Article 21 “Melts Down” Section 43-D(5) UAPA in Prolonged Incarceration: K.A. Najeeb as Binding Stare Decisis and the Impermissibility of Smaller-Bench Dilution

1. Introduction

In SYED IFTIKHAR ANDRABI v. NATIONAL INVESTIGATION AGENCY, JAMMU (2026 INSC 503, decided on 18-05-2026), the Supreme Court examined the constitutional limits of the statutory bail embargo under Section 43-D(5) of the Unlawful Activities (Prevention) Act, 1967 (“UAP Act”) in the face of prolonged pre-trial incarceration and a trial unlikely to conclude within a reasonable time.

The appellant, a government employee and a political activist, was arrested on 11.06.2020 in a prosecution combining NDPS and UAP Act allegations (described by the parties as “narco-terrorism”/terror-funding through narcotics). His regular bail was rejected by the Special NIA Court (10.08.2024) and the rejection was affirmed by the High Court of Jammu & Kashmir and Ladakh (19.08.2025). By the time the Supreme Court heard the matter, the appellant had undergone about 5 years 11 months in custody, with the prosecution indicating 350+ witnesses still to be examined.

The case raised two interconnected issues: (i) how constitutional courts should reconcile the UAP Act’s stringent bail regime with Article 21 (personal liberty and speedy trial), and (ii) whether smaller Benches may effectively dilute a larger Bench precedent (notably Union of India v. K.A. Najeeb) without formally disagreeing and without making a reference.

2. Summary of the Judgment

  • The Supreme Court allowed the appeal and granted regular bail to the appellant.
  • It held that Union of India v. K.A. Najeeb (three-Judge Bench) remains binding law: statutory restrictions such as Section 43-D(5) UAP Act do not oust the constitutional courts’ power to grant bail where Part III (especially Article 21) is infringed due to gross delay and prolonged incarceration.
  • The Court expressed concern about two-Judge Bench decisions GURWINDER SINGH v. STATE OF PUNJAB and Gulfisha Fatima v. State (Govt. of NCT of Delhi) insofar as they depart from or “hollow out” K.A. Najeeb, reiterating that the only proper course for a smaller Bench that disagrees with a larger Bench is to refer the matter.
  • On facts, it found the appellant’s continued detention oppressive given: (i) no recovery from his person/premises, (ii) heavy reliance on police statements/confessional material, (iii) lack of prior antecedents, (iv) compliance during earlier interim bail, and (v) the trial’s remote prospect of early completion.
  • It refused the suggestion that bail be deferred by directing trial completion within a fixed time, relying on High Court Bar Association, Allahabad v. State of Uttar Pradesh and Roop Bahadur Magar alias Sanki alias Rabin Vs. State of West Bengal.
  • It additionally referenced NCRB-derived conviction statistics under the UAP Act (2019–2023) to underscore the constitutional risk of prolonged pre-trial detention in a regime with low conviction rates.

3. Analysis

3.1 Precedents Cited (and How They Shaped the Outcome)

(A) The constitutional “speedy trial” line

Shaheen Welfare Association v. Union of India was used as an early articulation of the dilemma posed by stringent anti-terror bail regimes: excessive trial delay harms both societal interest (fading evidence) and individual liberty (innocents languishing in custody). The present judgment uses Shaheen to set the conceptual frame: anti-terror statutes cannot become instruments of indefinite pre-trial punishment.

Javed Gulam Nabi Shaikh v. State of Maharashtra reinforced that Article 21 applies irrespective of the nature of crime, and that the State cannot resist bail merely by invoking seriousness when it cannot ensure speedy trial. The Court relied on this to reject the prosecution’s “seriousness of offence” argument as a complete answer to prolonged incarceration.

Sheikh Javed Iqbal v. State of U.P. was pivotal in two ways: it reiterated that prolonged incarceration with no end in sight makes seriousness an insufficient ground to deny bail, and it clarified the correct reading of NIA Vs. Zahoor Ahmad Shah Watali, stating Watali cannot be treated as a precedent to deny bail in long-incarceration cases. The present judgment draws directly from this approach and adopts its interpretive stance.

ARVIND DHAM v. DIRECTORATE OF ENFORCEMENT (though under PMLA) was treated as doctrinal reinforcement: stringent statutory bail conditions do not eclipse Article 21, and prolonged pre-trial detention risks converting custody into punishment. The Court used it to show cross-statute consistency of Article 21 reasoning.

