In June and October 2021, the Delhi and Bombay high courts had passed two important judgments on bail under the Unlawful Activities Prevention Act (UAPA). The Delhi high court’s judgment(s) granted bail to Asif Iqbal Tanha, Devangana Kalita, and Natasha Narwal, who were accused of various UAPA offences arising out of the February 2020 Delhi riots. The Bombay high court granted bail to Iqbal Ahmed Kabir Ahmed, who was accused of criminal conspiracy involving members of the banned ISIS group.
The significance of these judgments – I had written at the time – lay in how they articulated a “jurisprudence of liberty” within the stringent confines of the UAPA. The high court justices addressed themselves to the fact that section 43(D)(5) of the UAPA – as interpreted in the Supreme Court judgment of Watali – effectively makes the grant of bail impossible, and ensures that people are jailed without trial for years. In response, the courts formulated two principles that would guide judicial determinations of bail under the UAPA. In a post about the Supreme Court’s (similar) judgment in Thawha Faisal’s case, I summarised these principles thus:
Principle 1: The definitional clauses of the UAPA must be given a strict and narrow construction. This was what the Delhi high court did with respect to the meaning of “terrorism” in Asif Iqbal Tanha, what the Bombay high court did with respect to Section 20 of the UAPA in Iqbal Ahmed Kabir Ahmed, and what the Supreme Court has now done in Thwaha Faisal.
Principle 2: The allegations in the chargesheet must be individualised, factual, and particularistic. The gap between what an individual is accused of, and the actual events, cannot be filled by inferences or speculation. This, again, is exhibited excellently in the analysis in the Thwaha Faisal judgment.
At the time, one hoped that these judgments – especially supported by the Supreme Court ruling in Thwaha Faisal – marked the beginnings of a consistent judicial pushback against state and prosecutorial impunity under UAPA. This was not to be.
This week, two other bail judgments – again, from the Bombay and Delhi high courts – reveal that UAPA adjudication continues to be inconsistent and judge-centric, and that individual liberty is, essentially, subject to the outcome of a judicial lottery. Not only is there inconsistency within the same court, but – as we shall see – inconsistency in the pronouncements of the same judge within the same court. Unfortunately, the costs of this inconsistency are measured in weeks, months, and years in jail.
Jyoti Jagtap and the Bombay high court
In Jyoti Jagtap vs NIA, a division bench of the Bombay high court denied bail to Jyoti Jagtap, a member of the Kabir Kala Manch (KKM) troupe. The case arose out of the violent events around the Elgar Parishad march on December 31, 2017. The prosecution’s case was that Jyoti Jagtap – and the KKM – had made various provocative speeches and performed provocative plays in the days leading up to the Elgar Parishad, and on the day of the event. The Prosecution argued that the KKM’s actions were tied to a larger conspiracy under the aegis of the banned CPI (Maoist), with the goal of overthrowing the State. The Prosecution also relied upon certain witness statements from 2011 (which, of course, had not been subjected to cross-examination at this stage of the proceedings) to the effect that Jyoti Jagtap had been seen in the forest, and in meetings with Naxalites.
In its analysis, the Bombay high court held, first, that the witness statements showed Jyoti Jagtap is an “active member” of the banned CPI (Maoist), by virtue of her presence in meetings with Naxalites, and alleged involvement in arms training; receipts and documents that showed her organisational role in the Elgar Parishad event; and “incitement of hatred and passion” on the day of the event, which showed that KKM was seeking to “overthrow” and “ridicule” the government. Evidence for this “incitement of hatred and passion” – according to the high court – included ridiculing the phrase “acche din”, referring to the prime minister as an “infant”, and “atrocities on Dalits in modern India.”
The High Court then referred to the National Investigative Agency’s account of the internal structure and functioning of the CPI (Maoist), as was recorded in the previous bail order in Hany Babu’s case, and noted that, for these reasons, Jyoti Jagtap’s acts would have to be assessed in the context of the “larger conspiracy” carried out by the CPI (Maoist). The court held:
The documents referred to herein above clearly highlights the active role of Appellant in so far as organising the Elgar Parishad event but more importantly it is the association of Appellant with the prominent members of CPI(M) which is a designated terrorist organisation which cannot be lost sight of. [Emphasis supplied]
Bail was accordingly denied.
