Case Summary: Sukanya Shantha v Union of India & Ors [WP (C) No 1404 of 2023]

Sukanya Shantha v. Union of India & Ors
WP (C) No. 1404 of 2023 | Supreme Court of India | Decided: 3 October 2024

Full Case Name: Sukanya Shantha v. Union of India & Ors
Citation: WP (C) No. 1404 of 2023; 2024 INSC 753
Court: Supreme Court of India
Bench: Dr. Dhananjaya Y. Chandrachud, C.J.I., Justice Jamshed B. Pardiwala, Justice Manoj Misra
Date of Judgment: 3 October 2024
Authored by: Dr. Dhananjaya Y. Chandrachud, Chief Justice of India
Citation: Sukanya Shantha v Union of India & Ors WP (C) No 1404 of 2023 (Supreme Court of India, 3 October 2024)

Background and Facts

This case began with a piece of investigative journalism. On 10 December 2020, Sukanya Shantha published an article in The Wire titled ‘From Segregation to Labour, Manu’s Caste Law Governs the Indian Prison System.’[1] What the article documented was not a historical relic but a live administrative practice: caste-based discrimination in India’s prisons was not merely persisting informally — it was being written into official rules and manuals. Shantha identified three main categories. First, prisoners were being physically separated into different barracks based on caste. Second, Dalit prisoners and members of Denotified Tribes (DNTs) were being assigned menial and sanitation work, including manual scavenging. Third, certain communities were being labelled ‘habitual offenders’ — a classification whose origins lie directly in the British-era Criminal Tribes Act of 1871, a law that deemed entire communities criminal by birth.

Drawing on Shantha’s findings, the petitioner filed a writ petition under Article 32 of the Constitution, challenging specific provisions in State prison manuals across Uttar Pradesh, West Bengal, Maharashtra, Karnataka, Tamil Nadu, Andhra Pradesh, Orissa, and Jharkhand. The petition also challenged the Model Prison Manual 2016[2] and the Model Prisons and Correctional Services Act 2023[3] issued by the Union government. The language in these manuals is worth dwelling on for a moment: ‘prison-cooks of suitable caste,’ prisoners with a ‘natural tendency to escape,’ prisoners of a ‘superior mode of living.’ The petitioner’s point — and it seems to me an obvious one — was that this was not neutral administrative shorthand. These were ways of encoding caste hierarchy into official rules while giving it the appearance of practical necessity.

Legal Issues

The Supreme Court identified three main questions to be decided:

1. Whether caste-based segregation of prisoners, the allocation of prison labour along caste lines, and the labelling of certain communities as ‘habitual offenders’ in State prison manuals and central model documents violate Articles 14, 15, 17, 21, and 23 of the Constitution of India.
2. Whether terms like ‘habit,’ ‘custom,’ ‘superior mode of living,’ and ‘natural tendency to escape’ are constitutionally acceptable ways to classify prisoners, or whether they are simply indirect ways of discriminating based on caste.
3. Whether the Model Prison Manual 2016 and the Model Prisons and Correctional Services Act 2023 adequately address caste discrimination or whether they need to be amended.

Summary of Arguments

A. For the Petitioner

Senior Advocate Dr. S. Muralidhar and Advocate Disha Wadekar made a case that was, at its core, quite simple: this was not accidental. Dalit prisoners and DNT members being assigned to sanitation and menial work was not an unfortunate by-product of neutral administration — it was how these institutions were designed. The distinction the petitioner drew between direct discrimination (manuals explicitly using caste to assign work) and systemic discrimination (vague criteria like ‘habit’ and ‘custom’ that consistently produce the same caste-stratified outcomes) is the argument the Court ultimately accepted, and it is not hard to see why.

The ‘habitual offender’ classification was also challenged as colonial thinking that had simply never been cleaned up — targeting DNTs without proper statutory definition or legislative backing. On the Model Prison Manual 2016, the petitioner’s point was direct: it nominally prohibited some discrimination but had no binding force over States, which meant it was essentially advisory wallpaper. The asks were correspondingly concrete — a declaration that the relevant provisions were unconstitutional, directions to States and UTs to revise their manuals, and removal of caste references from prison registers.

