Compelling an Accused to Unlock their Mobile Phones: A Critique of the Kerala and Karnataka High Court Judgments

On 29th January 2022, the High Court of Kerala handed down a brief judgment, holding that the Prosecution, in criminal cases, is entitled to access the data on an accused person’s mobile phone, and that this does not violate the constitutional guarantee against self-incrimination (P. Gopalakrishnan alias Dileep v State of Kerala). In doing so, the High Court purported to rely upon a recent judgment of the Karnataka High Court in Virendra Khanna v State of Karnataka. Both Dileep and Virendra Khanna, in turn, relied upon the eleven-judge-bench judgment of the Supreme Court in State of Bombay v Kathi Kalu Oghad, and considered themselves to be bound by it.

In my respectful submission, both the Karnataka and the Kerala High Courts misread Kathi Kalu Oghad, and arrived at an incorrect conclusion (see also this detailed critique of the Karnataka High Court judgment on The Proof of Guilt Blog, which arrives at the same conclusion). In paragraph 5 of its judgment, the Kerala High Court reproduced the findings in Oghad, but did not explain how that case – which concerned the compelled taking of fingerprints and handwriting samples – was applicable to the data on an individual’s mobile phone. In a somewhat longer analysis, the Karnataka High Court noted that:

The XI Judge Bench of the Apex Court in Kathi Kalu Oghad’s case has categorically held that providing of a thumb impression or impression of the palm or foot or fingers or specimen in writing or exposing a part of the body of an accused person for the purpose of identification would not amount to testimonial compulsion. Mere providing of an access of to smartphone or e-mail account would not amount to being a witness, the information that is accessed by the Investigating officer on the smartphone and or the e-mail account being only access to the data and/or documents, it is for the Investigating officer to prove and establish the same in a Court of law by following the applicable Rules of evidence.

It is submitted that this is evidently incorrect. The fact that a particular piece of evidence has to be proved in accordance with the procedure set out in the Evidence Act is not relevant to the question of self-incrimination; if that was the case, then even direct oral self-incrimination would, paradoxically, not fall within the prohibition of Article 20(3), since even that would still have to be “proved and established” in a Court of law in accordance with the Evidence Act.

A little later in its judgment, the Karnataka High Court then went on to agree with a submission that compelling production of a password to a phone was “akin” to fingerprinting, which had been upheld in Kathi Kalu Oghad. The analogy, however, proceeds upon a fundamental misreading of Kathi Kalu Oghad. In Oghad, the reason why fingerprints and handwriting samples were held to be outside the scope of testimonial compulsion was that, in themselves, they were neutral. A fingerprint or a handwriting sample did not by itself incriminate an accused, but only when it was compared with, say, a fingerprint from the scene of the crime, or the handwriting on a forged letter. This is obviously not the case when it comes to accessing data on the phone, because it is the data itself that has the potential to incriminate the accused. Indeed, both Dileep and Virendra Khanna appear to ignore the part of Kathi Kalu Oghad where – endorsing the judgment in M.P. Sharma – it was explicitly held that documentary evidence falls within the ambit of Article 20(3). The moment we accept that documentary evidence falls within the ambit of Article 20(3), it is impossible to reconcile the compelled unlocking of mobile phones – in order to access the data therein – with the guarantee against self-incrimination. Once again, the Karnataka High Court’s argument that data on a mobile phone is akin to a murder weapon, in that by itself it does not incriminate an accused, but still have to be “proved” in a court of law would, if taken to its logical conclusion, effectively destroy Article 20(3) altogether, because every piece of evidence – including oral testimonial compulsion – needs to be “proved” in a court of law before the Prosecution can rely upon it.

The Karnataka High Court then went on to recite a parade of horribles that would follow if mobile phone data was protected: “no blood samples can be taken, no body sample for the purposes of DNA analysis could be taken, no search of a house or office could be undertaken … offences like cyber crime could not be investigated.” To start with, as I have already demonstrated above, comparing mobile phone data to blood samples is an analytical error. But perhaps what is more disquieting about this particular form of consequential analysis is how it subordinates fundamental constitutional rights to the convenience of the investigative officer. The Karnataka High Court never stopped to ask why Article 20(3) exists in the Constitution, or to consider the possibility that the whole reason why we have a guarantee against self-incrimination is because there should be some limits upon what the State can do to an individual in the name of “efficient investigation.” In the Karnataka High Court’s analysis, it is simply assumed – instead of demonstrated – that something as invasive as DNA analysis of an individual should simply be a part of routine criminal investigation; and when regard for individual rights against the overbearing power of the State is so low, it is but the smallest of steps to go from forcible DNA extraction to extracting statements through torture, and justifying it in the name of investigatory efficiency.

It is, indeed, quite evident that the Karnataka High Court was aware of the shortcomings of its analysis, because it repeatedly went from the passcode or the password to a phone not itself being a “document”, to how the data on the phone would still have to be proven under The Evidence Act – while completely ignoring the fact that the data on the phone that could be accessed as a result of having the passcode would, under any definition, fall within the ambit of Article 20(3). It was only by artificially creating this gap in chronology that the Karnataka High Court – and the Kerala High Court – could justify compelling an accused to allow access to the contents of their mobile phone to investigative agencies.

The judgments of the Karnataka and Kerala High Courts are particularly concerning, because at a time when mobile phones are becoming more and more an extension of our interior lives rather than simple accessories, criminal procedure law should be moving towards greater protection of mobile phone data rather than a position where the State has free access to it. In Selvi v State – a judgment ignored by both Courts (even though it was specifically cited before the Karnataka High Court) – the Supreme Court grasped this fundamental point when it held that one of the goals of Article 20(3) – read with Article 20(1) – was to protect an accused’s mental privacy from invasion by the State.

Lastly, it may perhaps be pointed out that the accused before the Kerala High Court was a particularly unsympathetic and unsavoury character, and the case itself was one where the accused’s lawyers – in all likelihood – were raising the Article 20(3) as a delay tactic. That may or may not be true, but it is crucial to remember that the purpose of criminal law safeguards is not to protect the “good guys”, but to protect everyone. The inevitable consequence of that is that, on occasion, the “bad guys” will also be able to take advantage of these procedural safeguards; but that is not a reason to abolish or dilute the safegards themselves.

The question of law in these cases will, of course, undoubtedly reach the Supreme Court at some point (for a deeper overview, see this blog post by Abhinav Sekhri, and this paper). It will be interesting to see how the top Court – that, with the brief exception of Selvi – has historically been hostile to Article 20(3) – will deal with the issue.

Mobile Phones and Criminal Investigations: The Karnataka HC Judgment in Virendra Khanna [Guest Post]

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

(This post connects to a primer available on SSRN which deals with these issues, which is available here).

Earlier this month, a Single Judge Bench of the Karnataka High Court delivered the judgment in Virendra Khanna v. State of Karnataka and Anr. [W.P. No. 11759/2020 (Decided on 12.03.2021)]. The decision is likely to prove the first in a series of cases in the near future, in which courts grapple with issues posed to criminal investigations by mobile phones and similar digital devices. These issues require courts to not only consider the scope of constitutional protections but also interpret existing provisions of the Criminal Procedure Code 1973 [“Cr.P.C.”] and Information Technology Act 2000 [“IT Act”]. 

In this post, I argue that the conclusions arrived at in Virendra Khanna — in respect of the applicability of the fundamental right against self-incrimination, of the Cr.P.C. search and seizure provisions, and of the fundamental right to privacy, all in context of accessing digital devices — are incorrect. The interpretation that is endorsed by the High Court is also deeply problematic, as in the face of advancing technology, it seeks to restrict rather than enhance the contours of our constitutional rights to equip individuals with the means to protect themselves against unlawful incursions into enjoyment of one’s personal liberty by state agencies. The post does not engage with the guidelines provided by the Court, or its reiteration of the law laid down in Selvi [(2010) 7 SCC 263] that compelled administration of a polygraph test was is illegal.

The Facts and Issues Before the Court

The Petitioner was a person caught in the crosshairs of law enforcement agencies, and his mobile phone was allegedly important to advance the investigation into offences. In September 2020, the police went before the trial court asking for court orders to direct the Petitioner to unlock his mobile phone and grant access to email accounts, as the Petitioner had refused to cooperate. The court duly passed this direction and it appears the Petitioner complied. Then, the police moved another application before the trial court, this time asking for directions that the Petitioner be subjected to a polygraph test to confirm the mobile / email passwords, as it appeared that the Petitioner had been lying about the same during investigation. The court allowed this application as well and directed the polygraph tests be conducted — orders which, according to the Petitioner, were passed without ever giving him an opportunity to be heard and without considering if he had indeed consented to undergoing such tests. The Petitioner challenged this order and the consequent direction to undergo a polygraph test.

Following the judgment in Selvi, no court can direct any accused person to undergo polygraph tests unless such person consented to the same, and if seen from that perspective Virendra Khanna was an open-and-shut case requiring that the order be set aside. But the High Court was more indulgent with the legal issues placed before it and considered questions that lay beneath the surface as well. Out of these, I focus on the following three points taken up in the judgment:

  • What is the specific legal regime under which police can seek access to a digital device for pursuing its investigation?
  • What is the interplay between Article 20(3) and directions issued to an accused person for unlocking a digital device? 
  • What are the legal limits, if any, upon law enforcement agencies while they “explore” the contents of a digital device for investigation purposes?

Issue 1: The Relevant Legal Regime 

The High Court first considered whether there was any legal basis to root the actions of police officers in accessing a digital device for purposes of investigation. It observed that an officer could always ask an accused to open the device, but to direct compliance required some basis in law. This legal basis was found in the existing search and seizure regime of the Cr.P.C., concluding that the regime — which it admitted only applied to a “place” — was also applicable for accessing a digital device. Accordingly, the police would have to apply for search warrants to access a phone under Sections 93 / 94 of the Cr.P.C., and in emergent circumstances they could dispense with this requirement and act under Section 165 Cr.P.C. The obligations of the accused would be the same in both scenarios, i.e. assist in providing access to any locked space as provided by Section 100 Cr.P.C. 

Extending the existing search and seizure regimes from the realm of physical space to that of electronic / digital space is a path that many countries are taking. Erecting a need for judicial supervision by requiring search warrants to be sought before digital devices can be accessed helps redress the imbalance of power in such situations and also helps to keep law enforcement activity tailored to the needs of investigation and avoid roving inquiries into personal data. Viewed in the abstract, then, the choice of the Court does not seem problematic at all but a pragmatic solution. 

The problems only arise when we move beyond the realm of abstraction into practice. The Indian search and seizure regime does not mandatorily require search warrants; instead, police liberally use their powers under the “emergent circumstances” exception to conduct searches. The result is a situation where privacy is at the mercy of police. This is not to say that the search warrant regime itself, when invoked, supplies the necessary bulwark. The Cr.P.C. 1973 adopted wholesale the search regime that was present in the old British codes, the avowed purpose of which was to maximise scope of interference with personal liberty and not to safeguard it. Under this antiquated regime, general warrants allowing a roving search at a place are the rule, and a court may “if it thinks fit” restrict the scope of this search expedition.  