Let us briefly recall the second of the two principles that I had set out above: “The allegations in the chargesheet must be individualised, factual, and particularistic. The gap between what an individual is accused of, and the actual events, cannot be filled by inferences or speculation.” The importance of this principle has already been discussed, and we are now in a position to see how the Bombay high court’s bail order evidently violates it. The actual “events” in question involve the violence that occurred in the aftermath of the Elgar Parishad event. Admittedly, Jyoti Jagtap was not accused of any violent act. What she had done – allegedly – was to play a role in the organisation of the event, and to perform on the day. But to link her to the violence – and deny her bail – the following inferences (none of which were demonstrably proven) had to be made:
- That the Elgar Parishad event was a CPI (Maoist) conspiracy, and that the violence was the design of the CPI (Maoist).
- That Jyoti Jagtap’s prior association with “prominent members” of the CPI (M) – dating back seven years and more – demonstrated her participation in the specific conspiracy set out in point (1). Note that “association” is a very broad term, and indeed, it was for this reason that the Supreme Court in Arup Bhuyan’s case had clarified that, for the purposes of the UAPA, membership meant “active membership”, i.e., incitement to violence; mere attendance at meetings, for example, wouldn’t do. The High Court did not engage with Arup Bhuyan’s judgment.
- That the speeches and performances on the day were in furtherance of this “conspiracy”, and caused the violent events that followed.
Indeed, in order to fill in the gaps in the Prosecution’s case through inferences, the Bombay high court resorted to increasingly strained logic: ridiculing the Prime Minister, his catchphrases, and his policies, statements on Shivaji and Tipu Sultan, and on atrocities against Dalits, were all construed as “inciting passion”, and in furtherance of the “larger conspiracy.” The high court was forced to do this, because the actual evidence against Jyoti Jagtap, as we have seen, was negligible; thus, the only way to establish the “prima facie” case against her and deny her bail under the UAPA was for the prosecution and the court to create a set of inferences that would connect her (innocuous) acts, such as playing an organisational role in the event and speaking at it, to the (actual) events – the violence – through the (unproven) vehicle of a “larger conspiracy.”
But it is in the very nature of an accusation of “conspiracy” that silences are damning: absences in the prosecution’s case can be explained away on the assumption that the accused is simply a good conspirator who covered their tracks. This is why judicial scrutiny – and the resistance to filling in the gaps through inferences – becomes particularly important. What is happening here can be illustratively compared to dropping a stone into still water. The stone causes a splash, and ripples emanate from the point of contact. The ripples grow in size but lessen in intensity, until the point at which the calmness of the water is restored.
If we think of the splash as the event in question, the legal standard articulated in the previous high court judgments was that you must at least have a visible ripple that you’re linking back to the splash (i.e., the link between the event and the accused actor cannot be tenuous beyond a certain degree).
What we have in Jyoti Jagtap’s case is someone coming along an hour after the splash, and saying – without demonstrating – that a particular movement on the surface of the water is somehow traceable back to the stone. This is not a sustainable judicial standard.
Umar Khalid and the Delhi high court
In March 2022, the sessions court denied bail to Umar Khalid, in a case arising out of the 2020 Delhi riots. I had written at the time that the court’s order was like being a “stenographer for the prosecution”: the prosecution’s statements were not checked even for internal consistency, gaps were explained away through inferences (this was especially stark, as Khalid was not even in Delhi when the riots took place), vague witness statements were interpreted to the detriment of Khalid, and any aspect of the case that benefited the defence was not considered on the basis that it was a “matter for trial” (the problems with this last approach have been discussed in some detail earlier).
The Delhi high court’s judgment upholding the order of the sessions judge, is – by and large – a replica of that order; interested readers, therefore, can look at the ‘Stenographer for the Prosecution‘ piece for more detailed analysis. Indeed, the court notes that it is in “full agreement with the findings of the Ld Sessions Judge” on the question of appreciation of evidence, and “does not wish to burden this judgment” through reiteration – thus exempting itself from the burden of independent analysis as well.
Be that as it may, there are a few parts of the judgment that deserve to be flagged, as they demonstrate something quite similar to the Bombay high court’s judgment in Jyoti Jagtap: that as the gap between the event and the accused grows larger, the court needs to resort to an increasingly stretched set of inferences in order to fill it.
These inferences begin in paragraph 52. The high court notes that, after the passage of the Citizenship Amendment Bill, (a) a WhatsApp group called the “Muslim Students of JNU” is formed, of which Umar Khalid is a member (emphasis mine); (b) the day after, the United Against Hate group conducts an agitation against the CAA, which Umar Khalid attends (emphasis mine), and – allegedly – supports a call for “chakka jam”; (c) another WhatsApp group called “CAB Team” is formed, of which – again – Umar Khalid is a member (emphasis mine).