(B) The controlling authority on UAP Act bail and Article 21

Union of India v. K.A. Najeeb is treated as the binding constitutional lodestar. The Court foregrounded Najeeb’s holding that: “the rigours of such provisions will melt down where there is no likelihood of trial being completed within a reasonable time”. Importantly, the Court framed this not as an equitable one-off, but as an authoritative constitutional limitation on Section 43-D(5)’s operation.

In practical terms, Najeeb supplied the decisive test the Court applied here: (i) incarceration already very long, (ii) trial not completing in reasonable time, and (iii) continued custody would breach Article 21.

(C) Watali and the “prima facie true” standard

NIA Vs. Zahoor Ahmad Shah Watali was explained through the lens supplied by Najeeb and Sheikh Javed Iqbal v. State of U.P.: Watali restrains courts from conducting a mini-trial at the bail stage, but does not mandate indefinite detention despite delay. The present judgment treats Watali as context-specific, refusing its use as a universal anti-bail trump.

(D) Smaller-bench deviation and judicial discipline

GURWINDER SINGH v. STATE OF PUNJAB and Gulfisha Fatima v. State (Govt. of NCT of Delhi) were addressed directly. The Court held that both reflect a “divergent view” from Najeeb’s clear trajectory and expressed “serious reservations” about aspects of Gulfisha Fatima (including “foreclosing” bail applications for a year).

The critical doctrinal contribution is the Court’s restatement of stare decisis and bench-strength discipline: a two-Judge Bench cannot dilute or circumvent a three-Judge Bench precedent; disagreement must trigger a reference. This is presented as a systemic safeguard preventing “progressive hollowing out” of constitutional holdings.

(E) Trial time-limits as a substitute for bail

The prosecution’s proposal—deny bail now, try to finish trial in one year, then reconsider—was rejected using High Court Bar Association, Allahabad v. State of Uttar Pradesh (Constitution Bench) and Roop Bahadur Magar alias Sanki alias Rabin Vs. State of West Bengal. These cases caution that superior courts should not routinely impose fixed time-lines on trial courts, both because it distorts judicial prioritisation and because it privileges litigants who can reach constitutional courts. The present judgment adopts this as a normative rule: time-bound trial directions are not an adequate remedy for prolonged incarceration except in extraordinary situations.

(F) NDPS bail rigour and prolonged detention

Though the main constitutional conflict was framed around UAP Act Section 43-D(5), the case also involved NDPS offences and the stringent bail provision under Section 37 NDPS Act. The Court’s survey included Satender Kumar Antil v. CBI (as cited within Javed Gulam Nabi Shaikh v. State of Maharashtra) for the idea that principles like Section 436A CrPC (long incarceration) and Article 21 can cut across special statutes.

It further relied on Chintan Rajubhai Panseriya v. State of Maharashtra to show that even in serious NDPS matters, prolonged custody combined with an unmanageably large witness list can justify bail, without “undermining seriousness.”

(G) Case-specific parity references

The appellant invoked parity via bail granted to co-accused in the same prosecution, including: Romesh Kumar v. Union of India, Special Leave Petition (Criminal) No. 13829 of 2024, decided on 07.02.2025, Islam Ul Haq Peer v. Union of India, Mudasir Ahmed Dar Vs. NIA, and Amin Allaie Vs. NIA. While the Supreme Court ultimately granted bail on constitutional and factual grounds, it used these orders to demonstrate that lengthy incarceration and trial delay were already being treated by courts as constitutionally relevant—thus strengthening the reasonableness of granting bail to an accused with even longer custody.

(H) Reinforcement of “bail as rule” as constitutional principle

The Court cited Jalaluddin Khan v. Union of India to reaffirm that denial of bail in deserving cases can violate Article 21. In a notable doctrinal move, it declared that “bail is the rule and jail is the exception” is not merely a CrPC slogan but a constitutional principle grounded in Articles 21 and 22 and the presumption of innocence—even under the UAP Act, subject to case-specific denial.

3.2 Legal Reasoning

(A) Harmonisation, not subordination: Section 43-D(5) under Article 21

The Court’s reasoning proceeds from hierarchy: the UAP Act is subordinate to the Constitution. It accepts that at the “commencement” of proceedings courts should respect the legislative policy of restraint under Section 43-D(5), but holds that where incarceration becomes prolonged and trial completion is not reasonably foreseeable, the statutory rigour “melts down” (Najeeb) to prevent Article 21 from being reduced to a dead letter.

(B) Rejecting the “invented enemy” critique of Najeeb

A central portion of the judgment rebuts the interpretive move (attributed to Gurwinder and Gulfisha Fatima) that Najeeb creates a simplistic “passage of time = bail” formula. The Court clarifies that Najeeb’s warning was the opposite: not that delay must be the sole ground for bail, but that Section 43-D(5) must not become the sole ground to justify endless detention while ignoring Article 21.