As noted above, this regime was designed to maximise state interests. Importing this regime in 1973 was a dubious decision. Applying it in 2021 to digital devices which are nothing short of portable vaults full of sensitive personal data, is a disastrous one. 

Issue 2: The Right against Self-Incrimination and Unlocking Mobile Phones

The High Court held that compelling a person to give up a password and / or biometrics to unlock a digital device did not attract the fundamental right against compelled self-incrimination as it was not the kind of evidence protected by the prohibition: providing a password did not disclose anything incriminating, and it was not the “testimonial compulsion” which Article 20(3) sought to protect. As a result, adverse inferences could be drawn if a person refused to comply with court orders. This analysis was coupled with portraying disaster if the view was taken to its logical consequence, which according to court: 

“… would result in a chaotic situation: no blood sample could be taken; no sample for DNA analysis could be taken; no handwriting samples can be taken; no other body sample for the purpose of DNA analysis could be taken; no search of a house or office could be undertaken; the data of a laptop or computer or server cannot be accessed by the investigating officer; offences like cyber crime could never be investigated; offences like pornography, child pornography which are more often than not, on the internet, could not be investigated.”

Both these conclusions of the High Court are, unfortunately, incorrect. The legal position is misstated, and the approach on the factual aspects is deeply misguided and troubling.

The High Court called upon the “testimonial compulsion” concept in its reasoning and concluded that the furnishing of a password / biometrics was not of this nature, but was akin to “physical evidence”. This binary logic was engrafted upon Article 20(3) by the Supreme Court in Kathi Kalu Oghad [AIR 1961 SC 1808], according to which there is a kind of material called “physical evidence” which falls outside the scope of the protection and persons can be compelled to furnish it. This includes blood samples, hair samples, or even asking an accused to wear specific clothing. Then, there is “testimonial compulsion” which is the material Article 20(3) covers, which traditionally makes one think of confessions. What is the basis behind this distinction? “Physical evidence” is only relevant for purposes of comparison and so by itself it is not incriminating — police take the sample to compare it with other material. “Testimonial compulsion” is incriminating by itself, and conveys to the police information that is the direct product of testimony. The key then is whether the testimonial act — be it speaking, or making gestures — conveys information that can help furnish a link in the chain of evidence, by its own merit.

Is giving the password / biometrics really not conveying any information? Is it of no value as testimony by itself? Surely, the answer is no. At its most basic formulation, the testimonial value in having an accused person unlock the phone lies in the many inferences that can be drawn from this act. Not only does it lend support to the inference that the accused owns the phone, but also to the inference that the accused was in control of its contents. And where the contents of this device are what are potentially incriminatory, surely this is as obvious a link in the chain of evidence as any. 

On a more specific point, while the Court relied upon Kathi Kalu Oghad, it seemingly ignored the decision in Shyamlal Mohanlal Choksi [AIR 1965 SC 1251]. Otherwise, the High Court could not have observed that giving a password is not testimonial compulsion because “it is only in the nature of a direction to produce a document.” After all, Shyamlal specifically held that a direction of this nature could not be issued to an accused person as it would run contrary to Article 20(3).    

Since the High Court was incorrect in comparing the giving of a password with giving of bodily samples and the like, it is already obvious that the “heavens will fall” approach to the consequences of concluding an Article 20(3) violation are an exaggeration unfounded in the law. But let’s ignore that for a minute, and take up the assertion on its face value. What it reveals is a troubling state of affairs where the High Court assumes that cooperation by an accused is necessary to secure any or all of these obviously legitimate investigative aims. This cannot be further from the truth and, in fact, the Court itself alludes to this when at a later point in the decision it recommends that police proceed to “hack” a device to gain access where the accused refuses to cooperate. What is troubling here is that this kind of piggybacking upon an accused to secure investigations is what a protection against self-incrimination, in its myriad forms, was designed to reduce. Technological advances have made it more possible for police to do their job independently and have helped to usher in a situation where investigations are not subject to the sweet will of an accused, and at the same time are free of any potential taints of accused persons being assaulted to secure information. It is unfortunate that the High Court endorsed a view which still sought to place the accused as the focal point of a police investigation, without appreciating the well-established perils of this approach.  

Issue 3: The Right to Privacy

The High Court in Virendra Khanna was keenly aware of the potential invasion of one’s privacy at stake considering how much data is found on our digital devices. It acknowledged that once police gain access to a device, even if for a specific reason, that often enables full-blown access to all aspects to a person’s life. After heading in this direction, the High Court simply noted that the use of any such data during the course of investigations would not amount to a violation of the right to privacy, as it was protected under the exceptions carved out. At the same time, the High Court observed that unlawful disclosures of this material with third parties could certainly amount to an actionable wrong.

With all due respect, the High Court’s analysis of the privacy issues barely scratched the surface and, in effect, simply placed the cart before the horse. Yes, a criminal investigation can certainly require invasions of the right to privacy that are otherwise prohibited, but to confer a blanket protection over all all kinds of activities that may be done under the pretext of an investigation effectively extinguishes the fundamental right altogether. Let’s take an example. The police allege that an accused spoke to other conspirators over email and this correspondence is evidence to show the existence of an agreement to commit a crime. This is as genuine a law enforcement need as can be justifying going inside an email account and one’s private chats. 

According to Virendra Khanna, when faced with this situation a court should support untrammelled access for police agencies to the email account. Such an approach is hardly the only way out and actually asks courts to forsake their responsibility of crafting a proportionate intrusion to best safeguard law enforcement interests without sacrificing one’s privacy altogether. Rather than confer a carte blanche upon the police, an approach which took privacy seriously — the respect a fundamental right deserves — would have a court consider if the police could demonstrate with reasonable particularity what they hoped to find or if it was just a hunch and, importantly, create a time-limit so that the individual is not forever beholden to police snooping through her inbox. 

The seemingly benign way in which the High Court viewed potential breaches of the right to privacy can be seen not only from how it viewed state interests as an unquestionable concept, but also in how it failed to address what remedies may lie in the event of a breach. As mentioned above, the High Court did note that disclosures to third parties were possible and could constitute a breach, but it neither offered nor suggested a remedy to the aggrieved accused in this regard. What’s worse, the High Court in its support for getting search warrants endorsed the regular position that the fruits of an illegal search could still be admissible as evidence. It failed to engage with the small but significant line of recent cases where another High Court took strong exception to searches being conducted without following procedures, noted that this amounted to a breach of the right to privacy, and excluded material gathered pursuant to search from being considered as evidence. 

Conclusion: Setting Back the Clock, by Some Measure

On its face, the petition in Virendra Khanna offered a straightforward issue — administering polygraph test without consent. The High Court looked past this simplicity to address the underlying legal questions which are becoming critical in their relevance to law enforcement needs and ordinary life. That it chose to do so and contributed to the discourse by offering clear answers to some questions was a welcome move. The problem is that the answers themselves are severely wanting, either proceeding on an incorrect legal basis or drawing exaggerated hypothetical conclusions. 

How this judgment is treated by the other benches in the Karnataka High Court, the state police, as well as other courts, will be interesting to see. 

The Supreme Court on Mandatory Voice Samples – II: The Rise of the Executive Court

[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

In the previous post, Abhinav Sekhri highlighted some of the flaws with the Supreme Court’s judgment on the mandatory taking of voice samples from an accused party. In this post, I want to build upon Sekhri’s arguments, and make two further points.

To start with, in order to prevent misunderstanding, let us get a few basic points out of the way. The debate before the Court wasn’t so much about whether the compelled taking of voice samples would violate the guarantee against self-incrimination under Article 20(3) of the Constitution. The two disagreeing judges who referred the case to this three-judge bench did not disagree on that point, and most of the High Courts did not either. After Kathi Kalu, indeed, it would be difficult (although not impossible) to argue that the taking of voice samples is hit by the Article 20(3) bar.

Nor was the case about whether compelled voice samples violated the fundamental right to privacy. After Puttaswamy, this would be an interesting legal question to analyse, but it was not the question that the Court was called upon to answer. No: what the case was about was whether there existed a statutory basis upon which voice samples could be taken without consent. It was this question that the Court answered in the negative, and then went on to “fill the gap” by using Article 142, and allowing for it anyway.

Why does this matter? It matters because this background is key to understanding just how flawed – and how dangerous – paragraph 24 of the Court’s judgment is. Paragraph 24 states:

Would a judicial order compelling a person to give a sample of his voice violate the fundamental right to privacy under Article 20(3) of the Constitution, is the next question. The issue is interesting and debatable but not having been argued before us it will suffice to note that in view of the opinion rendered by this Court in Modern Dental College and Research Centre and others vs. State of Madhya Pradesh and others, Gobind vs. State of Madhya Pradesh and another and the Nine Judge’s Bench of this Court in K.S. Puttaswamy and another vs. Union of India and others the fundamental right to privacy cannot be construed as absolute and but must bow down to compelling public interest. We refrain from any further discussion and consider it appropriate not to record any further observation on an issue not specifically raised before us.

There is a lot to unpack in this paragraph. The first thing to note is that this is a rank sleight of hand. The reason why Puttaswamy and privacy was not raised before the Court was that there was no reason for it to be raised! As I mentioned above, the questions before the 2012 Bench – where the two judges disagreed – were two-fold: whether the taking of voice samples was hit by the Article 20(3) bar, and if not, whether it was affirmatively authorised by any provision in the Code of Criminal Procedure. The two judges agreed that 20(3) did not apply, and disagreed on whether statute applied. Now, if the Supreme Court intended to override the absence of statutory authorisation by invoking Article 142, it would have to – necessarily – examine first whether there existed any other statutory or constitutional bar that would prevent it from doing so (because recall that Article 142 cannot be applied contrary to statute or the Constitution). In other words, therefore, in applying Article 142, the Court could not simply dodge the privacy question by saying that it was “not specifically raised before us.” That excuse went out of the window the moment Article 142 came in through the back door.

In other words, the Court would have it both ways. Expand the scope of the case, of its own accord, to issues that were never in the frame, and then dodge an examination of those issues by claiming they were never argued. The only possible reaction to this sleight of hand begins at 30 seconds of this video:

Consequently, it was not enough for the Court to give us the boilerplate phrase of “privacy is not absolute and must bow down to compelling public interest.” Nobody has ever made the claim that privacy is “absolute”, and the Court gets no brownie points for tearing down that little straw-man for the umpteenth time.