The high court concludes that therefore, “a collective reading of the events that unfolded on each day after 04.12.2019 cannot be shrugged aside and it cannot be said that nothing incriminating has been.” (emphasis mine)
Thus, we can once again see the gap between fact (membership of WhatsApp groups and participation in an agitation) and event (riots), a gap that is large enough to drive a coach-and-four through, and this gap is effectively filled by stipulation: the high court says that all of this is incriminating, without an explanation for why it is so. In fact, not only does this gap not seem to bother the high court, but the court appears to positively revel in it. It notes:
It may be reminded that under the UAPA, it is not just the intent to threaten the unity and integrity but the likelihood to threaten the unity and integrity; not just the intent to strike terror but the likelihood to strike terror; not just the use of firearms but the use of any means of whatsoever nature, not just causing but likely to cause not just death but injuries to any person or persons or loss or damage or destruction of property, that constitutes a terrorist act, within the meaning of section 15 of UAPA. Moreover, under section 18 of UAPA, not merely conspiracy to commit a terrorist act but an attempt to commit or advocating the commission or advising it or inciting or directing or knowingly facilitating commission of a terrorist act that is also punishable. In fact, even acts preparatory to commission of terrorist acts are punishable under section 18 of UAPA. Thus, the objection of the appellant that a case is not made-out under UAPA is based on assessing the degree of sufficiency and credibility of evidence not the absence of its existence but the extent of its applicability; but such objection of the appellant is outside the scope and ambit of section 43D(5) of the UAPA. [Emphasis supplied]
Quite apart from the fact that the final sentence is borderline incoherent, the entire reason why the courts exist – and why judicial review exists – is precisely to introduce an element of the rule of law oversight over state action, so that words with boundless scope – such as “likelihood”, “likely to cause”, “of whatsoever nature”, “attempt”, “advocating”, and “preparatory” – are given a clear and precise meaning, and – most importantly – a limit (think of the stone-in-the-water image again).
In its reproduction of these words, however, the high court does the opposite: it uses their fuzzy nature and unclear boundaries as a justification for the inferences that it fills the gap with: in other words, imprecision begets imprecision: in the mind of the court, the vagueness of the legal language justifies the vagueness of the allegations.
Indeed, the entire judgment is shot through with similar vagueness. The high court says that the fact that Umar Khalid’s speech about Donald Trump was delivered despite permission for it being rejected gives “credibility” to the prosecution’s accusation that this same speech “heralded” the Delhi riots. This is a sine qua non to end all sine qua nons: there is no reference to the contents of the speech (for good reason), but the mere fact that the speech was given in defiance of the administration is used by the court as a basis for buttressing the prosecution’s accusation that it was meant to “herald” the riots.
The court also refers to a “flurry of phone calls” between the accused after the riots began; but the only circumstance in which a “flurry of phone calls” after a riot begins becomes suspicious is if you have already decided that the accused were conspirators, and their “flurry of phone calls” was in furtherance of their conspiracy. In all other circumstances, a “flurry of phone calls” between activists in the immediate aftermath of a riot is one of the most natural things imaginable; in fact, it would be silence that would be infinitely more suspicious.
Having probably realised that its inferential chain stretches credulity in the extreme, the high court is now forced to look at the actual evidence on record, and make something out of where nothing exists. Thus, the high court seizes upon Umar Khalid’s words – “inquilabi salam” (revolutionary greetings) and “krantikari istiqbal” (revolutionary welcome) – and finds incriminating material there. In a highly confusing passage that cites Robespierre and Nehru, the court says:
“Revolution by itself isn’t always bloodless, which is why it is contradistinctly used with the prefix – a ‘bloodless’ revolution. So, when we use the expression “revolution‟, it is not necessarily bloodless.”
What? Is the court trying to say that henceforth, whenever anyone uses the word “revolution”, they have to preface it with the word “bloodless”, otherwise they will be caught and thrown in jail under the UAPA, and denied bail? The very absurdity of this paragraph should make clear that the high court is clutching at straws here: what it has before it are a set of facts (Umar Khalid’s conduct) that has no conceivable relationship with what actually happened (the Delhi riots); and to establish the relationship, not only does the high court have to construct an elaborate labyrinth of assumptions and inferences, but also put some of our most common political turns of phrase into a torture rack, until they crack and confess to any meaning the court wishes them to confess to.
Concluding its analysis, the court notes the existence of a “pre-meditated conspiracy” to do a chakka jam and incitement to violence, a “pre-planned” attack, Umar Khalid’s membership of WhatsApp groups and speeches, the “flurry of calls”, and Umar Khalid’s active involvement in the protests against the CAA. As with the Jyoti Jagtap bail order, let us identify the assumptions at work, that the court needs to fill in the gaping holes in the prosecution’s case:
- That calling for a chakka jam logically entails incitement to violence and riots.