(C) Stare decisis and institutional integrity

The judgment treats bench-strength discipline as a constitutional-institutional necessity: smaller Benches must follow larger Bench law or refer. The Court explicitly frames “progressively hollowing out” as a structural problem—especially acute in liberty cases—because it allows constitutional protections to erode without an open doctrinal confrontation.

(D) Why the Court still considered some factual factors

While careful not to conduct a “mini-trial,” the Court listed features supporting release: no recovery from the appellant’s person/premises; heavy reliance on police statements implicating the appellant; no prior antecedents shown; his compliance with interim medical bail; and the political/government-service background noted in the preventive detention dossier. This was not a merits adjudication, but a proportionality assessment relevant to the constitutional necessity of continued detention during a trial that will not end soon.

(E) The statistical turn

Unusually, the Court used publicly furnished arrest/conviction statistics (NCRB-based, as provided in Parliament) to highlight the systemic risk of long pre-trial detention under UAP Act amid low conviction rates—particularly in Jammu & Kashmir. The statistics were not used to prejudge the appellant’s innocence, but to intensify the Article 21 concern: if most cases end in acquittal, prolonged detention becomes constitutionally more suspect absent speedy trial.

3.3 Impact

  • Re-centering K.A. Najeeb: The judgment is a strong reassertion that Najeeb is the governing authority on the Article 21 override to Section 43-D(5), and that it is not a narrow “exception” but a constitutional rule of limitation.
  • Bench-strength discipline in liberty jurisprudence: It crystallises a warning against doctrinal erosion by smaller Benches. This may influence both Supreme Court and High Court bail practice by discouraging reliance on narrower readings of Najeeb without reference.
  • Constraint on “time-bound trial” substitutes: Courts may become less willing to deny bail while issuing “finish trial in X months” directions, in light of High Court Bar Association, Allahabad v. State of Uttar Pradesh.
  • Cross-statute resonance: By drawing from NDPS and PMLA bail-delay cases, the judgment strengthens a general constitutional approach: stringent bail clauses are not permission for punitive pre-trial detention.
  • Potential evidentiary implications (indirect): The Court’s emphasis on the primacy of Article 21 alongside references to police-confession constraints (Section 25 Evidence Act) and contested “disclosure/recovery” narratives may encourage more careful judicial scrutiny at the level of whether continued detention remains proportionate.

4. Complex Concepts Simplified

Section 43-D(5) UAP Act (“prima facie true”)
It restricts bail where, based on the case diary/charge-sheet, the court finds reasonable grounds to believe the accusation is prima facie true. This judgment reiterates that even if that threshold is met, constitutional courts can grant bail if Article 21 is violated by prolonged incarceration and delayed trial.
“Non-obstante clause”
A phrase like “notwithstanding anything contained…” gives the provision overriding effect over conflicting laws (e.g., CrPC), but it cannot override the Constitution. That is the hierarchy the Court enforces.
Stare decisis and bench strength
Decisions of a larger Bench bind smaller Benches. If a smaller Bench disagrees, it must refer the matter for a larger Bench’s reconsideration; it cannot dilute the precedent indirectly.
“Bail is the rule” as a constitutional principle
The judgment frames this not as a mere CrPC default but as flowing from Article 21/22 and the presumption of innocence—meaning special statutes can tighten bail but cannot normalise indefinite detention.
Section 25 and Section 27 of the Indian Evidence Act, 1872
Section 25 generally bars confessions made to police from being proved against the accused. Section 27 is a limited exception: it allows proof of that part of information leading distinctly to discovery of a fact. The Court notes the case’s reliance on police statements while avoiding a final merits ruling at the bail stage.
“Time-bound trial” directions
Superior courts sometimes refuse bail while directing the trial court to finish quickly. The Constitution Bench in High Court Bar Association, Allahabad v. State of Uttar Pradesh cautions such directions should be exceptional, because they distort trial-court priorities and create inequality among litigants.

5. Conclusion

This judgment is significant for two reasons. First, it operationalises the constitutional proposition that Section 43-D(5) UAP Act cannot justify indefinite pre-trial incarceration where trial delay makes continued custody inconsistent with Article 21; in such cases the statutory rigour “melts down” as declared in Union of India v. K.A. Najeeb.

Second, it delivers an institutional message: smaller Benches cannot “hollow out” larger Bench constitutional holdings without reference. By granting bail after nearly six years of custody amid an unfinishable witness list, and by rejecting “finish trial in one year” as a routine substitute for bail, the Court reasserts liberty-centric constitutionalism in the administration of stringent penal statutes.