But there is a deeper problem with paragraph 24, which would preclude the above discussion from even happening in the first place. And that is – as Sekhri pointed out in his post – the established position that the word “law” under Articles 19 and 21 does not include judicial orders. This point has been discussed at some length before, and I summarise it here: our constitutional scheme envisages a two-step process before an infringement of rights can be justified. First, there must exist State-enacted law that, prima facie, is found to infringe a right. And secondly, upon a challenge, a Court must review that law to check whether it is a reasonable restriction upon the fundamental right – and strike it down, if not. The point of this two-step process is to ensure that rights are adequately safeguarded, and that the supreme deliberative and elected body – Parliament – must take the first decision in that regard. For very good reason, therefore, the Constitution does not grant power to the Courts to infringe rights of their own accord.

The implication of this, therefore, is that it was not for the Court to say that privacy was not absolute, and would have to bow down to compelling public interest. That analysis, in the first instance, would be Parliament’s to make when deciding whether or not to amend the CrPC – and for the Court to review if Parliament’s assessment was challenged. The moment that it was found that the use of Article 142 would prima facie violate privacy – which it undoubtedly does – there was no question of the Court using Article 142 powers. That – as Sekhri pointed out – is the lesson of Kharak Singh and every other case after it, until Rupa Ashok Hurra.

This brings me to my second, broader point. The Supreme Court’s judgment reflects perhaps most starkly a phenomenon that we can label as the “Executive Court.” What began as the Supreme Court stepping in to fill a vacuum left by an inactive Executive, is now transforming into the Supreme Court becoming indistinguishable from the Executive. Instead of the rule of law, protection of rights, precedent, the separation of powers, and judicial reasoning (all hallmarks of a constitutional court) guiding its decisions, the executive logic of an undefined, larger “public interest” (that only the Court knows), which is to be achieved by tossing aside all other constraints that may stand in the way, has come to dominate important cases (the NRC is perhaps the starkest example of this). So: Parliament has made no provision for the taking of voice samples, and without Parliament’s approval, executive agencies cannot act? No problem – the Court will step in and sanctify those actions judicially. This, for obvious reasons, is dangerous in many respects.

In the present case, this is revealed most starkly by the Court’s repeated invocation of the mantra, “procedure should be the handmaiden of justice.” As Sekhri pointed out in his post, this is yet another sleight of hand: the Court effectively borrowed this phrase from the context of civil litigation – conducted between parties of equal standing – and applied it to criminal litigation, where procedural safeguards are literally all that stand between the individual and the unchecked might of the State. The fact that the Court viewed procedural safeguards as impediments to achieving “true justice” is the clearest statement yet of how indistinguishable it seems to have become from the Executive.

Perhaps the Court should have dug a little deeper into its own history. Had it done so, it would have found another, telling quote: “the history of personal liberty,” wrote a certain Justice Hans Raj Khanna, in a certain well-known dissent, “is largely the history of insistence upon procedure.”

Guest Post: The Supreme Court on mandatory voice samples – I: Some glaring conceptual errors

[Editorial Note: Justice is an indivisible concept. We cannot, therefore, discuss contemporary Supreme Court judgments without also acknowledging the Court’s failure – at an institutional level – to do justice in the case involving sexual harassment allegations against the Chief Justice. This editorial caveat will remain in place for all future posts on this blog dealing with the Supreme Court, until there is a material change in circumstances.]

(This is a guest post by Abhinav Sekhri, the first in a two-part series. On 2nd of August, a three-judge bench of the Supreme Court decided the case of Ritesh Sinha v State of UP, dealing with the question of whether an accused could be compelled to provide voice samples during the course of a criminal investigation. The issue had divided many High Courts – some of which had returned detailed judgments – and had also divided a two-judge bench of the Supreme Court, where the judges had disagreed whether the Code of Criminal Procedure authorised this form of compulsion. The three-judge bench was called upon to resolve the issue. The Bench found that there was no statutory authorisation for the process; but it then noted that “procedure was the handmaiden of justice”, and decided to “fill in the gap” by invoking its powers under Article 142 of the Constitution to do “complete justice”, and authorised Magistrates to demand voice samples anyway.)

“In the light of the above discussions, we unhesitatingly take the view that until explicit provisions are engrafted in the Code of Criminal Procedure by Parliament, a Judicial Magistrate must be conceded the power to order a person to give a sample of his voice for the purpose of investigation of a crime. Such power has to be conferred on a Magistrate by a process of judicial interpretation and in exercise of jurisdiction vested in this Court under Article 142 of the Constitution of India. We order accordingly and consequently dispose the appeals in terms of the above.”

This decision by a three-judge Bench of the Supreme Court in Ritesh Sinha v. State of U.P. [Crl. Appeal 2003 of 2012, decided on 02.08.2019 (“Ritesh Sinha”)], is the latest iteration of the Court using its extraordinary powers under Article 142 for “doing complete justice”. This time, “complete justice” has been done by (i) ignoring elementary differences between civil and criminal procedure; (ii) creating new law without any modicum of precision or clarity, and; (iii) by ignoring past precedent on using Article 142 itself, with the Court choosing, of its own volition, to give the law a course that necessarily infringes fundamental rights. 

These features have adorned previous decisions of the Supreme Court as well. But very rarely do we get an example of all three working together with such destructive potency, as it has been done in Ritesh Sinha to achieve the result mentioned at the outset of this post.

Civil v. Criminal Procedure   

Throughout Ritesh Sinha, the Court is at pains to remind us that procedural law mustn’t become a thorn in the side of substantive justice — instead, procedural law must serve its role as the “handmaiden” of justice. The opinion turns to old cases, as if to suggest that its views are hardly new but rather a reflection of time-honoured tradition. But this is a ruse, dear reader: By relying upon the distracting prose of prior precedent, Ritesh Sinha tries to throw sand in your eyes. 

Why? Because all of those flowery words are germane to the context of civil litigation. Here, the idea of procedural law is to help ensure litigation places parties on an equal footing for the Court to do justice. It does so by removing privilege and secrecy, bringing before the Court all possible material and giving both sides equal opportunities to advance their case. Therefore, when there are doubts about what course procedural law should take, the answer is clear: take the path that helps the litigation to proceed.

If civil procedure is planet Earth, then criminal procedure is the moon. Police investigations are, by definition, not a situation with two equal players. Till the time a case is under investigation, the police hold all the cards. If it is a cognizable case, then the police can go and arrest anyone, question anyone, practically conduct searches anywhere, to gather material. And there is no duty to share this material with the subject of a future prosecution: an accused person does not even have the right to get a copy of the FIR registered against her.

Procedural law is not a handmaiden of justice in this context: it is the only sword and shield in the hands of an ordinary person to resist the truly awesome might of the State. It is the only rope I can cling onto for securing some measure of restraint against the State when it proceeds down the brutally invasive path of a criminal investigation. It is, also, the golden thread that protects an individual from being condemned unfairly to a conviction and the badge of guilt and shame that it must carry.

But in its earnest efforts at “doing complete justice”, the Supreme Court seems to have ignored these differences. If not, then the Court seems to have imagined a different criminal justice system from the one that most ordinary Indians are subject to. Perhaps it would help if rather than only quote prior precedent, the Court resume past practices of actively engaging with the various facets of the criminal justice system by doing surprise inspections at prisons and police stations. Maybe then the Court might see the police practices that coerce an innocent person to implicate herself falsely just to end the brutal harassment of interrogation. 

Where is the Law?

Ritesh Sinha is not the first or the last time that the Court decides to “fill the gaps” and create law rather than wait for Parliament to do its job. But, if you are deciding to deal another hammer blow to the system of the separation of powers and create law, you might at least do a decent job of it. Previous instances have included detailed guidelines being passed in Vishaka, as well as a fully detailed draft legislation being blessed in Prakash Singh; so one would expect that the Court create judicial legislation with some measure of precision.

Sadly, that doesn’t happen. The paragraph quoted at the outset is all we have by way of a law. Rather than pen a paragraph with the many gaps left behind by the Court in its effort to plug a single one, below is an illustrative list of questions, based on all that is left unclear:

  • What kind of Judicial Magistrate is conferred the power?
  • Is the Court acknowledging that no existing provision in the Cr.P.C. could confer this power on Magistrates?
  • Will a Magistrate / Sessions court not have the power to compel persons to give voice samples during trial? (since the judgment specifies investigation)
  • Who will take the sample? Must it be before officials of the Forensic Science Laboratory or can the Investigating Officer take samples? Or will they be taken before Court?
  • Is custody a necessary requirement for taking voice samples? 
  • What kind of questions can be asked during this process? Will earlier guidelines offered by the Court in the Zee extortion case be valid?

None of these questions are immaterial, and all of them are bound to start coming up sooner or later before courts, to clog our system with more cases and contribute to rising delays. That is the cost of “doing complete justice”, and not waiting for Parliament to do its job.

Atlas, Shrugged?

The premise behind the title of Ayn Rand’s famous novel was simple: Atlas, the Greek Titan, is supposedly keeping the Earth aloft on his shoulders; so what happens if Atlas shrugged? Apply this in context of the Indian Constitution and the fundamental rights of citizens, and we find that the Supreme Court has been placed in the role of Atlas. It bears the brunt of our crazy system on its shoulders, to ensure that the fundamental rights of persons are protected as the cogs of this machine keep ticking. Ritesh Sinha seems to suggest that our Atlas has shrugged, and it is perhaps time to start rethinking our approach to protect fundamental rights in the Indian polity.

The signs had come at least a decade ago, when the Court used Article 142 in Shahid Balwa [(2014) 2 SCC 687)] to go ahead and completely remove the constitutional right of appeal under Article 226 of persons who had been embroiled in the “2G Scam” cases. On that occasion, this removal of right was justified on a balancing logic, keeping in mind the overwhelming public interest in a quick prosecution (but it still took almost a decade to complete the trial, which ended in acquittals). This resurfaced in the “Coal Scam” cases, where again the same tactic was used and justified away. 

What makes Ritesh Sinha different, is that while the “2G Scam” and “Coal Scam” orders also involved a clear derogation of rights, this was limited to a particular class of cases. This time around, the criminal procedure code itself stands amended by the Court through Article 142, in spite of having noted the possible derogation of fundamental rights that this involved: because in paragraph 24 of Ritesh Sinha, the Court concedes how, at least at the threshold level, its decision can infringe the fundamental right to privacy.

Now at this point, the Court was bound to follow prior precedent set by a Constitution Bench in Prem Chand Garg [AIR 1963 SC 996], where it was held that Article 142 cannot be used to pass any orders inconsistent with fundamental rights. But there is no mention of Prem Shankar Shukla in the judgment. Instead, after noting how the fundamental right to privacy can be infringed, the Court in Ritesh Sinha justifies the derogation on grounds of “compelling public interest.”

