Guest Post: Reflections on the High Court's Judgment in Fahad Shah's Case

(This is a guest post by Rajyavardhan Singh)

A recent post on this blog by the editor argued that the Supreme Court in Union Territory of J&K v. Peerzada Shah Fahad had erred in declaring the J&K and Ladakh High Court’s judgement (“Fahad Shah”) per incuriam merely for referencing the "clear and present danger" test from Schenck v. United States (“Schenck”) in the context of bail under Unlawful Activities Prevention Act, 1967 (“UAPA”). It argued that the High Court in Fahad Shah did not misuse Schenck to assess free speech so as to render it per incuriam, but instead adapted it to set a threshold for justifying arrests and detention under UAPA. Further, it criticised the Supreme Court’s dismissal as potentially hindering other courts from considering valid reasoning.

Reading the analysis led me down a fascinating rabbit hole, exploring a rather distinct yet related question: Whether the invocation of the "clear and present danger" doctrine was normatively sound in this context?

Upon careful perusal, I submit that the High Court's reasoning suffers from two interconnected flaws that extend beyond the mere citation of foreign precedent. First, it commits a false equivalence by conflating a doctrine specifically designed for post-facto speech restrictions with the preventive nature of the UAPA, thereby misaligning the temporal and causal requirements. Second, and more problematically, it attempts a flawed transposition of American free speech doctrine into an Indian anti-terror statute, disregarding not only the Supreme Court’s deliberate rejection of such transplants but also the carefully calibrated evolution of domestic anti-terrorism jurisprudence.

1. False Equivalence

The High Court's invocation of the "clear and present danger" test in interpreting Section 43D(5) of the UAPA marks a significant departure from the legislative intent of the statute. While Sekhri contends that the court merely borrowed evocative prose from Schenck to express a proximity requirement for arrests, this interpretation understates the conceptual confusion such borrowing creates. Although he is right in observing that the High Court was "crafting an idea to address a problem using remnants of old concepts from different contexts," the Court's seemingly harmless transplantation nonetheless obscures a fundamental category error.

For what must be understood is that the "clear and present danger" test, even when repurposed, carries with it specific analytical assumptions about causation and imminence that are fundamentally at odds with the preventive nature of the UAPA framework. As evidenced in paragraph 31 and 32 of the High Court's judgement:

"31. The legislative intent behind s. 43D (5) and its proviso was to ensure that those who were a “clear and present danger” to the society, whose relationship with the offence is proximate and direct, do not get bail during the pendency of the trial lest they take to their nefarious ways again, once released. It was not to keep incarcerated the unwary transgressor who found himself at the wrong place at the wrong time.”

32. "We hold, that the investigating agency, investigating a case under the UAPA, has the unbridled authority to arrest or not to arrest under the provisions of the UAPA. However, upon arrest, the investigating agency would have to justify the arrest on the anvil of 'clear and present danger' of the accused to the society at large, if enlarged on bail. The existence of prima facie evidence against the accused is to no avail if there is no justification for the arrest based on the doctrine of clear and present danger to the society." [Emphasis mine]

This borrowed standard implicitly requires a degree of temporal and causal proximity that the UAPA deliberately eschews. When the High Court states that authorities must "justify the arrest on the anvil of 'clear and present danger' of the accused to society at large" it imposes a reactive paradigm on a preventive statute. This, I submit, is not merely a semantic sleight-of-hand—but much rather represents a substantive reconfiguration of the UAPA's operational logic. The specific importation of Schenck's framework introduces evidentiary and temporal requirements that could effectively nullify the UAPA's preventive capacity, particularly in cases involving long-term planning or preparatory acts.

This is precisely why the UAPA deliberately moves away from such immediate temporal requirements, allowing intervention based on reasonable grounds even when threats aren't imminent.

2. Flawed Transposition

The High Court's use of Schenck's language to express established principles about justified arrests overlooks the dangers of decontextualised legal borrowing. While Sekhri quite rightly notes that "there is no complete ban on citing a case that may have been disagreed with or frowned upon in the past" so as to render it per incuriam, nevertheless I argue that the manner of citation matters. Otherwise, it risks oversimplifying the nuanced jurisprudential reasons why Indian courts have rejected the "clear and present danger" test in cases like Babulal Parate v. State of Maharashtra, State of Madras v, V.G. Row and more recently in Arup Bhuyan v. State of Assam.

