Authored By: Debangshi Paswan
University of Calcutta
Case Details
Case Name: Prajwala v. Union of India
Citation: 2026 INSC 609
Court: Supreme Court of India
Date of Decision: May 29, 2026
Bench Composition: Justice J.B. Pardiwala and Justice R. Mahadevan
Background
The Prajwala v. Union of India (2026) closes a case that opened in 2004. After two decades, the Supreme Court recognised that identifying and rescuing trafficking victims is only a part of the State’s duty, and that rehabilitation is owed to them as a matter of constitutional right rather than a policy grace. The court read Article 21’s guarantee of a dignified life together with Article 23’s prohibition on trafficking to establish this right. The judgment is worth close study not only for the right it declares, but the machinery it builds to secure that right.
III. Facts of the Case
Prajwala, an anti-trafficking organisation based in Hyderabad, filed Writ Petition (Civil) No. 56 of 2004, arguing that the law and institutions governing the rescue and rehabilitation of victims of trafficking were inadequate, and the survivors were treated as offenders rather than victims.[1] Hearings stretched over a decade, drawing in the National Legal Services Authority, the National Commission for Women, and the States. On December 9, 2015, the original petition was disposed of after the Union Government undertook to set up an Organised Crime Investigation Agency (OCIA) and to draft anti-trafficking legislations through an Inter-Ministerial Committee.[2]
Neither promise was kept. Proposed bills were introduced in 2016, 2017, 2018, and 2021; The 2018 bill successfully passed in Lok Sabha but did not proceed further due to the dissolution of the House. Prajwala withdrew its first attempt to enforce the 2015 order in 2018 once the Bill’s passage seemed imminent.[3] However, the OCIA was never created. Instead, a 2019 Amendment brought trafficking offences under the IPC into the Schedule of the National Investigation Agency Act, 2008.
In 2022, Prajwala filed the present application seeking enforcement of the 2015 order. The Union’s position shifted twice during the proceedings: first, it argued that NIA’s jurisdiction made a separate OCIA unnecessary. Later, it argued that the new criminal codes of 2023, namely BNS, BNSS, and BSA together with ITPA made a dedicated trafficking law unnecessary as well.[4] This change in the Union’s stance along with the decade of inaction behind it were examined by the court when it delivered the judgement on May 29, 2026.
Issues for Consideration
The following issues were considered by the Court:[5]
Whether Articles 21 and 23 of the Constitution read together, entitles the victims of trafficking for Commercial Sexual Exploitation to a right to rehabilitation?
Whether a gap exists in the legislative and institutional framework governing their rescue, prohibition, rehabilitation, and repatriation?
iii. Whether the Court should direct the creation of the OCIA?
Arguments Presented
For the Petitioner, Ms. Aparna Bhat, Senior Advocate, argued that nothing of substance had changed since 2015. The ITPA retained its original 1956 structure, while the NIA had addressed only a limited number of trafficking cases. Additionally, the Anti-Human Trafficking Units established under its jurisdiction lacked coordination across states and often operated without being designated as police stations in most regions.[6] She criticized the Shakti Sadan homes as being inadequately funded and unevenly allocated. Furthermore, she contended that legal actions seldom extended beyond those apprehended at the raid site, leaving financiers and traffickers unscathed.[7] Her argument was that the Union’s changed stance, delayed by a decade, amounted to an opportunistic withdrawal from a previously established commitment.
For the Union, Ms Aishwarya Bhati, Additional Solicitor General, catalogued the steps taken since 2015: the 2019 NIA Act amendment, the introduction of organised-crime and trafficking offences under the BNS, victim-compensation obligations under the BNSS, 827 functional Anti-Human Trafficking Units as of January 2025, over 14,000 Women Help Desks, and a national cyber-crime information platform.[8] Since policing is a state subject under the Seventh Schedule, she insisted that primary responsibility lay with the States. Moreover, she asserts that the existing framework, taken as a whole, was more robust than what had originally been envisaged for the OCIA and a standalone statute.
Court’s Reasoning
The Court rejected any single-cause account of trafficking, describing it as a phenomenon produced by poverty, social exclusion and increasing cyber-enabled recruitment, which no purely criminal or purely structural response could address alone.[9] In regards to the Constitutional question, the court read Article 21’s guarantee of a dignified life against Article 23’s prohibition on trafficking, and relying on its own bonded-labour jurisprudence, it ruled that rescue without rehabilitation forces the victims back into the very circumstances that caused them to be targeted in the first place.[10] The Court concluded there was no principled basis for confining that reasoning to bonded labour rather than extending it to trafficking, since both are forms of exploitation Article 23 was written to prohibit.
