Landmark Cases — Watali, KA Najeeb, Union of India v Yasin Malik

No trio of decisions explains the modern law of bail under the Unlawful Activities (Prevention) Act, 1967 better than NIA v. Zahoor Ahmad Shah Watali, Union of India v. K.A. Najeeb and the long-running prosecution of Yasin Malik. The first built the statutory wall of Section 43-D(5); the second showed where Article 21 lets a court climb over it; the third tests how all of this plays out for a high-profile terror-funding convict still fighting a death-penalty appeal. Read together, they are the spine of every UAPA bail argument an aspirant must master, and they recur constantly in judiciary and CLAT-PG papers. This note grounds each holding in its verified citation, isolates the ratio, and traces how the Supreme Court has since narrowed and re-balanced the doctrine.

The statutory backdrop: Section 43-D(5) and its proviso

The three cases cannot be understood without the text they revolve around. Section 43-D(5) of the UAPA opens with a non-obstante clause — “Notwithstanding anything contained in the Code” — and provides that no person accused of an offence under Chapters IV and VI of the Act shall be released on bail without the Public Prosecutor being heard. The crucial proviso then states that such an accused “shall not be released on bail… if the Court, on a perusal of the case diary or the report made under section 173 of the Code is of the opinion that there are reasonable grounds for believing that the accusation against such person is prima facie true.”

Two features make this an extraordinary bail bar. First, it inverts the ordinary presumption: instead of liberty being the default, the court must be affirmatively satisfied that the accusation is not prima facie true before bail can follow. Second, the inquiry is confined to the case diary and the chargesheet — the prosecution's own material. The interpretive battle in the landmark cases is exactly how deep a court may dig into that material, and whether the Constitution can ever displace the bar altogether. For the constitutional pedigree of these powers, see our note on the introduction, object and constitutional background of the Act, and for the underlying offence categories, the offences and penalties for terrorist acts.

Watali: the facts and the High Court's error

National Investigation Agency v. Zahoor Ahmad Shah Watali, (2019) 5 SCC 1, decided on 2 April 2019, arose from the J&K terror-funding investigation. Watali, a Srinagar businessman arraigned as Accused No. 10, was alleged to have acted as a conduit channelling funds from the Lashkar-e-Taiba founder Hafiz Muhammad Saeed and other Pakistan-based sources to Hurriyat leaders and secessionists to wage organised violence in the Kashmir Valley. The Special Court refused bail; the Delhi High Court reversed and granted it, principally on the view that much of the NIA's material — statements, documents and disclosure memos — would be inadmissible at trial and therefore could not sustain the accusation.

The Supreme Court (Justices A.M. Khanwilkar and Ajay Rastogi) held the High Court had committed a fundamental error: at the bail stage under Section 43-D(5), a court is not to weigh admissibility or conduct a mini-trial. The Bench restored the rejection of bail, holding the NIA's material disclosed reasonable grounds to believe the accusation against Watali was prima facie true. The High Court had, in the Supreme Court's view, effectively pre-judged the trial by discarding statements and documents on admissibility grounds — a function reserved for the trial court after evidence is led, not for a bail court reading the chargesheet.

Watali: the ratio that governs every UAPA bail

The enduring ratio of Watali is its definition of the prima-facie standard. The Court held that the expression “prima facie true” means that the material on record, taken at face value and accepted as it is, must show the complicity of the accused; the court must not analyse the evidence as if at trial, nor weigh probative value, nor consider whether the material is admissible or could be controverted in cross-examination. The totality of the case-diary and chargesheet material is to be read on a surface level, and if it discloses reasonable grounds, bail must be refused.

The Court also clarified that the degree of satisfaction is lighter than the standard for framing charge or for the prosecution to prove guilt — it is enough that, on a broad probability, the accusation is prima facie true. This made Watali a powerful prosecution tool: by foreclosing any examination of the quality of evidence at the bail stage, it effectively converted the bail hearing into an acceptance of the chargesheet's narrative. It is the single most-cited authority for denying UAPA bail and is the precedent every defence argument must distinguish or escape.

Three subsidiary propositions from Watali deserve close attention because they recur in answer scripts. First, the Court held that the word “believing” in the proviso signals a degree of satisfaction higher than mere suspicion but lower than proof; the court reaches an opinion, not a finding. Second, the material relied upon need not be admissible evidence in the strict sense — intercepts, disclosure statements and documents recovered during investigation may all be looked at on a face-value basis, even though their admissibility will be tested only at trial. Third, the Court cautioned that the bail order must be reasoned but must avoid detailed comments on the merits, lest it prejudice the trial. The cumulative effect was to tilt the bail calculus decisively towards the prosecution wherever a chargesheet narrates a coherent accusation. Critics argue this collapses the distinction between an accusation and proof; defenders say it reflects Parliament's deliberate choice in a security statute. Both framings are examinable.