Importantly, the High Court's judgement does more than merely reference Schenck; it attempts to transpose a specific legal test designed for post-facto speech restrictions into the realm of anti-terror legislation. This, in turn, creates two fundamental problems. First, it muddles the distinct jurisprudential frameworks governing speech restrictions and terrorism. Second, it disregards the Supreme Court's careful development of native doctrine around preventive detention, which has evolved to balance security imperatives with constitutional rights in India's specific context (As Shastri J. emphasised in V.G. Row, India's constitutional scheme demands "different techniques and approaches" that account for our unique security challenges).

The flawed transposition becomes even more apparent when we trace how the borrowed test would operate in practice. Under Schenck, the "clear and present danger" framework was designed to evaluate whether speech had sufficient proximity to harm to justify restriction. This required courts to analyse: (1) the immediacy of potential harm, (2) the direct causal link between speech and danger, and (3) the substantive likelihood of the feared outcome.

When the High Court attempts to transpose this framework to UAPA arrests, it reveals both the appeal and limitations of such doctrinal borrowing:

First, while the "clear and present danger" test might seem useful for cases like Mr. Shah's, where a significant temporal gap exists between alleged acts and arrest. It creates an impossible ‘temporal paradox’—requiring authorities to show a "present" danger for actions specifically aimed at preventing dangers from becoming "present." This framework might work when questioning decade-old publications but becomes incoherent when applied to UAPA cases more broadly. While the High Court's instinct to question the necessity of arrest for historical acts is sound, importing Schenck's test unnecessarily complicates what could be a straightforward proportionality analysis.

Second, even in cases involving journalistic activities, the causal chain analysis from Schenck proves problematic. The High Court's attempt to require a direct link between past publications and present security threats, while protective of press freedom, fails to account for how written content might contribute to security risks over time through gradual radicalization or by providing operational intelligence to hostile actors. The "clear and present danger" framework's insistence on direct causation could blind courts to these more subtle, long-term security concerns while offering little additional protection against arbitrary detention that existing proportionality standards don't already provide.

Thus, while the High Court's judgement admirably attempts to protect against arbitrary detention for dated allegations, its choice of doctrinal tool creates more problems than it solves. Rather than importing foreign standards without fully contextualising them—especially from reactive contexts to preventive ones—a more effective approach would involve creating guidelines specifically tailored to address temporal remoteness under UAPA. Such guidelines could consider factors like the ongoing relevance of old publications, shifts in context since their release, and the proportionality of current detention. This approach would better serve the Court’s objectives while preserving doctrinal coherence.

To Conclude

Section 43D(5) of the UAPA, while imposing strict conditions for bail, does not override the fundamental requirement that arrests must be necessary and proportionate. In Fahad Shah, the High Court could very well have examined whether the significant temporal gap between the alleged journalistic activities and the arrest met these essential standards (given the decade-long gap between Mr. Shah’s alleged acts and his arrest) without resorting to the analytically incompatible 'clear and present danger' test.

In Union of India v. K.A. Najeeb, the Supreme Court has already established that statutory restrictions like Section 43D(5) cannot displace constitutional safeguards enshrined under Part III in evaluating preventive detention:

18. “It is thus clear to us that the presence of statutory restrictions like Section 43­D (5) of UAPA per­ se does not oust the ability of Constitutional Courts to grant bail on grounds of violation of Part III of the Constitution. Indeed, both the restrictions under a Statue as well as the powers exercisable under Constitutional Jurisdiction can be well harmonised. Whereas at commencement of proceedings, Courts are expected to appreciate the legislative policy against grant of bail but the rigours of such provisions will melt down where there is no likelihood of trial being completed within a reasonable time and the period of incarceration already undergone has exceeded a substantial part of the prescribed sentence. Such an approach would safeguard against the possibility of provisions like Section 43­D (5) of UAPA being used as the sole metric for denial of bail or for wholesale breach of constitutional right to speedy trial.”

This approach would have allowed the court to conclude that an inordinate delay between the alleged acts and the arrest, without compelling justification, suggests either a diminished need for preventive detention or an abuse of process—both warranting bail. Such reasoning would not only have led to the same conclusion but also promoted a more principled development of bail jurisprudence under special legislations like the UAPA, ensuring doctrinal consistency.

The SC Order in the Fahad Shah Case

In November 2023, the High Court of J&K and Ladakh passed its judgment in Peerzada Shah Fahad v. Union Territory of J&K. Readers may recall that this judgment concerned a journalist, Mr. Shah, who had been prosecuted for offences under the Unlawful Activities Prevention Act 1967 ['UAPA']. In connection with which he had also been arrested and denied bail.  