The standard established by the Court defines true rehabilitation through a threefold model of dignity, namely material support, including shelter, healthcare, monetary-relief, and vocational-training, social recognition, countering marginalization and prejudice, and inherent respect, ensuring aid is guided by the survivors’s autonomy rather than State compulsion.[11] Assessed against this standard, the Court concluded that the Union’s purported legal progress had stagnated since 2015. Furthermore, ongoing welfare frameworks suffered from systematic failures in execution. Only eight states had implemented regulations under Section 23 of the ITPA, while major regions like Gujarat, Haryana, Uttarakhand, and Uttar Pradesh lacked even a single operational Shakti Sadan home.[12]
Addressing the ITPA’s rescue procedure, the Court observed that Section 17 fails to differentiate between victims of trafficking and adults engaging in consensual sex work, subjecting both to identical majesterial proceedings. Relying on its own interpretation of Section 17 alongside the principle of non-interference established in Budhadev Karmaskar, it determined that a magistrate is obligated to ascertain, at the threshold, whether the individual before them is a voluntary adult sex worker who opposes long term State custody, preceding any extensive enquiry intended for trafficking cases.[13] Regarding those actually trafficked, the Court emphasized that the victim’s choice must remain the primary determinant for detention or familial reintegration when deviations permitted only when necessary for physical safety or where the victim’s stated wishes appear to be influenced by her former exploiter.[14]
In the matter of the OCIA, the Court reasoned that its intended functions of investigation, coordination, and data sharing, were already dispersed across the NIA, State Anti-Trafficking Bureaus, and the CBI. Furthermore, it ruled that commanding Parliament to establish a new statutory body would trespass judicial boundaries, contravening its own precedent barring courts from mandating legislation.[15]
VII. Judgment
The Court ruled that Articles 21 and 23, read together, guarantees victims of trafficking a constitutional right to rehabilitation. Furthermore, it declared that the Union had failed to take reasonable measures to secure this right.[16] Exercising its power enshrined under Articles 32 and 142, the Court issued a comprehensive nine-principle Victim Protection Plan to remain in effect until the Parliament legislates. Built upon the foundational principles of human rights “primacy”, “non-criminalisation”, “informed consent”, “non-stigmatisation”, “Safety and Protection”, “Privacy & Confidentiality”, “Right to Information”, “individualised care”, and the victim’s “best interest”, it encompasses six distinct stages, namely, “pre-rescue”, “rescue”, “post-rescue”, “rehabilitation”, “reintegration”, and “prosecution”.[17] It declined to direct the creation of the OCIA, leaving that choice to the Union, but issued four narrower directions: that States notify recognised welfare institutions, prepare panels of social welfare workers, and designate ADGP-rank police officers and WCD Secretaries as nodal officers, all within three months, with compliance to be reported in September 2026.[18]
Ratio Decidendi: The case turns on two linked propositions. The first is substantive; Articles 21 and 23, read together, place the State under a real obligation to rehabilitate victims of trafficking, an obligation that begins where rescue ends and covers material support, protection from stigma, and respect for the victim’s own choices.[19] The second is institutional; a court may use its powers under Articles 32 and 142 to write interim rules that hold the field until Parliament acts, but that power stops short of directing Parliament to create a new statutory body, since that would cross from filling a gap in the law into designing the law itself.[20]
VIII. Analysis
Strengths: The judgment’s principal strength is doctrinal. It closes a gap that the Court’s own bonded-labour cases had left open, since none of those earlier judgments extended their reasoning to trafficking in terms. That extension holds up for two reasons: trafficking and bonded labour both engage Article 23, and the Court supplies working content for the right through its three dimensions of dignity, rather than leaving rehabilitation an undefined aspiration.[21] The threshold inquiry built into the rescue procedure is a further, more practical strength. By requiring a magistrate to ask, at the first stage, whether an adult before him is a voluntary sex worker, the Court gives Budhadev Karmaskar’s principle of non-interference a concrete procedural home, rather than leaving it as a general instruction police could construe as they wished.[22]
Criticism: The judgment is more uneven where it draws lines rather than declares rights. The Court states outright that sex workers’ rights can exist independently of any right to sex work, and it builds the threshold inquiry described above around that acknowledgment.[23] That acknowledgment sits, however, in the section headed Recommendations and Suggestions, addressed to the legislature, rather than in the operative Victim Protection Plan issued under Articles 32 and 142.[24] The threshold inquiry is enforceable, since it changes how Section 17 ITPA is read, while the underlying recognition of sex workers’ rights carries only the weight of advice rather than of a binding direction. A woman relying on that recognition to resist a removal order under Section 20 ITPA has less to stand on than a trafficking victim invoking the Victim Protection Plan, even though the Court’s own reasoning treats both as owed dignity.