Najeeb: prolonged incarceration without trial

If Watali built the wall, Union of India v. K.A. Najeeb, (2021) 3 SCC 713, decided on 1 February 2021 by Justices N.V. Ramana, Surya Kant and Aniruddha Bose, showed where a court may go over it. Najeeb was accused in the 2010 attack in which the palm of Professor T.J. Joseph was chopped off in Kerala over an allegedly blasphemous examination question, attributed to a banned outfit. He was arrested by the NIA in 2015 and charged under, among others, Chapter IV and VI offences of the UAPA. By the time the matter reached the Supreme Court he had spent over five years as an undertrial, with 276 prosecution witnesses still to be examined and no realistic prospect of the trial concluding soon. The Kerala High Court granted bail; the NIA appealed.

Najeeb: when the rigours of Section 43-D(5) 'melt away'

The Supreme Court dismissed the NIA's appeal and upheld bail, delivering the most important counterweight to Watali. The Court held that the embargo in Section 43-D(5) is a statutory restriction on the power of an ordinary court, but it does not oust the constitutional jurisdiction of the constitutional courts to grant bail where there has been a violation of the fundamental right to a speedy trial under Article 21. Where an undertrial has suffered prolonged incarceration with no likelihood of the trial concluding in reasonable time, the rigours of Section 43-D(5) “would melt down” and the constitutional court may grant bail notwithstanding the statutory bar.

Crucially, the Court reconciled the two regimes: the statutory restriction and the constitutional safeguard operate in different planes, and the latter prevails when continued detention itself becomes unconstitutional. The Bench drew on Shaheen Welfare Association v. Union of India, (1996) 2 SCC 616, and Hussain v. Union of India, (2017) 5 SCC 702, to anchor the speedy-trial principle. Najeeb thus did not dilute the prima-facie test of Watali at all; it carved out a parallel route resting wholly on delay and Article 21.

The reasoning rests on a careful jurisdictional point. The Court accepted that an ordinary criminal court, deciding bail under the Code as modified by Section 43-D(5), is bound by the statutory embargo and cannot grant bail merely because the trial is slow. But a constitutional court — the High Court under Article 226 and the Supreme Court under Article 32 or 136 — exercises a power that the legislature cannot curtail, namely the duty to enforce fundamental rights. When delay in trial defeats the Article 21 guarantee, refusing bail would itself perpetuate a constitutional violation, and no statute can compel a court to do so. The Court was also careful to fence the principle: it does not convert every delayed UAPA trial into an automatic bail; the constitutional court must weigh the gravity of the accusation, the period already undergone, and the realistic timeline for conclusion. Najeeb had spent over five years in custody with hundreds of witnesses pending — an extreme case — and that quantum of delay was decisive. The ratio is therefore narrow in form but immensely influential in practice, because protracted UAPA trials are common.

Watali and Najeeb reconciled: two doors, not one

Aspirants frequently mis-state these cases as being in conflict. They are not. Watali governs the merits door: can the court, reading the chargesheet at face value, say the accusation is not prima facie true? If the material crosses the threshold, this door is shut. Najeeb opens a separate delay door: even where the merits door is shut, a constitutional court may grant bail if prolonged detention without timely trial has breached Article 21. A defence lawyer who cannot win on Watali grounds therefore pivots to Najeeb grounds — length of incarceration, number of witnesses remaining, and the improbability of an early conclusion.

This two-door framework is now standard. It is why courts in cases such as the Delhi riots and Bhima Koregaon matters spend pages first on the prima-facie material and then, separately, on the period already spent in custody. Understanding which door an argument belongs to is the single most testable insight from this pairing.

Vernon Gonsalves: narrowing Watali from within

The merits door was not left entirely rigid. In Vernon v. State of Maharashtra, decided on 28 July 2023 by Justices Aniruddha Bose and Sudhanshu Dhulia, the Supreme Court granted bail to Bhima Koregaon–Elgar Parishad accused Vernon Gonsalves and Arun Ferreira. The Court read Watali carefully and held that the prima-facie test does not require a court to accept material mechanically: while it cannot weigh evidence as at trial, it must undertake “at least a surface-analysis of the probative value” of the material relied on. Documents of low or doubtful probative worth — here, literature and electronic material whose connection to any terrorist act was tenuous — cannot by themselves satisfy the prima-facie threshold.