The judgment of the High Court was the result of Mr. Shah challenging two orders, one by which charges had been framed against him for offences under the UAPA and the other by which his bail application had been rejected by the lower court. The High Court granted him bail, and also ruled that some of the charges against him, including the most serious UAPA charge, were unsustainable. [The judgment was covered on the Blog here and was also discussed in the press here, here and here].

This judgment was challenged by the Union Territory of J&K and Ladakh before the Supreme Court. The Union, presumably, sought that the bail so granted be denied and that the charges originally framed against Mr. Shah be restored. It appears that, on the first date of hearing in September 2024, the Union took an adjournment. After which the case came up for hearing on 14.10.2024, where the petition was disposed off. The two orders suggest that the Court did not issue any notice to the opposite side, i.e. Mr. Shah, before deciding the petition. 

That a petition was decided without hearing the proposed respondent is not uncommon as many petitions to the Supreme Court are dismissed at the outset as the Court considers no interference is being called for with the lower court's orders. In this case, though, the Court did not simply dismiss the petition. While it held that there was no reason to deny bail and return Mr. Shah to custody, the Supreme Court made a key observation in respect of the validity of the reasoned 25 page judgment of the High Court: It declared that the judgment was per incuriam and shall not be cited as precedent (para 6). In other words, no other court can henceforth rely upon the observations made by the J&K and Ladakh High Court, nor can counsel try to use them in support of their case.

Why did the Supreme Court arrive at such a conclusion? Paragraph 6 of the brief two page order passed on 14.10.2024 states that the per incuriam finding was because the High Court's judgment were contrary to judgments passed by the Constitution Benches of the Supreme Court. To understand which judgments, we must turn to Paragraph 4 of the same order, which extracts submissions made by the Solicitor General. To avoid any confusion I have copied the relevant part below:

"The learned S.G., Mr. Mehta has drawn our attention to the para 28 of the impugned judgment, whereby the High Court has placed reliance on Schenck Vs. United States (249 US 47 (1919) : 1919 SCC Online US SC 62). He submitted that the two Constitution Benches in Babulal Parate vs. State of Maharashtra (AIR 1961 SC 884) and in State of Madras vs. V. G. Row [(1952) 1 SCC 410] and one Three Judge Bench in Arup Bhuyan vs. State of Assam [(2023) 8 SCC 745] have rejected the application of the doctrine of “clear and present danger”, as laid down in the said judgment, i.e., in Schenck Vs. United States (supra), and therefore, the observations made in the impugned judgment would be per incuriam. ..." 

There is no doubt that the Indian Supreme Court does not fully approve of the test in Schenck and that if a High Court relies upon Schenck without understanding its clarifications by the Indian Supreme Court, it is proceeding in error. But underlying all of this analysis is a rather obvious premise: the High Court ought to have been using Schenck for the reasons why the Indian Supreme Court did not approve it, for it to have ever fallen into error. But if it was not using Schecnk for these purposes, there was no problem, because there is no complete ban on citing a case that may have been disagreed with or frowned upon in the past.

Let me explain this further. A judicial decision gives us an answer to a problem, but rarely does it give this answer in a straightforward or formulaic manner. Judges give their answer in ways that can help in future cases, and do so with their unique sense of flourish (or dose of dour). Schenck is one such judgment, in which the U.S. Supreme Court not only decided the problem before it — the validity of the Espionage Act — but did so by providing a test for future courts to decide issues of the legality of similar statutes. Now it is obviously open for future courts to disagree with this old test and come up with their own tests, which is what happens with India's treatment of Schenck. The result is that the test laid down in Schenck cannot be used to decide the kind of problems that it was intended to solve. It does not mean that no court can use the decision to aid its imaginative process of solving other problems.           

To decide, then, whether or not the High Court in Mr. Shah's case had fallen into error in citing Schenck, it is first necessary to determine just what was the problem that it was dealing with. Was the petition testing the legality of any statute on grounds of interference with freedom of speech and expression? It was not. It was deciding a challenge to arrest and detention by a person who had been arrested in 2022 on the basis of accusations against him pertaining to acts done in 2011. In other words, was an arrest in 2022 justified for alleged illegal acts committed in 2011? 

