problem of potential unfairness in starting trial while the investigation is ongoing. Police should and do have unfettered powers to continue investigating a case even after submitting a report, but this exercise must ordinarily cease when a court frames charges. This position helps strike a balance between ensuring the fullest opportunity for police to ferret out the facts, while assuring that the goalposts are fixed before proceeding to trial.

Presumably because courts usually refrain from commenting upon the investigative process, treating it as the domain of the police, there has not been much judicial discussion on what causes an investigation to linger. But the phenomenon of incomplete reports has become more frequent, especially in ‘sensitive’ cases, and that’s probably why the Supreme Court feels the need to confront the issue.

How might it do so? Well, one way could be to turn the law on its head and conclude that trial must start even if an investigation is pending. This would seemingly remove the roadblock to starting trials, right? I beg to differ, for as mentioned above, the prosecution and defence are both unlikely to want to proceed with investigations pending.

How about time limits, then? Section 187 of the Bharatiya Nagarik Suraksha Sanhita, 2023 will introduce a measure on these lines, placing a 90-day time limit on finishing further investigations. At the same time, it allows for extensions where a court deems fit. Considering that most further investigations already result in a report within a time frame because of the bail provisions, I’m unsure whether an additional time frame will make a real difference.

Here’s where I hazard to make a bold claim: the only meaningful way to address the issue would be to engage with the police process in a substantive fashion to discern the completeness of investigations. This is exactly what courts refrain from doing ordinarily because it would necessarily need them to confront the tricky question of bail. The court often stops at examining whether the report makes out the commission of an offence. If the straightforward answer is yes (as it often is), then there’s no question of default bail. 

Mixed signals from the Court

A good example of this approach is how the Supreme Court, in January 2024, decided an appeal filed by the Central Bureau of Investigation against the bail granted to Kapil Wadhawan by the Delhi High Court on grounds that the chargesheet filed by the agency was an incomplete one. The Supreme Court allowed the appeal and set aside the bail order, disavowing an approach which looked to the contents of the report to decide the issue. Instead, the Court held that it was enough for the report to meet all the formal requirements spelt out in the law (see here for a longer discussion on the judgement). 

The Kapil Wadhawan judgement is a good example of the peculiarity that a hyper-formalistic approach begets: an investigation is ‘complete’ upon filing a report, and this report is ‘complete’ upon merely complying with the technical requirements of the chargesheet. 

A less extreme example of this approach is courts deciding whether or not the chargesheet discloses enough material for a court to take cognizance of the offence. ‘Cognizance’ is a light inquiry, where the court must scrutinise the material only to discern whether the ingredients of an offence are made out. It is not an inquiry which considers how fully the case against a specific accused is made out. Cognizance, as the Supreme Court itself reminds us, is taken only once, and of the ‘offence’ not the ‘offender’. 

There are exceptions. A recent and infamous one being the orders passed in Ritu Chhabria (2023). There, a two-judge bench admitted applications for release on bail. It adopted an approach which substantively engaged with the ‘incomplete’ material to decide if the chargesheet was an exercise designed purely to frustrate bail, and found that it was. 

The judgement carried strong language condemning the practice of agencies to file reports in a piecemeal fashion. In doing so, the Court noted that agencies had defeated the very purpose behind the statutory provisions inserted in the Code of Criminal Procedure, 1973, i.e. to deter the filing of incomplete reports to continue pre-trial custody.

Such an approach itself was not unprecedented. Yet, what followed Ritu Chhabaria has cast a cloud. Within a few weeks, an application had been filed seeking a recall of the judgement. Even before the application made its way in the regular course, the Chief Justice’s bench found the issue critical enough to merit a hearing and pass directions that the judgement was not to be relied upon until further orders. The CJI’s bench did this without explaining what prompted such an extraordinary order in the first place. This continues to remain the status quo. 

A minimum level of certainty

A prolonged and protracted investigation has become alarmingly common. When the Supreme Court turns to the legal regime, it will find that the barren text has been laden with meaning by its own past decisions. The Court’s judgements have crafted rules which separate the stage at which the police file chargesheets and the stage at which a court is encouraged to qualitatively examine whether the investigation is complete. 

To repair the investigative and trial process in these contexts, the Court must face its worst enemy: itself. One approach could be to encourage a stricter assessment of the case being made upon the filing of a chargesheet. Move away from only looking at whether the ingredients of an offence are made out, but also look at the case being made against the specific accused; especially where that accused is in custody. 

Of course, such an approach would still need to contend with the question of what level of fluidity is permissible to proceed with trial. Some fluidity is impossible to eradicate. At the same time, it is not difficult for a court to satisfy itself about the minimum level of certainty needed to proceed to framing charges (or discharging an accused) and kicking off the trial. 

If the Court is truly ‘bothered’ and is looking for an opportunity to be definitive on this issue, it need not look any further than the batch of applications keeping Ritu Chhabaria in suspended animation.

Abhinav Sekhri is a Delhi-based lawyer. His main areas of interest are criminal law and history. He runs the Proof of Guilt blog discussing criminal law and procedure.