This is a significant refinement: Vernon did not overrule Watali (a coordinate-bench decision) but interpreted its surface-level reading to include an assessment of probative value, restoring some judicial scrutiny at the merits door. It is increasingly invoked alongside Najeeb to attack weak chargesheets.

The Court reasoned that the Watali direction not to weigh evidence cannot mean that a court must accept any document, however irrelevant, as establishing the accusation; otherwise the proviso would be satisfied by the mere act of filing a chargesheet, reducing the judicial inquiry to a rubber stamp. The phrase “prima facie true” must carry some content, and that content is a threshold assessment of whether the material, if believed, has the probative capacity to connect the accused to a Chapter IV or VI offence. On the facts, the alleged recoveries — books, pamphlets and electronic content — even taken at face value, did not demonstrate any incitement to or commission of a terrorist act, so the threshold was not met. Vernon therefore sits in productive tension with Watali: it accepts the no-mini-trial rule but insists that the material clear a minimal probative bar. Defence counsel routinely cite it to argue that ideological or associational material, without a demonstrable link to violence, cannot sustain a UAPA accusation at the bail stage.

Gurwinder Singh: the wall re-asserted

The pendulum swung back in Gurwinder Singh v. State of Punjab, 2024 INSC 92, decided on 7 February 2024. The Supreme Court refused bail to an accused alleged to have facilitated a terror module, and famously observed that the conventional maxim “bail is the rule, jail is the exception” does not apply with the same force to UAPA prosecutions — in this special-statute regime, the proviso to Section 43-D(5) means jail is, in effect, the rule and bail the exception. The Bench reaffirmed the Watali standard and laid down a structured twin-test: a court must first ask whether the statutory restriction in the proviso applies (the prima-facie inquiry), and only if it does not may it proceed to the conventional tripod tests of flight risk, tampering and influencing witnesses.

For aspirants, Gurwinder Singh is the case that confirms Watali remains good law and supplies the cleanest articulation of the sequence in which the tests are to be applied. It also shows the doctrine is contested terrain, oscillating between the Watali/Gurwinder rigour and the Najeeb/Vernon liberalisation.

The judgment distilled the bail inquiry into a two-stage exercise. At the first stage, the court applies the proviso to Section 43-D(5): reading the case-diary and chargesheet on a face-value basis as explained in Watali, is there a reasonable ground to believe the accusation is prima facie true? If yes, the statutory bar bites and bail is ordinarily refused; the court does not proceed further on the merits. Only if the bar is not attracted does the court reach the second stage, where the conventional considerations — gravity of the offence, likelihood of the accused fleeing, tampering with evidence or influencing witnesses, and the larger interest of society — come into play. The Court expressly noted that this special-statute sequence reverses the ordinary intuition that liberty is the starting point. It is worth observing the constitutional asymmetry: Gurwinder Singh speaks to the merits door and says nothing to displace the Najeeb delay door, which continues to operate independently. An accused failing the Gurwinder test on merits may still secure bail under Najeeb if incarceration becomes constitutionally excessive.

Yasin Malik: from guilty plea to life imprisonment

The third pillar of this note is the prosecution of Yasin Malik, chief of the proscribed Jammu and Kashmir Liberation Front (JKLF). In the NIA terror-funding case, Malik pleaded guilty to charges including offences under Sections 16, 17, 18 and 20 of the UAPA. In May 2022 a Special NIA Court convicted him, and on 25 May 2022 sentenced him to life imprisonment on two counts along with lesser concurrent sentences. Significantly, the trial court declined the NIA's prayer for the death penalty, holding the case did not fall within the “rarest of rare” category required since Bachan Singh v. State of Punjab, (1980) 2 SCC 684.

The conviction is a textbook illustration of the substantive terror-financing offences under Chapter IV and the consequences of membership of and support to a banned organisation — the same proscription machinery explained in our notes on declaration of an association as unlawful and the penalty for membership of an unlawful association.

Yasin Malik: the NIA's death-penalty appeal

Dissatisfied with life imprisonment, the NIA filed an appeal before the Delhi High Court seeking enhancement of Malik's sentence to death, contending that the gravity of the offences — described as waging organised secessionist violence — warranted capital punishment. The appeal has had a protracted procedural history, including the recusal of a judge and repeated listings, with the High Court directing Malik to file his response. As of mid-2026 the enhancement appeal remains pending and undecided; no appellate court has enhanced the sentence to death, and any statement that he has been sentenced to death would be incorrect.