Even if you disagree with the idea of precedent and that the sound restraint advised in Prem Chand Garg should not be controlling subsequent benches, the Court’s attempt in Ritesh Sinha to justify derogations from the fundamental right to privacy is flawed at a very basic level. In the haste to limit the fundamental right, the Court somehow forgets that such restraints must be through procedure established by law, and that judicial orders do not fall within the definition of “law” for this purpose. It is painfully ironic that the one privacy decision that ignored the Court’s attention in Paragraph 24, i.e. Kharak Singh, saw this issue being discussed at some length.

Conclusion: History Repeats Itself

In August 1961, an Eleven Justices’ Bench of the Supreme Court delivered separate opinions in State of Bombay v. Kathi Kalu Oghad [(1962) 3 SCR 10]. In Oghad as well, the Court chose to limit a fundamental right [Article 20(3)] by operating on unrealistic presumptions about criminal investigations involving an exchange between equal participants. 

In August 2019, nearly sixty years later, the Supreme Court has made the same mistake. As I have argued recently, some reasons behind the outcome in Oghad were probably the peculiar politics of 1961 — police reform was considered imminent, emboldening the Court to show more faith in the police by allowing it more investigative leeway. Maybe the only way to make sense of the judgment in Ritesh Sinha is also by turning to the peculiar politics of our time, and removing naive notions of the Court being a counter-majoritarian protector of our civil rights.  

The Gujarat High Court’s Voice Spectrography Judgment, and the Limits of Selvi vs State

On the 18th of January, in Devani vs State of Gujarat, a single judge of the Gujarat High Court handed down an interesting ruling on the constitutionality and legality of voice spectrography as an investigative technique during criminal proceedings. The writ petitioner had been charged with offences under the Prevention of Corruption Act. The main piece of evidence was a telephone conversation between the petitioner and the person from whom he had allegedly demanded a bribe. The Investigating Agency wanted to subject the Petitioner to a voice spectrography test, so that it could compare the two voice samples. The Petitioner challenged this.

The Court’s decision was delivered in the context of a split opinion by a two-judge bench of the Supreme Court, delivered in 2012, in Ritesh Sinha vs State of U.PJustice Aftab Alam had held that in the absence of express statutory authorisation, investigative agencies could not compel an accused to undergo a voice spectrography test. Justice Ranjana Desai had disagreed. While the two judges were in apparent agreement over the proposition that voice spectrography did not violate the right against self-incrimination under Article 20(3) of the Constitution, in view of the disagreement between them, they referred the case for resolution by a three-judge bench. This bench has not yet been constituted.

In Devani, the Gujarat High Court agreed with Justice Alam and disagreed with Justice Desai, holding that the investigating authorities could not legally compel an accused to undergo a voice spectrography test. The Court based its judgment on statutory interpretation, finding no warrant for the taking of voice samples under the existing criminal procedural framework. However, it also found that the taking of voice samples did not violate Article 20(3) of the Constitution. It is to this finding that we turn first.

Article 20(3)

The Court’s examination of the self-incrimination issue took place – as it had to – in the context of the Supreme Court’s 2010 decision in Selvi vs State (previously, on this blog, we have discussed Selvi and the right against self-incrimination under the Indian Constitution in some detail). Readers will recall that Selvi – a three-judge decision of the Supreme Court – had clarified some of the conceptual underpinnings of the right against self-incrimination that had been left open in the previous judgment of Kathi Kalu Oghad. The question in this case was how the Gujarat High Court would interpret Selvi, and how it would thread the needle between Kathi Kalu Oghad, which had held the taking of fingerprints and handwriting samples was consistent with Article 20(3), and Selvi, which had held that narco-analysis, brain mapping, and the polygraph test, were not. In other words, what was the deeper conceptual basis that underpinned the spectrum between fingerprints and narco-analysis, and where did voice spectrography fall on this spectrum?

The Gujarat High Court’s reading of Selvi was as follows: Selvi, it held, had prohibited the compulsory psychiatric examination of the accused. However, a voice spectrography test – which only involved the subject speaking something into a recorder, and subsequent analysis of his voice patterns – did not come within the ambit of “psychiatric examination“. Consequently, voice spectrography was not hit by the Selvi bar, and was therefore constitutional (paragraph 44).

It is submitted, with respect, that the High Court’s reading of Selvi is unnecessarily narrow, and is not borne out by the decision and reasoning in that case. Recall that, in the last analysis, in Selvi, the intellectual foundation of the Court’s judgment was the idea of mental privacy. In paragraph 160 of that judgment, the Court held:

Even though the actual process of undergoing a polygraph examination or a BEAP test is not the same as that of making an oral or written statement, the consequences are similar. By making inferences from the results of these tests, the examiner is able to derive knowledge from the subject’s mind which otherwise would not have become available to the investigators. These two tests are different from medical examination and the analysis of bodily substances such as blood, semen and hair samples, since the test subject’s physiological responses are directly correlated to mental faculties. Through lie-detection or gauging a subject’s familiarity with the stimuli, personal knowledge is conveyed in respect of a relevant fact.” (Para 160)

On, in other words, the right against self-incrimination (among other things) was aimed at protecting the mental inviolability of an accused during the course of a criminal proceeding.

The Court in Selvi went on to note:

“The compulsory administration of the impugned tests impedes the subject’s right to choose between remaining silent and offering substantive information. The requirement of a `positive volitional act’ becomes irrelevant since the subject is compelled to convey personal knowledge irrespective of his/her own volition.”


While the ordinary exercise of police powers contemplates restraints of a physical nature such as the extraction of bodily substances and the use of reasonable force for subjecting a person to a medical examination, it is not viable to extend these police powers to the forcible extraction of testimonial responses. In conceptualising the `right to privacy’ we must highlight the distinction between privacy in a physical sense and the privacy of one’s mental processes… so far, the judicial understanding of privacy in our country has mostly stressed on the protection of the body and physical spaces from intrusive actions by the State. While the scheme of criminal procedure as well as evidence law mandates interference with physical privacy through statutory provisions that enable arrest, detention, search and seizure among others, the same cannot be the basis for compelling a person `to impart personal knowledge about a relevant fact’. The theory of interrelationship of rights mandates that the right against self-incrimination should also be read as a component of `personal liberty’ under Article 21. Hence, our understanding of the `right to privacy’ should account for its intersection with Article 20(3)… a conjunctive reading of Articles 20(3) and 21 of the Constitution along with the principles of evidence law leads us to a clear answer. We must recognise the importance of personal autonomy in aspects such as the choice between remaining silent and speaking. An individual’s decision to make a statement is the product of a private choice and there should be no scope for any other individual to interfere with such autonomy, especially in circumstances where the person faces exposure to criminal charges or penalties. Therefore, it is our considered opinion that subjecting a person to the impugned techniques in an involuntary manner violates the prescribed boundaries of privacy. Forcible interference with a person’s mental processes is not provided for under any statute and it most certainly comes into conflict with the right against self-incrimination.” (Paras 190 – 193)

Consequently, Selvi went much further than simply prohibiting compelled “psychiatric examinations”. It located its decision within the framework of personal autonomy, the right to remain silent, and mental privacy.

However, it is precisely within this context that the issue of voice spectrography attains an almost intractable difficulty. If you look at the issue from one perspective, it seems clear that voice spectrography falls squarely within the Selvi prohibition. The mind is centrally involved in the act of speaking, and if “mental privacy” is to be understood in this sense, compelling a person to speak, and then using his voice to determine other relevant facts in a criminal proceeding, is certainly an invasion of mental privacy. Compelling a person to speak certainly violates their choice between speaking and remaining silent. It is certainly an interference with her “mental processes“. And finally, it does seem that what happens in a voice spectrography test is very similar to polygraph tests and brain mapping – bodily stimuli are mapped and measured. The fact that in one case, it is stimuli from the brain, and in another, voice patterns, ought not to make a difference, because – as discussed above – speaking is a direct result of mental activity.

Examined another way, however, the issue is much more complex. While speaking is a result of brain activity, the crucial thing to note in a voice spectrography test is that it is not about what the speaker is saying – that is, not about the content of her voice – but about her voice as a bodily, physical phenomenon. In this sense, voice spectrography seems much closer to fingerprints, blood samples, and DNA, than it does to narco-analysis. In a similar sense, it seems a stretch to say that the examiner is deriving knowledge from the subject’s mind. Although speaking is a result of a mental process, given that the brain is, ultimately, responsible for keeping the human body alive, every other bodily function can, ultimately, be traced back to the brain. Here, again, given that the content of what the speaker says is not at issue, it does seem that voice samples are physical phenomena, in the same sense that blood or DNA is.

What this reveals, I suggest, is that ultimately, Selvi’s neat distinction between the physical and mental, between the taking of physical samples (blood, DNA, hair swabs, fingerprints) on the one hand, and interference with mental processes and invasion of mental privacy on the other, breaks down on closer examination. It breaks down in cases such as voice spectrography, where the physical/mental binary simply loses its valence as an explanatory device.

As I had argued in my essay on Selvi, the distinction between the physical and mental in that case was drawn in an attempt to strike a balance between two models of the criminal process: the crime-control model, which seeks to remove obstacles from investigating agencies in their attempt to discover the “truth”, and the due process model, which places great stress on the fundamental rights of the accused. By limiting the right against self-incrimination under Article 20(3) to invasions of mental privacy, the Court attempted to balance the two models. A case like Devani suggests that more work needs to be done to achieve a clear and coherent balance.

Statutory Analysis

Although not strictly within the scope of this blog, a word ought to be said about the second part of the judgment. The Court undertakes a thorough analysis of laws such as the Identification of Prisoners Act, the Code of Criminal Procedure, and the Evidence Act, all of which allow for various situations – and methods – by which investigating agencies and judges can act upon the body of the accused (compelling her to yield blood samples, DNA, sputum, hair swabs, fingerprints etc.) The Court finds that nowhere is there an express authorisation to take a voice sample, and nor can such an authorisation be read into the statutes. Consequently – and despite its misgivings on this issue as a matter of policy – the Court holds that, under the existing legal regime, compelled voice spectrography is not permissible. The Court’s analysis – from paragraphs 47 to 100 – repays close study, not simply because it is an excellent piece of statutory interpretation, but also because the Court subordinates its own sense of what might be desirable as a matter of legal policy to an autonomous reading of the existing law as it stands, and refuses to step an inch beyond the law. This is a phenomenon that has been falling distressingly out of fashion in recent years.




Guest Post: The Right Against Self-Incrimination and its Discontents – II

(In this second and concluding post, Abhinav Sekhri, a Delhi-based criminal lawyer, discusses the application of Article 20(3) to persons “accused of an offence”)


Previously, we talked about how the words ‘person accused of an offence’ present in Article 20(3) of the Constitution of India have been interpreted by the Supreme Court. We saw that the Court had understood this phrase as describing a person subject of a formal accusation, akin to a FIR. This naturally created a lacuna for the time it takes for an informal accusation against one to become a formal accusation. While ordinary investigations for IPC offences may confer little investigative mettle to the Police before a formal accusation, the same is not the case in statutes creating socio-economic offences, such as smuggling. Statutes such as the Customs Act 1962, and the NDPS Act 1988 vest officers with extensive investigative powers before a formal accusation is levelled.