A reading of the High Court's judgment from November 2023 makes it clear that there are only three places in which Schenck, or the idea of 'clear and present danger' which the Supreme Court order cites, are referred to. The first of these is paragraph 28 of the High Court's judgment (which is also mentioned in the Supreme Court's order): 

The charge against the Appellant is basically associated with his right to freedom of speech and expression, which prima facie appears to have gone wrong. In Schenck Vs. United States, the charge against the Appellant was under the provisions of the Espionage Act that impeded the US war efforts against the German Empire in world war I, by mailing letters to discourage conscripts who had qualified to give their service as soldiers against Germany, from joining the US Army. The Appellant sought the quashing of the Espionage Act as it infringed his first amendment right of free speech. Justice Oliver Wendel Holmes writing for the court held “The question in every case is whether the words used are used in such circumstances and are of such a nature as to create a clear and present danger that they will bring about the substantive evils that congress has a right to prevent. It is a question of proximity and degree”.

The second and third references are not to Schenck, but only to the idea of a 'clear and present danger' at paragraphs 31 and 32 of the judgment — and neither of these have been referred to by the Supreme Court. What do these paragraphs show us? That the High Court did not use 'clear and present danger' as a test to decide the validity of interference with free speech and expression, but as a kind of placeholder to create a new threshold to determine circumstances where arrest and detention would be justified under the UAPA. 

Paragraph 31: " The legislative intent behind s. 43D (5) and its proviso was to ensure that those who were a “clear and present danger” to the society, whose relationship with the offence is proximate and direct, do not get bail during the pendency of the trial lest they take to their nefarious ways again, once released. It was not to keep incarcerated the unwary transgressor who found himself at the wrong place at the wrong time. ...". [Emphasis mine]

Paragraph 32: "We hold, that the investigating agency, investigating a case under the UAPA, has the unbridled authority to arrest or not to arrest under the provisions of the UAPA. However, upon arrest, the investigating agency would have to justify the arrest on the anvil of “clear and present danger” of the accused to the society at large, if enlarged on bail. The existence of prima facie evidence against the accused is to no avail if there is no justification for the arrest based on the doctrine of clear and present danger to the society. If the investigating agency does not satisfy this Court and is unable to justify the arrest (as warranted in Joginder Kumar) the same would result in the violation of the rights of the accused under part III of the Constitution as adumbrated in K.A Najeeb's case, and the accused may be enlarged on bail. In order to assess whether the accused is a clear and present danger, there can be no rule of thumb and it must be seen in the backdrop of the specific facts and circumstances of each case." [Emphasis mine] 

What we have here, is a court crafting an idea for solving a problem with remnants of old concepts from different contexts. And in crafting this new idea the J&K and Ladakh High Court did not simply cite from Schenck, but it turned to a concept which the Supreme Court of India has wholeheartedly endorsed time and against: that a person ought not to be arrested merely because he can, but there must be grounds for an arrest. 

Moreover, it would be a disservice to say that this is an entirely new idea to begin with. Within the Indian context, there were already decisions, by various courts, which stressed upon the need for some proximity between the allegations and the exercise of powers of arrest. All that the J&K and Ladakh High Court did was to express this rationale in different, more evocative, prose. That it did so by bringing to use the prose from a decision which had been disagreed with in a different context, could not be a reason to deprive all other courts from even considering the High Court's reasoning. 

If the the only reason to hold that the judgment of the J&K and Ladakh High Court in Mr. Shah's case is per incuriam is because it decided to adopt the same catchy phrase from a decision which earlier benches of the Indian Supreme Court had disagreed with for different reasons—and I say 'if' because the petition itself may have raised many other grounds, none of which are captured in the order unfortunately—then it appears that the Supreme Court order dated 14.10.2024 has committed a most serious error. 

Unpacking the Judgment in Fahad Shah v. Union

A Division Bench of the Jammu and Kashmir High Court delivered its judgment in Fahad Shah v. UT of J&K & Anr. on 17.11.2022 — a common judgment in petitions assailing orders passed by the trial court that had framed charge and denied bail to Mr. Shah. The ultimate result of the judgment was to partially set aside the order on charge, taking out the serious offences punishable under the Unlawful Activities (Prevention) Act 1967 ['UAPA'] and the Indian Penal Code 1860 ['IPC'] while retaining some less serious offences under the UAPA and the Foreign Contribution (Regulation) Act ['FCRA']. And, significantly, to grant him bail and sanction his release from prison where he had spent more than six hundred days, being detained in multiple cases and pursuant to preventive detention orders.