The appeal is doctrinally instructive because it pits the prosecution's “rarest of rare” argument against the trial court's reasoned refusal, and tests whether a guilty plea and the absence of a contested trial affect sentencing severity. It is a live demonstration of how UAPA convictions feed into the general sentencing jurisprudence of Bachan Singh and Machhi Singh v. State of Punjab, (1983) 3 SCC 470.

The sentencing question is genuinely contested. The Bachan Singh framework requires a court to balance aggravating and mitigating circumstances and to record special reasons before imposing death, reserving it for cases where the alternative of life imprisonment is unquestionably foreclosed. The trial court treated the absence of direct loss of life attributable to the convict in the specific terror-funding charges, and the existence of life imprisonment as an adequate alternative, as decisive mitigating factors. The NIA's enhancement appeal contends that organised financing of secessionist violence is itself among the gravest offences against the State and meets the threshold. Because the conviction rested on a plea of guilt, the appellate court must also weigh whether such a plea operates as a mitigating circumstance, as the general sentencing jurisprudence suggests, or whether the gravity of the offences overrides it. Until the Delhi High Court rules, the case stands as an unresolved test of how far the death penalty extends to terror-financing under the UAPA.

Yasin Malik before the Supreme Court: the fair-trial dimension

Malik also features in Supreme Court proceedings arising from older J&K cases — the 1989 kidnapping of Rubaiya Sayeed and the 1990 killing of Indian Air Force personnel — where the central agency sought his physical production before a Jammu court. In April 2025 a Bench of Justices Abhay S. Oka and Ujjal Bhuyan declined physical production, citing a prohibitory order restricting Malik's movement out of Delhi, and instead permitted him to participate and cross-examine prosecution witnesses through video-conferencing from Tihar Jail. The Court emphasised that even a high-security undertrial is entitled to a fair trial, including a meaningful opportunity to test the prosecution's witnesses.

This thread links the Malik saga back to Article 21 and the fair-trial guarantees that animate Najeeb: security imperatives may shape the mode of trial, but cannot extinguish the accused's procedural rights. It is a useful counterpoint to the perception that UAPA cases suspend ordinary due-process norms.

The video-conferencing direction also illustrates how courts reconcile two competing public interests — the State's interest in not transporting a high-security convict across the country, and the accused's constitutional right under Article 21 to confront and cross-examine the witnesses arrayed against him. The right to cross-examination is an integral facet of a fair trial, and the Court's solution — remote participation with adequate technical facilities — preserves the substance of that right while accommodating legitimate security concerns. For aspirants, the takeaway is that the procedural protections of a fair trial are not suspended by the gravity of the charge or the notoriety of the accused; they are accommodated, not abrogated. This is consistent with the broader principle that the special UAPA regime modifies, but does not eliminate, the ordinary criminal-justice safeguards.

Drawing the threads together

Across these decisions a coherent map emerges. Watali fixes the merits standard — prima-facie truth on a face-value reading of the prosecution material, no mini-trial. Vernon refines it to require a surface analysis of probative value, blunting reliance on weak material. Najeeb supplies the constitutional escape valve, allowing bail on Article 21 speedy-trial grounds when incarceration becomes prolonged and the trial stalls. Gurwinder Singh re-asserts that, within the statutory regime, jail is the rule and bail the exception, and sequences the tests. And the Yasin Malik prosecution shows the full life-cycle of a UAPA case — conviction on a guilty plea, a pending death-penalty enhancement appeal, and persistent fair-trial protections even for the most serious accused.

For a structured revision sequence, pair this note with the foundational material in the UAPA notes hub and the conceptual building blocks in the definitions note. Mastery here is less about memorising paragraph numbers and more about knowing which door — merits or delay — an argument must walk through.

Exam pointers and common traps

Three traps recur in examinations. First, candidates wrongly assert that Najeeb overruled or diluted Watali; it did neither — it operates on the independent plane of Article 21 and delay. State this distinction explicitly to score. Second, candidates over-state the Yasin Malik outcome; he was sentenced to life imprisonment in May 2022, and the death-penalty appeal is pending, not decided. Third, candidates forget that Watali bars only a probing of admissibility and weight, not a basic reading of the material — Vernon's “surface analysis of probative value” qualifies but does not abolish that bar.

A model answer on UAPA bail should: (i) set out the proviso to Section 43-D(5); (ii) state the Watali prima-facie standard; (iii) note the Vernon refinement; (iv) explain the Najeeb Article 21 carve-out; and (v) close with the Gurwinder Singh sequencing of tests. That five-step structure converts these landmark cases into a complete, defensible answer.