The Supreme Court during the 1960s consistently held that no self-incrimination could arise if persons were compelled to give evidence against themselves at these stages. I have expressed deep reservations about this approach, and in this part of my argument, I flesh out a possible alternative approach to answering the problem. While nothing would be better than an amendment to either the Constitution or the Cr.P.C. for providing clarity, we know how remote the possibility of such a non-political amendment getting tabled in Parliament is today, and must make do with innovating from the existing morass of laws. I thus propose that the phrase ‘person accused of an offence’ must be understood to include detention in custody by any authority during an investigation. I will elaborate on the benefits of this later, and require the reader to be content with just understanding my proposal for now. Simply put, if an authority seeks your detention for more than the 24-hour minimum period (before it needs to produce you for the first time before a magistrate), you necessarily must be able to exercise your right against self-incrimination.

Deepak Mahajan and a Lost Opportunity

The judicial history of interpreting Article 20(3) reflects a certain reluctance to re-evaluate basic premises and stick to formulas. The near-vehement consistency in the Supreme Court’s decisions M.P. Sharma v. Satish Chandra [AIR 1954 SC 300] onwards on the point is remarkable. There is not one stray decision in all those years. The scope for innovation when arguing this issue before a Division Bench in the 1990s was thus quite slim. Still, Directorate of Enforcement v. Deepak Mahajan [AIR 1994 SC 1775], is a landmark decision in its own right.

The offence in question was under the erstwhile Foreign Exchange Regulation Act 1973 [FERA], which followed in the footsteps of all socio-economic offences in vesting great investigative powers before filing of a formal accusation., including that of detention in custody [Section 35 FERA]. The Respondent here had been arrested and detained under Section 35 FERA – the issue was whether a Magistrate could remand him to judicial custody. Section 167 Cr.P.C. provided for remand by magistrates, but applied during investigation. But the Supreme Court itself had created a fiction that what officers did under the FERA and other such offences is not an “investigation”. To now apply Section 167 Cr.P.C. would mean that it was investigation, and that the person detained was a ‘person accused of any offence’ since investigation presumed formal accusation. Interestingly, the Delhi High Court held that Section 167 Cr.P.C. did not apply for these reasons.

Reading the decision, it is clear that the Supreme Court knew exactly how tricky the issue was. It reversed the conclusions of the High Court and held that Section 167 Cr.P.C. would apply to allow those arrested under Section 35 of FERA and other socio-economic offences to be remanded to judicial custody. While doing so, the Court did not extend the protection of Article 20(3) to such persons who are remanded to custody. With due respect, the basis for this decision is utter rubbish, and the Court horribly let itself down. The refusal to extend Article 20(3) was despite the admission that the “words ‘accused’ or ‘accused person’ is used only in a generic sense in Section 167(1) and (2) denoting the ‘person’ whose liberty is actually restrained on his arrest by a competent authority on a well-founded information or formal accusation or indictment.”

This was the first time in nearly 50 years that the Supreme Court had to consider the text of Section 167 Cr.P.C. together with the right under Article 20(3). The decision shows how the Court goes to absurd lengths to try and avoid this connection, where at one point it distinguishes the binding Constitution Bench decisions on Article 20(3) by saying that none of them applied to the ever-so slight issue of actual detention but only with admissibility of evidence. This is the reason why I term the decision a lost opportunity, for I argue that a far more consistent and coherent approach to Article 20(3) lies in considering it together with Section 167 Cr.P.C. and the idea of remand.

Re-Drawing Lines in Article 20(3)

The illogical approach of the Supreme Court was pushed to theoretically unbelievable limits in Deepak Mahajan where it held that persons arrested under laws such as the erstwhile FERA could even be remanded to judicial custody under Section 167 Cr.P.C. yet remain beyond the pale of Article 20(3). If nothing else, this by itself should make the reader a bit concerned about how the basic premise of Article 20(3) is being understood today. I propose a simpler alternative not located in the judicial precedent – connect the idea of remand to custody with terming someone as ‘accused of any offence’.

Why do we have a right against self-incrimination in the way we do under Article 20(3)? There is no one-size-fits-all rationale here, but several that apply. An interpretation of the right that tries to be in sync with these different claims is bound to bring a more wholesome solution to the problem. By providing this right, the legal system attempts to expel potentially unreliable evidence obtained through coercion. But protection from being compelled to incriminate oneself is also a basis to ensure protection from coercion itself, and all the necessary evils a system using coercion brings. At the same time, having an unbridled right to stay silent can naturally dent any investigation and we often see potentially sensitive legislation contain relaxations from certain legislative expressions of self-incrimination [the erstwhile Prevention of Terrorism Act 2002 allowed judges to draw an adverse inference against persons refusing to tender voice samples]. Not providing the right to every person but only those accused of any offence is the Constitutional balancing act.

The fact that Section 167 Cr.P.C. is the first place in the criminal process that the word ‘accused’ finds a mention to describe the affected person is not pithy phrasing. The reason being that a request by an officer seeking further remand to custody of any person shows that at that time that there is something in the allegations against that person. Otherwise what is the need to seek further custody in the first place? It is irrelevant whether the person so detained is not ultimately proceeded against. To hold that self-incrimination should only protect such eventual accused is solely looking to the evidentiary purposes of the right and completely ignores the idea of personal protection it entails. 24 hours also protect the needs of investigation, and is a Constitutionally prescribed limit [Article 22].

Furthermore, such a statutory approach allows for a more wholesome take on the investigative and evidentiary process at large. Having begun by looking at the Cr.P.C., we must now turn to the Evidence Act 1872; Sections 25 and 27 more specifically. These provisions address two concerns that arise from my proposal from the standpoint of both prosecution and defence. If you are forced by the police to confess before expiry of 24 hours, does that render you defenceless? Section 25 negates such a conclusion, for it says that “no confession made to a police officer shall be proved as against a person accused of any offence”. This ‘person accused’ is at the trial, remember, and would cover our initial un-protected suspect. On the other hand, what investigation can the police conduct with Article 20(3)? Section 27 of the Evidence Act becomes relevant here, and allows the limited use of information if any fact is discovered consequent to such information being provided by the accused.


The current approach to the words ‘person accused of any offence’ in Article 20(3) established from the decision in M.P. Sharma was useful but short-sighted. It was useful because it provided a simple solution to a difficult problem. It was short-sighted for it failed to consider the problem from all angles. This became evident with the growth of a peculiar form of socio-economic offences. The interaction of the right against self-incrimination with the procedure created under these offences that have come to represent a vast section of the criminal law in India, is illogical at best and horrendous at worst leaving the right utterly redundant. I am the first to admit that my alternative setup to Article 20(3) suffers from critical flaws of design. I do consider, however, the need for other ideas imperative, and the essence of my own argument as being sound.

Guest Post: The Right against Self-Incrimination and its Discontents

(In a two-part series, Abhinav Sekhri, a criminal lawyer, explores some of the problematic issues with the operation of the constitutional right against self-incrimination)

Article 20(3) of the Constitution of India is extremely fascinating. Tersely worded, it can be quoted in full: No person accused of any offence shall be compelled to be a witness against himself. Three components can be identified: (i) the protection applies to a person accused of any offence, (ii) it prohibits said person being compelled to be a witness, and (iii) this prohibition applies only to the person concerned. The issues also become clear, therefore. How do we understand ‘accused of any offence’? What amounts to one being compelled? And, of course, when is someone a ‘witness against himself’? In this series of posts, I wish to address only the first of these three issues. In this post, I look at three decisions of the Supreme Court on the point to show how the approach adopted by the Court towards interpreting ‘accused of any offence’ suffers from serious flaws, and leads to seriously problematic consequences. In the next post, I’ll continue on this thread and suggest a different approach to answering the interpretive conundrum posed here.

Phase I

Phase I concerns the first proper decision on Article 20(3), M.P. Sharma v. Satish Chandra [AIR 1954 SC 300]. The facts were simple. Investigations ordered by the Registrar of Joint Stock Companies revealed certain companies were engaging in fraudulent transactions. The Delhi Special Police Establishment (now CBI) required to seize the books of the concerns, and separate First Information Reports (FIR) were registered against different companies. Search warrants under Section 96 of the Cr.P.C. 1898 were issued and documents seized. The proprietors filed petitions under Article 226 challenging the searches as violating their fundamental rights under Articles 20(3) and 19(1)(f) of the Constitution. Only the Article 20(3) question was considered important. Eight judges delivered a unanimous decision, rejecting the claim. The Court noted the historical development of self-incrimination in India and mentioned how the idea always contained problems of balancing the needs of investigation with the protection. With this in mind, the Court explained all three of the facets identified above. I’ll restrict myself to the aspect of Article 20(3) being limited to a person accused of any offence. The Court held that the right extends to beyond the courtroom (yes, I’m cheating. This came in context of the ‘being a witness’ part), and then had to draw a line on how far did the right extend. In rather vague fashion, the Court said limited it to “a person against whom a formal accusation relating to the commission of an offence has been levelled which in the normal course may result in prosecution”. Importantly, the Court observed “whether it is available to other persons in other situations does not call for a decision in this case”.

The importance of this decision cannot be reduced despite the years that have passed. In choosing to expand the scope of Article 20(3) beyond the courtroom despite it having a qualifier (accused of any offence), the Court took an important decision. However, I argue that nevertheless the Court got this one slightly wrong. The problem is lies in the vantage point the Court adopts while viewing Article 20(3). The Court decides against a narrow approach, yes, but it still bases its decision by looking at the protection from that standpoint of the eventual proceedings in court. Compelled evidence is inherently unreliable for trial (for people lie when tortured), and such tainted evidence can arise from compulsion exercised beyond the courtroom. This reasoning is commonly touted as the basis for the protection only for such claims to be convincingly rebuffed time and again [For instance, Friendly, The Fifth Amendment Tomorrow: The Case for Constitutional Change, 37(4) U. Cin. L.Rev. 671 (1968); Amar & Lettow, Fifth Amendment First Principles: The Self-Incrimination Clause, 93 Mich. L. Rev. 857 (1995)]. The reason why focusing too much on eventual proceedings is a problem is not hard to fathom. Imagine a situation where you are not named in the FIR as an accused but are picked up by the police, forced into writing a confession, but the matter ultimately never goes to trial and are thus released from custody. The detention is not unlawful – the police can arrest suspects – so no Article 21 violation occurs. You were forced to incriminate yourself. But, according to M.P. Sharma, it is difficult to claim this as a violation of the right under Article 20(3). Difficult and not impossible, only because of the caveat, which was to get forgotten soon after.