Indian law on bail generally, and especially in statutes such as the UAPA, makes it far easier for courts to deny bail in non-bailable cases. The Criminal Procedure Code 1973 ['Cr.P.C.'] does not demand reasoned orders for rejecting bail, only for granting it; similarly, the UAPA encourages rejecting bail if accusations are simply 'prima facie true'. At some level, this makes any bail order an act of courage. The judgment in Fahad Shah is mightily courageous, and has garnered deserved praise. This post is not another footnote in that celebratory procession, though, and wishes to engage with the judgment from the practitioner's lens. It only looks at procedural issues, and does not ponder upon the implications of the government drawing the contours of terrorism so widely as to bring within it alleged defamation of national interests. 

The first point of note is the procedural posture of the case — this was a case where charges had already been framed. To conclude that charge ought to be framed is, essentially, a finding by court that the police materials make out a 'prima facie' case against the accused. Determining whether the case is 'prima facie true' is precisely the threshold identified by Section 43-D(5), UAPA for courts to reject bail. So is there no chance of being granted bail where charge has been framed? 

Since the High Court in Fahad Shah ruled that there was no material to frame charge for the offences that triggered the restriction under Section 43-D(5), it did not need to answer this issue as such. It nevertheless expounded on a related problem — if accusations are 'prima facie true' does that act as an insurmountable hurdle for release? The conclusion it arrived it was 'no', and it located the reasoning for its conclusion in Article 21 of the Constitution and the view in Supreme Court decisions such as K.A. Najeeb (2021) which held that the direction of constitutional courts to grant bail on constitutional grounds, where such grounds are made out, cannot be fettered by the text of the UAPA. Using this prong, it latched on to another line of prior cases which had held in a general context that while accusations may be made out in many cases and confer a power of arrest, the mere existence of this power did not automatically validate its exercise in all cases. Similarly, when it comes to the UAPA, even where the accusations may be 'prima facie true', that did not validate the fact of arrest itself which the court had to separately inquire into in every case. If, as it held in Fahad Shah, the arrest itself was not made out, then there was no question of Section 43-D(5) of the UAPA acting as a barrier to release.

At its highest, the judgment in Fahad Shah has reinforced what ordinary law already demanded of courts — a thorough scrutiny of the validity of arrest prior to authorising any further remand to custody. Strong reminders of the judicial obligation to seriously scrutinise the police version and examine the need for an arrest came from the Bombay High Court in Chanda Kochhar (an IPC case) and more recently by the Supreme Court in Pankaj Bansal (a PMLA case). Fahad Shah reiterates the logic in a UAPA context. The judgments show how courts can read more bite into the existing statutory yardsticks for effecting arrests if they try. That Mr. Shah remained in custody for almost two years in a case where the High Court finally held that he needn't have been arrested sums up how hard courts end up trying.

Where Fahad Shah does not develop the law is its choice to ground its reasoning in the Constitution and not the text. While reaffirming law that allows constitutional courts to interfere with orders and grant bail, it did not clarify about the extent to which trial courts can proceed on the same route. One might argue that this is a bogus point — the Constitution reigns supreme, and all courts must enforce its mandate. But the specific reference to constitutional courts has allowed prosecutors to raise an objection and for trial courts to pass the buck. Had the Court tethered its reasoning to the statutes more than the constitution, it would have helped diffuse such claims from being raised in the everyday application of the laws. 

Fahad Shah reminds us that the only avenues for bail in such cases appear to be moving (constitutional) courts by invoking Article 21, or by challenging the validity of the accusations itself. Thus, what the High Court does not do, is develop the law on bails in UAPA cases where charge has been framed, where the court does not conclude that the charges or bogus or arrest was improper. To repeat, the issue is an obvious one: when a court frames charge, it delivers a finding accepting that accusations are prima facie true, which is the threshold identified for rejecting bail under Section 43-D(5) of the UAPA. In ordinary non-bailable cases, the absence of a Section 43-D(5) means that even if charges are framed bail can still be granted if the circumstances so warrant, since there is no peremptory bar operating. For instance, where the case has dragged on for years and there is no threat to evidence being recorded. This is seemingly not available in the UAPA which can cause serious prejudice to those persons unable to exercise remedies by going to a High Court. 

To explain the issue through the logic that is endorsed by the High Court — while it held in Para 32 that the issue of arrest ought to be scrutinised on the anvil of a 'clear and present danger' test, it did not go so far as to hold that the continued detention of the accused also ought to be tested on that anvil, and that is a problem.