- That membership of WhatsApp groups is indicative of participation in a conspiracy.
- That a “flurry of calls” after a riot has started – between activists who have been engaged with the issue in question – is indicative of a conspiracy.
- That Umar Khalid’s involvement in protests against the CAA is indicative of his participation in a conspiracy to cause riots.
The High Court needs to do all this because – as the evidence showed very clearly – at no point did Umar Khalid call for violence, publicly incite violence, commit violence, or participate in riots. Thus, all we have is membership of WhatsApp groups indicating a vague “association”, a “flurry of calls” that also indicates an “association”, and involvement in protests; the high court weaves this together into a boundlessly flexible mesh called “larger conspiracy”, and tops it off with a dash of Robespierre and Nehru. Thus, an individual who has now been in jail for two years without a trial is condemned to a further – uncertain – spell in prison, while we still wait for the trial to begin.
Post-script: The contradictions of Justice Siddharth Mridul
Before concluding, it is important to flag a final point. The two-judge bench of the Delhi high court that denied bail to Umar Khalid had, as its members, Justices Bhatnagar and Siddharth Mridul. The June 2021 bail judgment in Asif Aqbal Tanha’s case had, as its members, Justices Bhambani and Siddharth Mridul.
Under even the loosest standards of intellectual consistency, it is simply inexplicable how the same learned justice can – without further explanation – be party to two bail judgments that not only arose out of the same set of facts, but took polar opposite approaches to the issue. Let us begin with the most glaring and blatant contradiction: in paragraphs 49-58 in Asif Iqbal Tanha’s judgment, the bench – of which Justice Siddharth Mridul was a member – gives a narrow interpretation to the word “terrorism” under the UAPA, and notes that the prosecution’s argument that the anti-CAA protest was designed to threaten the foundations of the nation is based on “inferences”, and that therefore, there is prima facie no case made out of terrorism, or conspiracy, or the commission of acts “preparatory” to terrorism.
In paragraphs 62-67 of Umar Khalid’s judgment, the bench – of which Justice Siddharth Mridul is also a member – explicitly notes that the anti-CAA protest was not an “ordinary protest”, and then goes on to argue how the protests and the riots were linked, and how it all constitutes a terrorist act.
How can the same judge simultaneously appear to believe both X and not-X?
And if the same judge has changed his mind, and now believes not-X where at one point he so firmly believed X that he signed a judgment to that effect, are we not owed the courtesy of an explanation? Walt Whitman might claim the luxury of contradicting himself because he contains multitudes, but it is not open for a high court judge to contain contradictory multitudes, at least not while he sits in his official capacity.
Similarly, in paragraph 35 of the Natasha Narwal order (which accompanied the Asif Iqbal Tanha order), the high court bench – of which Justice Siddharth Mridul was a member – noted that “we can discern no specific or particularised allegation, much less any material to bear out the allegation, that the appellant incited violence, what to talk of committing a terrorist act or a conspiracy or act preparatory to the commission of a terrorist act.”
In the Umar Khalid judgment, as we have seen, the bench – of which Justice Siddharth Mridul is also a member – dispenses with the need for any “specific or particularised” allegation at all, going – instead – by membership of WhatsApp groups and “inquilabi salam” to justify the denial of bail. Once again, the two approaches to the issue are simply irreconcilable.
It is one thing to have a polyvocal court that speaks different tongues (although that is problematic enough). But it is quite unheard of to have a polyvocal judge. We can, perhaps, only shake our heads at this marvel, and agree with Hamlet when he wisely said, “There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy.”
The judgments in Jyoti Jagtap and Umar Khalid show that the courts continue to be sites of contestation when it comes to UAPA, state and prosecutorial impunity, and under-trial incarceration. These two judgments belong to the “executive court” tradition, where the language of the court resembles – and often goes beyond – the language of the executive. In UAPA bail cases, the executive court’s judgments are marked by how judicial reasoning fills in the gaps in the prosecution’s case with inferences and assumptions, and innocuous and political legitimate forms of dissent are rendered illegal by transplanting them into a “larger conspiracy”, and how the issue of the conspiracy itself remains an assumption.
As we have seen, however, this is not the only way under the UAPA: the 2021 and 2022 bail judgments – that also come from the Bombay and Delhi high courts – show how a judiciary that is sensitive to the claims of individual liberty can act under the confines of the UAPA.
Much, therefore, will depend upon which of these two approaches, over time, finally transforms into “settled law”: in the meantime, each individual case represents an important site of the legal and constitutional struggle against the UAPA’s entrenchment of state impunity.