 Phase II

The formal accusation referred to by the Supreme Court in M.P. Sharma must also be one that normally may result in prosecution (Thomas Dana v. State of Punjab, AIR 1959 SC 375). Prosecution refers to initiation of legal proceedings, i.e. a trial in this case. A formal accusation appears to have been used with the intention to create different classes of accusations. Formal here would mean ‘official’ (the other definition of the word being based on etiquette must be excluded), and the Court perhaps wanted to separate official accusations from those by private individuals. The police registering an FIR, is different from me filing a complaint against you at the local police station. So we find a gap emerge in the process; in the time when an accusation becomes a formal accusation. Here, remember, the Court issued a caveat – its judgment did not preclude an extension of the right under Article 20(3), but only provided some sort of minimum. This gap might yet be covered under Article 20(3), as the Court didn’t discuss such cases. The Court would have been wise to clothe the caveat with more flesh, as hindsight would suggest.

Phase II will look at two cases. The first is the unanimous decision of 5 judges in Raja Narayanlal Bansilal v. Maneck Phiroz Mistry [AIR 1961 SC 29]. The case was similar on facts to M.P. Sharma; the only difference being that here there was no FIR against the companies or proprietors. An Inspector appointed under the Companies Act, 1956, called upon the appellants to furnish certain information. The Appellant challenged this investigation as being inter alia, contrary to Article 20(3). While observing the earlier judgment of M.P. Sharma, the 5 judges concluded that “the effect of this decision thus appears to be that one of the essential conditions for invoking the constitutional guarantee enshrined in Art. 20(3) is that a formal accusation relating to the commission of an offence, which would normally lead to his prosecution.” Did the inquiry here amount to such an accusation? No, it was more like a fact-finding commission and “the fact that a prosecution may ultimately be launched against the alleged offender will not retrospectively change the complexion or character of the proceedings”. The unanimous answer thus being, no violation of Article 20(3). The Court seems to have been referring to the caveat to the end of its decision. It observed “even if the clause ‘accused of any offence’ is interpreted in a very broad and liberal way it is clear that at the relevant stage the appellant has not been, and cannot be, accused of any offence.” Only to then say that such a broad interpretation does not appear to be “consistent with the tenor and effect of the previous decisions of this Court”.

This leads us to the second case, another 5 judge unanimous decision in Romesh Chandra Mehta v. State of West Bengal [AIR 1970 SC 940]. This case arose out of the old Sea Customs Act, 1878 (replaced by the Customs Act, 1962). The Appellant had been searched at Dum Dum Airport, Calcutta, and this search lead to recoveries of jewels and currency worth several lakhs of rupees. He was questioned under Section 171-A of that Act (after possibly being arrested, which is not entirely clear from the judgment), where disclosures lead to further recoveries. These discoveries made pursuant to this inquiry were assailed as being the result of the Appellant being compelled to incriminate himself. The Court’s reasoning while denying that Article 20(3) extended to the case is important: “a person arrested by a customs officer because he is found in possession of smuggled goods or on suspicion that he is concerned in smuggling is not, when called upon by the Customs Officer to make a statement or to produce a document or thing, a person accused of an offence within the meaning of Article 20(3) of the Constitution. The Customs Officer does not at that stage accuse the person suspected of infringing the provisions of the Sea Customs Act … he is not accusing the person of any offence punishable at a trial before a Magistrate”. With this it would appear that the Court has taken a clear and consistent view on how to consider that gap we identified in the process. Anything prior to the formal accusation means no protection.

The Problem

Have a look at the procedure under the Customs Act 1962 and other socio-economic offences such as the Foreign Exchange Management Act 1999, and the Narcotics, Drugs and Psychotropic Substances Act 1985. The formal accusation in these cases is a Complaint filed by the authorised officer, which is the result of an investigation. Here, there is no pre-investigative accusation drawn up by the authority like an FIR required under the Cr.P.C. The reason for this is a ridiculous deeming fiction which requires that we don’t call these officers police officers and by extension call these investigations enquiries, despite the officers having the same powers of investigation as conferred under Chapter XII of the Cr.P.C. [on this, see Sekhri, Confessions, Police Officers and Section 25 of the Indian Evidence Act, 1872, 7 NUJS L. Rev. 1 (2014)].

These socio-economic offences thus have extensive pre-trial powers during the ill-phrased enquiry stage. The Customs Act 1962 (successor to the Sea Customs Act, 1871), allows customs officers powers of ‘searches, seizure and arrest’. A customs officer can summon any person [Section 108] and examine her during the course of an enquiry into the smuggling of goods [Section 107]. Persons summoned are “bound to state the truth upon any subject respecting which they are examined or make statements”, and these enquiry proceedings are deemed to be judicial proceedings for Section 193 and 228 of the Indian Penal Code. So if you lie, perjury charges can follow. On top of this, the customs officer can arrest [Section 104] upon having reasons to believe that the person committed an offence punishable under Section 132, or 133, or 135, or 135-A or 136 of that Act. Detention can and will follow, and could possibly extend to 60 days as specified under Section 167, Cr.P.C. What the Supreme Court has done, is to exclude the right of self-incrimination from this entire process because of its focus on the actual realisation of proceedings as first seen in M.P. Sharma.

 The steady growth of powers during this enquiry stage under socio-economic offences supports the theory that the Legislature is not blind to the line-drawing adopted by the Supreme Court. So what do we draw from this? Today you can be questioned under the threat of prosecution to supply potentially incriminating information. You can be arrested and detained during this process, but still the right under Article 20(3) remains beyond reach. Why? Because the accuser might not have found all that was needed from you to incriminate yourself.



Guest Post: Admissibility of Confessions to the Police: Normalising a State of Exception – II

(In this second post in a two-part series, Manasa Sundar Raman examines how, true to Agamben’s thesis, the “exceptional” remedy of making confessions to the police admissible has spilled over into “normal” judicial practice.)

In the previous post, the scope of right against self-incrimination and its dilution by Kartar Singh and PUCL were discussed. In this post, we will examine how the guidelines issued in Kartar Singh were inadequate to protect the right against self-incrimination resulting in police confessions increasingly becoming a norm.

Dilution subsequent to Kartar Singh

The guidelines suggested to the government in Kartar Singh were without adequate clarity on their applicability to future cases or even their rationale. Predictably, they have suffered dilution by subsequent Supreme Court judgments which condone non-compliance with them and the rules framed under Section 15 of the TADA.

S.N. Dube v. N.B. Bhoir, for instance, held that even if the Kartar Singh guidelines were not strictly adhered to, it would not affect the admissibility of the evidence. Shockingly, even though the accused had alleged coercion by the police officer, the plea was rejected since no specific instance of coercion was alleged against the concerned police officer. Thus, not only does the court discard the importance of the guidelines, it places the burden on the accused to show his confession was recorded in an atmosphere that was not free.

In Jameel Ahmed v. State of Rajasthan, the confessional statements so recorded by the police were not sent to the CMM or the CJM as required under Rule 15(5) of the Rules, and as per the Kartar Singh guidelines. However, the Court underplays the importance of the Magistrate and the guidelines by stating the CJM or CMM has to merely forward the confessional statement and, hence, the procedural lapse may be overlooked. It goes on to describe the Rules and guidelines as being “only directory and not mandatory.” This way, an important safeguard against the police extrapolating the confession was diluted.

Finally, in Gurdeep Singh v. The State (Delhi Admn.), one can clearly notice the kind of coercive and intimidating circumstances that can be wrought by the preceding rulings. Even though the accused was handcuffed with a policeman holding him in chains and armed guards surrounding the room, the Court dismissed the contention of a confession being non-voluntary stating that these measures were intended and essential for security purposes. As a way of justification, once again, the Court relies on the objects for the enactment of TADA and the necessity of having stringent provisions to assist the prosecution.

From the above case-laws we see how the courts in their enthusiasm to convict terror-suspects, dispense with crucial guidelines issued by Kartar Singh and Rules under TADA. Thereby, even if the ratio in Kartar Singh read with the guidelines have the effect of securing a meagre right against self-incrimination (in reality, it did not), the rhetoric of terrorism has undermined even this.

Characterising the Judicial Response

In most of the above cases, the rhetoric that is used prominently includes the image of the innocent victim of terror. This way the Court is able to depart from the traditional paradigm of a trial where the accused stands against the overwhelming might of the state machinery (and therefore has certain rights such as right against self-incrimination, right to silence, presumption of innocence) to a different paradigm where the accused has wronged the several innocent victims. The role of the trial process under the new paradigm is to strike a balance between the rights of these two individuals. By portraying the State’s citizens as victims of terror, the Court is able to turn a blind eye to his presumption of innocence or right against self-incrimination.

Moreover, this has changed the role of the higher judiciary. As pointed out by Prof. Mrinal Satish and Prof. Aparna Chandra, the judiciary has shifted from being the “sentinel on the qui vive” (as famously observed in State of Madras v. V.G. Row), duty-bound to zealously guard fundamental rights, to a pragmatic mediator that balances competing claims of ‘national security’ and ‘fundamental rights’. The pragmatic mediator attitude can once again be seen when the Court endorses appeals to observe human rights during the investigation and trial stages. This is not done as a reiteration of the State’s legal or moral duty but for instrumental reasons of preventing further terrorism. As a consequence, even in cases where the violations are blatant, there is a tendency for judges to defer to the legislature as it concerns ‘terrorism’ and prescribe guidelines, without actually striking it down.

Admittedly, the judiciary may not possess the same expertise as the legislature or the executive in evaluating counter-terror measures. However, it would be far more respectable and healthy for the rule of law if the judiciary candidly conceded that it is ill-equipped to deal with the problem of terror and, thus, give a wide leeway or ‘margin of appreciation’ to the State in such cases. Accordingly, it could have formulated a lower level of review when the object of the statute was to counter terrorism or read in a national-security exception to Article 20(3). However, we see no such admission of its inadequacy or a well-defined exception to terror-cases. In contrast, in their eagerness to uphold the statute, the judgments simply refer to the legislative competence of the state, echo the State’s high-pitch alarmist attitude to terror and dilute the entirety of the Article 20(3) guarantee.

The above argument can also be illustrated using Article 14 analysis used in Kartar Singh. In evaluating whether it was open for the legislature to make confessions before the police admissible, the judges reduce the issue to one where ‘different mode of proof’ is prescribed by the legislature for a certain class of offences. The rationality behind legislative classification is not questioned at all, instead the analysis is limited to the competence of the legislature to make such a classification without evaluating the rational nexus of such a classification to the object sought to be achieved. In not doing so, the judges leave open the question of whether it would be open for the legislature to make classifications such as ‘sexual offenders’ and ‘other offenders’; ‘white-collar crimes’ and ‘other crimes’ and make confessions before the police selectively admissible for one set? Can different classes of crimes have ‘different modes of proof’? The judgment reduces it to a question of legislative competence.