A second criticism sits in the Court’s restraint over the OCIA. It declines a mandamus for a single coordinating body, citing the rule against directing Parliament to legislate, then proceeds, in the same judgment, to prescribe the composition of every district Anti-Human Trafficking Unit, the sequence of a rescue operation, and the wording of a magistrate’s threshold inquiry.[25] Both are exercises of the same Article 142 power, and the second is, on any fair reading, the more intrusive of the two. If institutional restraint counselled against creating one coordinating body, the same restraint might equally have counselled against drafting what is, in effect, a working manual for anti-trafficking policing.
Legacy: The judgment’s legacy will turn on compliance rather than on the content of the Plan itself, since this is not the first attempt of its kind. In 1990, Vishal Jeet v. Union of India directed the creation of Central and State Advisory Committees on trafficking; Prajwala’s own submissions recorded that the Central Advisory Committee had not functioned in twelve years.[26] The 2015 order in this very case produced a similar outcome. The binding directions issued this time, four modest administrative steps with a three-month deadline, are considerably narrower than the Victim Protection Plan they are meant to anchor, and only those four are tied to the September 2026 compliance hearing.
Whether the wider Plan takes hold, or joins the Vishal Jeet advisory committees in disuse, will depend largely on litigants willing to return to Court, the same dependence that produced two miscellaneous applications and a two-decade delay this time.
Conclusion
Prajwala v Union of India confirms that Articles 21 and 23 impose a judicially enforceable obligation on the State to rehabilitate trafficking victims rather than merely rescue them, and gives that obligation working content through material, relational and autonomy-respecting dimensions of dignity. Its Victim Protection Plan is among the most administratively detailed uses of Article 142 in recent years, and its threshold inquiry for voluntary sex workers marks a genuine procedural advance on Budhadev Karmaskar. The judgment’s own history, however, should temper optimism about what follows. This is the second time the Court has issued detailed anti-trafficking directions, the Vishal Jeet committees having lapsed into disuse, and the compliance mechanism built for its own order is narrower than the Plan it is meant to secure. Whether Prajwala succeeds where Vishal Jeet and the 2015 order did not will depend less on the content of the Plan than on whether the September 2026 hearing, and those that follow it, are used to hold the State to it.
Reference(S):
[1] 1 Prajwala v Union of India, Miscellaneous Application No 530 of 2022 in Writ Petition (Civil) No 56 of 2004, 2026 INSC 609 (SC, 29 May 2026) [29]–[31].
[2] ibid [39]–[43].
[3] ibid [45]–[47].
[4] ibid [50]–[53].
[5] ibid [80].
[6] ibid [57]–[64].
[7] ibid [65]–[67].
[8] ibid [73]–[78].
[9] ibid [90].
[10] ibid [278]–[280], relying on Bandhua Mukti Morcha v Union of India (1984) 3 SCC 161; Neeraja Chaudhary v State of Madhya Pradesh (1984) 3 SCC 243; Public Union for Civil Liberties v State of Tamil Nadu (2004) 12 SCC 381.
[11] ibid [286]–[289].
[12] ibid [297]–[300].
[13] ibid [334]–[336], applying Budhadev Karmaskar v State of West Bengal (2022) 20 SCC 220.
[14] ibid [339]–[349].
[15] ibid [378]–[382], relying on Union of India v K Pushpavanam (2023) 20 SCC 736.
[16] ibid [439]–[446].
[17] ibid [303], [362], [447].
[18] ibid [454], [456]–[458].
[19] ibid [439]–[446].
[20] ibid [378]–[386].
[21] ibid [278]–[289].
[22] ibid [334]–[336], applying Budhadev Karmaskar v State of West Bengal (2022) 20 SCC 220.
[23] ibid [455]
[24] ibid, section VII ‘Recommendations and Suggestions’ [455].
[25] ibid [362], [378]–[386].
[26] ibid [37], [68], citing Vishal Jeet v Union of India (1990) 3 SCC