No analysis is presented with respect to Article 20(3). While the intention may have been to dilute the standards for terror-related cases alone, the judgment by not making that intention explicit jeopardises the right as a whole. Even if the Court had chosen to explicitly hold ‘terror cases’ as an exception where it is acceptable for confessions made to the police to be admissible, the substantial dilution of Article 20(3) could have been prevented. The result is that the diluted right to self-incrimination has now become the normalised standard under the Constitution.

Normalisation of police-recorded confessional statements

Over the years, we see that the Courts have become oblivious to the so called “terrorism” exception. Thus, evidence obtained under such exceptional statutes are being used for the trial of regular offences.

In State v. Nalini, the accused was charged with offences both under TADA and IPC. The Court, in paragraph 81 of the judgement held that confessional statements recorded under Section 15 of TADA would apply and become admissible as substantive evidence for all offences including the non-TADA ones. In holding this, it specifically overruled another SC decision (Bilal Ahmed Kaloo v. State of Andhra Pradesh) on point. Further, the evidence so recorded could also be used against the co-accused.

In view of the conflict between the above two cases, the issue was referred to a constitutional bench in State of Gujarat v. Prakash Kumar which upheld the Nalini rationale. It said that the confession statement recorded under Section 15 of TADA is admissible for the offences under any other law which were tried along with TADA offences notwithstanding the fact that the accused was acquitted of offences under TADA in the same trial.

In State of Gujarat v. Mohammad Atik, the issue was whether a confessional statement recorded by a police during investigation under TADA could be used for a completely different trial where investigation was done by a different set of police officers against the same accused. The Court held that in the absence of any statutory inhibition, there is no need to introduce a further fetter against the admissibility of the confessional statement. Thus, there is no cognisance of the fact that a police-recorded confession is admissible in only very exceptional cases. Once admissible, it is treated on par with any other evidence and there is no re-examination of the voluntariness of that evidence in consideration of the circumstances under which it was recorded.

From the above line of cases, we notice a disturbing trend where the Courts adopt a dangerous and cavalier approach to confessional statements recorded by the police. They are unmindful of the exceptional circumstances under which this exception was sought to be created and constitutionally justified. Instead, there is an increasing trend to issue guidelines to be complied with. These too are not rigorously enforced. This way, the extraordinary standards which are judicially condoned for addressing ‘terrorism’ slowly seep into the ordinary criminal justice system, and are normalised.

The blame for this undoubtedly falls on the majority opinion in Kartar Singh. Not only does it depart from the progressive interpretation given to the rights of the accused in Nandini Satpathy, it does not adequately frame admissibility of confessions recorded by the police as a strict exception for terror-cases. Indeed, by making it wholly a question of legislative competence, the legal position is such that if Section 25 and 26 of the Evidence Act were replaced completely by a provision akin to Section 32 of POTA, the existing precedent would make it wholly constitutional. Fortunately, the POTA has been repealed and the legislation which replaces it – the Unlawful Activities (Prevention) Act, 2008 contains no such provision. However, the constitutional guarantee has been undoubtedly diluted.

Concluding Remarks

In view of the progressive interpretation given to the right of accused against self-incrimination at the interrogation stage in Nandini Satpathy, and the absence of any exception to Article 20(3), the judgment of Kartar Singh was grossly regressive. It can even be argued that the judicially created exception for admissibility of confessions made before the police is unconstitutional.

However, even assuming it is constitutional, Kartar Singh, by not carving out a specific and deliberate exception for terror cases and by employing vague rhetoric about the necessity felt by the Parliament has severely diluted the right as a whole. Read with subsequent judgments where confessional statements made in coercive atmospheres have been held to be admissible, it is safe to say that guarantee against self-incrimination has virtually become meaningless.

More alarmingly, Courts show no hesitancy in applying confessions obtained under such extra-ordinary statutes designed to meet exceptional ends in normal criminal justice administration. This results in a normalisation of these standards.

(Manasa is a Vth Year student at the National Law School of India University)

Guest Post: Admissibility of Confessions made to the Police: From Exception to the Norm – I

(Previously on this blog, we have discussed Giorgio Agamben’s theory of the state of exception, according to which “extraordinary laws” ostensibly meant for “extraordinary times”, characterised by concentration of power in the executive, bypassing of parliamentary oversight, and incremental erosion of procedural safeguards, gradually become “normalised” parts of the existing legal landscape. In a two-part series, Manasa Sundar Raman discusses the use of extraordinary law to achieve the gradual normalisation of confessions made to police officers, that were originally considered inadmissible under the Indian Evidence Act.)

One of the chief goals of any liberal democratic Constitution, including India’s, is to guard against the growth and abuse of police power of the state and corrosion of civil liberties. Yet, it is no secret that in the recent decades, basic constitutional guarantees have increasingly come under threat and have been justified in the name of counter-terrorism. Despite struggling to define “terrorism”, most counter-terror legislations and judgments that interpret them, comfortably place terrorism as an exceptional or extraordinary crime that justifies infringements on civil liberties that are otherwise prohibited.

Characteristically, counter-terrorism legislations give enormous and overriding powers to the executive. Further, in accordance with the extraordinary concern that is sought to be addressed by such laws, they also have extraordinary provisions in procedural aspects such as bail, arrest, detention, confessions, sentencing etc. A recurring ‘extraordinary’ feature in these statutes is the admissibility of confessions made to police during interrogation.

Such extraordinary provisions inevitably lead to constitutional concerns. However, over the years, it can be observed that the judiciary is inclined to save them, paying undue deference to the government rather than declaring them as being unconstitutional. The judgments in this regard invoke the stereotypical imagery of brutalities, fear psychosis and loss of life inflicted by terrorism. This is used as a ground in itself to undermine constitutional guarantees.

In this context, one can consider the extent of deference given to legislative wisdom by the judiciary when these statutes permit confessions made to police as being admissible in trial and are challenged for being violative of Article 20(3). It is my thesis that the judiciary, by allowing for the so-called extraordinary statutes to undermine the right under Article 20(3) without constitutional basis, has gradually aided in the erosion of the right for ‘ordinary’ situations, as well. Therefore, admissibility of confession to police officer is no longer a shocking exception, but well on its way to becoming the norm.

Right against Self-Incrimination in India

Prohibition against self-incrimination is undoubtedly one the cardinal principles of criminal law. Under Article 20(3) of the Indian Constitution, the safeguard, framed as a right, extends to all accused persons to not be compelled to be a witness against themselves.

The Article 20(3) guarantee is further codified under Sections 161, 162, 163 and 164 of the Criminal Procedure Code, 1973 and such guarantees are also present in statutes pre-dating the Constitution in the form of Section 25 and Section 26 of the Indian Evidence Act, 1872. Additionally, India has also, without any reservations, ratified the ICCPR which under Article 14 provides for the right against self-incrimination “in full equality”.

The extent of the right against self-incrimination was fully recognized by the Supreme Court in Nandini Satpathy v. P L Dani. In this case, the court relied on Miranda v. Arizona, to extend the embargo against compulsion in testimony to the investigation stage as well. Further, the judgment recognises that compulsion may come in many forms i.e. not just by way of physical torture, but also in the form of psychic pressure or a coercive atmosphere. Most importantly, the decision holds the right under Article 20(3) and under the Criminal Procedure Code to be co-terminus in their protection. Thus, such strong language used by Nandini Satpathy was merely one step behind holding that provisions enabling the admissibility of confessions recorded by the police would be unconstitutional. However, we notice that subsequent judgements on Article 20(3) regress from this holding and dilute the protection altogether.

Judicial Response to Art. 20(3) challenges to Extraordinary statutes

The Terrorists and Disruptive Activities (Prevention) Act, 1987 (“TADA”) was introduced in May, 1985 as a temporary and extraordinary measure to deal with insurgent movements felt in several areas in India. However, over the years, it was notified in almost all states and became infamous as one the most abused laws.

Many provisions of the TADA, including Section 15 were challenged before a constitutional bench in Kartar Singh v. State of Punjab. Section 15 of the TADA provided that a confessional statement recorded by a police officer is admissible as evidence. In a 3-2 split, the Court upheld the section.

The judgment is prefaced with the historical background and circumstances that led to the enactment of the TADA. For a large part, the Supreme Court dwells on the legislative competence of the State to enact such a law. It then proceeds to hold the section as valid under Article 14 on the basis of the legislative classification that differentiates between ‘terrorists and disruptionists’ and ‘ordinary criminals’. Rejecting the reliance on State of W.B. v. Anwar Ali Sarkar, it reasons that classification of offences is constitutional as long as they are legislatively defined and not left to the arbitrary and uncontrolled discretion of the executive. The issue with this reasoning is that the Court refuses to dig deeper into the rationality of the legislative classification itself. Even assuming that there is an intelligible differentia between the ‘terrorists and disruptionists’ as determined by the police and ‘ordinary criminals’, the Court fails to elaborate on the rational nexus between such a difference and lesser degree of constitutional protection against self-incrimination.

Further, it holds the procedure to be just and fair under Article 21. In the reasoning for this, the majority cite a National Police Commission report that recommends that confessions made to the police to be made admissible so that it will “remove the present feeling of the police that they have been unjustly discriminated against in law”! Thereafter, once again the Court invokes the spectre of terrorism. It paints terror and disruptions as those which endanger the sovereignty and integrity of the country, the normal life of the citizens, and by linking it to the difficulty in obtaining evidence, upholds it entirely.

There are several things that are inexplicable and downright absurd in the Court’s reasoning in saving Section 15. First, it is beyond comprehension why the entire reasoning is based on Article 14 and 21 even though the Court earlier admits that Article 20(3) concerns were implicated by such a provision. There is absolutely no analysis on the extent of protection guaranteed by Article 20(3) and to what degree can it be infringed, if at all. Admittedly, the Court issues guidelines for recording of a complaint of torture by Magistrate if the accused so complains. However, it omits to state the consequence of torture on the probative value of the ‘confession’. Further, the judges limit their understanding of involuntary confessions to those made under torture. There is no mention of other methods by which a confession may be extracted such as by inducement or threats. As rightly pointed out in Nandini Satpathy, compelled testimony is not limited to those made by “physical threats or violence” alone but also as a response to “psychic torture, atmospheric pressure, environmental coercion, tiring interrogative prolixity, overbearing and intimidatory methods” as well. Therefore, there is no way for the Court to determine if the accused did in fact make the confession freely and voluntarily. Second, the judges acknowledge truth of custodial torture to obtain evidence and then shockingly, in the same breath, turn a blind eye to it by invoking the imagery of terrorists being a threat to the nation’s sovereignty.

Thus, despite powerful dissents by Justice Ramaswamy and Justice Sahai, it is unfortunate that the majority chooses to substitute constitutional analysis with rhetoric on barbarity of terrorism. The saving grace of the majority judgment is the guidelines issued by the majority such as appearance of the accused before the Magistrate and forwarding the confessional statement recording. Ostensibly, the guidelines were aimed at ensuring the voluntariness of the confession and as a protection mechanism against torture. However, this rationale is not explicitly mentioned, perhaps because it would become an implicit admission of the constitutional infirmities of the section.

The TADA was subsequently repealed but was re-incarnated as the Prevention of Terrorism Act, 2002. Section 32 of the POTA, akin to Section 15 of the TADA, allowed for admissibility of confessions made to a police officer. However, Section 32 in sub-sections (2) to (5) statutorily incorporated the guidelines issued in Kartar Singh. In PUCL v. Union of India it was argued that the since the accused has to be produced before the Magistrate within forty-eight hours, there is no reason why the police are authorised to collect confessions. The Supreme Court rejected this contention stating that a similar provision was upheld in Kartar Singh and also pointed out that it incorporates the guidelines issued therein. Once again, the Court falls into the trap of excessive deference to the determination of “necessity” by Parliament, by characterising this encroachment of a fundamental right as an issue of policy rather than principle. Moreover, just as in Kartar Singh, the Court makes no reference to the fact that the recording of confession by the police is an exceptional case, departing from the well-established rules under Evidence Act, Criminal Procedure Code and norms under Article 20(3). Thus it is uncertain from the reasoning whether the Court does view Section 32 as a justified derogation from Article 20(3) in view of terrorism or if recording of confession by police is normally allowed even for regular offences under Article 20(3).

(Manasa is a Vth year student at the National Law School of India University)

Privacy, Self-Incrimination and the Constitution – IV: Selvi and the Middle Way

Selvi vs State, decided by a three-judge bench in 2010, is the Supreme Court’s most recent – and by far most detailed – engagement with Article 20(3). Selvi involved a batch of appeals challenging the constitutionality of three investigative techniques: narco-analysis, the polygraph test, and the Brain Electrical Activation Profile. The nature of these processes is important. In narco-analysis, an intravenous injection of a drug into a subject’s blood-stream sends her into a hypnotic state, lowering her inhibitions, and making her more likely to divulge information.  In discussing the comparative jurisprudence on narco-analysis, the Court quoted both Horvath vs The Queen, which we had discussed in the last post, and the American Supreme Court case of Townsend vs Sain, in which Warren J. held that “if an individual’s “will was overborne” or if his confession was not “the product of a rational intellect and a free will,”, his confession is inadmissible because coerced.”

A polygraph test, on the other hand, measures various physiological responses (respiration, blood pressure, blood flow etc.) during questioning, and makes determinations about the truth or falsity of the subject’s statements, based on the changes in those responses.  Similarly, the Brain Electrical Activation Profile (BEAP) test measures responses within the brain, in order to ascertain whether or not the subject has recognised the stimuli to which she has been exposed.

Obviously, out of these three tests, only narco-analysis involves “testimony” in the classic sense: communicating information through words, written or spoken. One of the central issues in the case, therefore, was whether recording physical stimuli amounted to compelling a person to be a “witness against himself”. It is in this backdrop that the Court embarked upon a detailed analysis of whether these investigative procedures fell foul of Article 20(3). Up front, it laid out the conceptual foundations of the Article:

“Its underlying rationale broadly corresponds with two objectives – firstly, that of ensuring reliability of the statements made by an accused, and secondly, ensuring that such statements are made voluntarily… when a person is compelled to testify on his/her own behalf, there is a higher likelihood of such testimony being false. False testimony is undesirable since it impedes the integrity of the trial and the subsequent verdict. Therefore, the purpose of the `rule against involuntary confessions’ is to ensure that the testimony considered during trial is reliable… the concerns about the `voluntariness’ of statements allow a more comprehensive account of this right. If involuntary statements were readily given weightage during trial, the investigators would have a strong incentive to compel such statements – often through methods involving coercion, threats, inducement or deception. Even if such involuntary statements are proved to be true, the law should not incentivise the use of interrogation tactics that violate the dignity and bodily integrity of the person being examined. In this sense, `the right against self-incrimination’ is a vital safeguard against torture and other `third-degree methods’ that could be used to elicit information. It serves as a check on police behaviour during the course of investigation. The exclusion of compelled testimony is important, otherwise the investigators will be more inclined to extract information through such compulsion as a matter of course. The frequent reliance on such `short-cuts’ will compromise the diligence required for conducting meaningful investigations.” (paras 91 and 92)

The first of the Court’s rationales – reliability – straightforwardly corresponds to the crime-control model. The second – voluntariness – seems, at first sight, to correspond to the due process model, but in answering the question, “why is voluntariness important?”, the Court complicates the issue: it mentions bodily integrity and dignity (in the context of torture and other third-degree methods), but also notes that the “diligence” required for “meaningful investigations” will be “compromised” if the investigators can take short-cuts such as torture. It is unclear, therefore, that even within the “voluntariness” framework, whether the focus is on (due-process based) concerns of dignity and integrity, or of (crime-control based) diligent investigations.

After examining a plethora of precedents on self-incrimination the world over, the Court then clarified some of the basic tenets of Article 20(3): “accused of an offence” covered a wide ambit, that included people formally charged of offences, as well as people whose answers could expose them to criminal charges (paragraph 109); incriminatory statements included statements that the prosecution could directly rely upon to further its claims, as well as derivative statements. The corse of the Court’s analysis, however, was in examining whether “testimonial compulsion” was involved in the three impugned techniques. Precedent – as we have seen – drew a distinction between testimony and physical evidence (fingerprints, blood samples etc.) In Kathi Kalu, this distinction had been rationalised on the ground of “volition”, in the sense of unchangeability. Under this logic, narco-analysis would be borderline unconstitutional (depending on whether or not you take answers given under hypnosis to be “volitional” or not), whereas polygraph tests and brain-mapping would be definitively constitutional.

This argument was buttressed by the fact that Ss. 53 of the CrPC allows for the examination of the accused on the request of the police, of “blood, bloodstains, semen… sputum… sweat…” etc., through the use of “modern and scientific techniques including DNA profiling and such other tests…” It was argued that the three impugned techniques ought to be read into this Section. The Court, however, rejected this argument, holding that S. 53 clearly referred only to the examination of “bodily substances”, whereas the impugned techniques involved “testimonial responses” That, however, is not self-evident: what is it about physiological reactions of the brain and other parts of the body, that put them within the category of “testimonial responses”?

The Court answered this question by going back to Kathi Kalu, and holding that a testimonial act is equivalent to “the imparting of knowledge by a person who has personal knowledge of the facts that are in issue.” (paragraph 158) And then the Court noted:

Even though the actual process of undergoing a polygraph examination or a BEAP test is not the same as that of making an oral or written statement, the consequences are similar. By making inferences from the results of these tests, the examiner is able to derive knowledge from the subject’s mind which otherwise would not have become available to the investigators. These two tests are different from medical examination and the analysis of bodily substances such as blood, semen and hair samples, since the test subject’s physiological responses are directly correlated to mental faculties. Through lie-detection or gauging a subject’s familiarity with the stimuli, personal knowledge is conveyed in respect of a relevant fact.” (Para 160)

And, immediately afterwards:

“The compulsory administration of the impugned tests impedes the subject’s right to choose between remaining silent and offering substantive information. The requirement of a `positive volitional act’ becomes irrelevant since the subject is compelled to convey personal knowledge irrespective of his/her own volition.”

Of course, in a trivial way, being forced to provide a blood sample or a fingerprint also involves conveying personal knowledge (of your blood group or your fingerprint pattern) irrespective of your volition. In earlier cases, the Court had gotten around that by arguing that a blood sample or a fingerprint was innocuous in itself – only when subsequently corroborated with another piece of evidence (fingerprints at the scene of the crime), did it become incriminatory. However, that reasoning is not open to the Court here, because physiological responses to stimuli are also innocuous in themselves. On the other hand, it seems clear from paragraph 160, that what the Court was concerned about was – in a phrase – the privacy of the mind. This becomes clearer subsequently, when after an excursion into the constitutional right to privacy under Article 21, the Court noted:

While the ordinary exercise of police powers contemplates restraints of a physical nature such as the extraction of bodily substances and the use of reasonable force for subjecting a person to a medical examination, it is not viable to extend these police powers to the forcible extraction of testimonial responses. In conceptualising the `right to privacy’ we must highlight the distinction between privacy in a physical sense and the privacy of one’s mental processes... so far, the judicial understanding of privacy in our country has mostly stressed on the protection of the body and physical spaces from intrusive actions by the State. While the scheme of criminal procedure as well as evidence law mandates interference with physical privacy through statutory provisions that enable arrest, detention, search and seizure among others, the same cannot be the basis for compelling a person `to impart personal knowledge about a relevant fact’. The theory of interrelationship of rights mandates that the right against self-incrimination should also be read as a component of `personal liberty’ under Article 21. Hence, our understanding of the `right to privacy’ should account for its intersection with Article 20(3). Furthermore, the `rule against involuntary confessions’ as embodied in Sections 24, 25, 26 and 27 of the Evidence Act, 1872 seeks to serve both the objectives of reliability as well as voluntariness of testimony given in a custodial setting. A conjunctive reading of Articles 20(3) and 21 of the Constitution along with the principles of evidence law leads us to a clear answer. We must recognise the importance of personal autonomy in aspects such as the choice between remaining silent and speaking. An individual’s decision to make a statement is the product of a private choice and there should be no scope for any other individual to interfere with such autonomy, especially in circumstances where the person faces exposure to criminal charges or penalties. Therefore, it is our considered opinion that subjecting a person to the impugned techniques in an involuntary manner violates the prescribed boundaries of privacy. Forcible interference with a person’s mental processes is not provided for under any statute and it most certainly comes into conflict with the right against self-incrimination.” (Paras 190 – 193)

The shift from Kathi Kalu is crucial. In that case, “volition” – in the sense of changeability – played the crucial role, and we saw how it was conceptually connected with the crime-control model: information that you had no power to change could not possibly be fabricated. In Selvi, although the Court embarked upon a discussion of the reliability of the investigative techniques, ultimately, the distinction it drew was between “physical privacy” (blood samples, fingerprints) and “mental privacy”; linking the word “witness” to testimony, and then understanding “testimony” as the impartation of information present within a person’s mental sphere, the Court placed a certain conception of mental privacy – understood as autonomous mental processes – at the heart of the guarantee against self-incrimination.

While the result of Selvi was the unconstitutionality of three specific investigative procedures, its implications for criminal/constitutional jurisprudence are more significant. In Selvi, we have a strong recognition of the role of the due process model as the foundation of criminal procedure and associated constitutional guarantees. It therefore provides a template for future cases where the crime-control model and the due process model pull in opposite directions, and the Court is obliged – as it was in Selvi – to balance the two.