B. For the Respondents

Additional Solicitor General Aishwarya Bhati, for the Union, pointed to the Model Prison Manual 2016 as evidence that the Centre had already moved — it explicitly prohibited caste and religion-based discrimination in kitchen duties. She acknowledged that prisons fall under State jurisdiction (Entry 4, List II, Seventh Schedule) and said the Union had done what it could through advisory guidelines. The States were less conciliatory. Maharashtra, Uttar Pradesh, and Jharkhand argued that caste-based separation of prisoners, where it existed, was a security measure to prevent conflict — not discrimination. They also argued, with varying degrees of conviction, that whatever discriminatory language remained in their manuals was outdated and not being actively applied.

Judgment and Ratio Decidendi

The Supreme Court, unanimous, struck down the challenged provisions in the State prison manuals, the Model Prison Manual 2016, and the Model Prisons and Correctional Services Act 2023. The judgment was authored by Chief Justice Chandrachud and held that these provisions violated Articles 14, 15, 17, 21, and 23 of the Constitution. Wide-ranging directions for reform followed.

Under Article 14, the Court’s central finding was that indirect and systemic discrimination is as prohibited as the direct kind — neutral-looking criteria that consistently fall hardest on the same marginalised communities cannot be dressed up as administration. The manual terms ‘habit,’ ‘custom,’ and ‘natural tendency to escape’ were found vague and rationally disconnected from any legitimate prison purpose.

Article 15 followed the same logic. Caste-based classification — whether openly stated or working through proxies — is prohibited. The only exception is where it operates in favour of marginalised groups to achieve substantive equality, which is obviously not what the prison manuals were doing.

On Article 17, the Court’s position is, arguably, the most significant part of the judgment — and the most honest. It held that the abolition of untouchability is not a formal textual command to be discharged and forgotten. It is a living directive that the State must actively uphold within its own institutions, including prisons. Assigning manual scavenging and sanitation duties to Dalit prisoners on the basis of their caste was found to amount to enforcing untouchability in an institutional form. The Court drew on the vision of Dr. B.R. Ambedkar and the Constituent Assembly debates to ground this interpretation — and rightly so, because Ambedkar was clear that formal abolition without substantive transformation was not what the Constitution was designed to achieve.

The Article 21 holding is more unusual. The Court expanded the right to life and personal liberty to include a ‘right to overcome caste prejudices’ — framing prisons as spaces meant for rehabilitation, not for entrenching the hierarchies prisoners arrive with. Caste-based segregation, the Court found, does exactly the latter.

The ratio comes down to this: segregating prisoners by caste, assigning labour on caste lines, and labelling marginalised communities as ‘habitual offenders’ through constitutionally vague criteria has no rational connection to any legitimate penological objective. These practices violate Articles 14, 15, 17, 21, and 23. The constitutional bar on caste-based discrimination applies inside prisons, and the State cannot dress up discrimination as security or administrative convenience.

Directions Issued:
(i) All States and Union Territories directed to revise their prison manuals within three months in line with the judgment.
(ii) The Union government directed to amend the Model Prison Manual 2016 and the Model Prisons and Correctional Services Act 2023 to remove all caste-based discriminatory provisions.
(iii) All caste-based columns and references in prison registers directed to be deleted.
(iv) The Court registered the broader issue of discrimination inside prisons as ‘In Re: Discrimination Inside Prisons in India’ for ongoing judicial oversight.

Critical Analysis

This judgment matters, and it is worth trying to say precisely why — both what it does doctrinally and what it reveals about the gap between constitutional promise and institutional reality. The analysis below works through the key contributions and the most serious limitations.

A. Doctrinal Contribution: Systemic Discrimination and the Right to Overcome Caste Prejudice

The most significant doctrinal development in this judgment is the recognition of a ‘right to overcome caste prejudices’ as part of the right to life under Article 21. What makes this interesting is not just that it is new, but the reasoning behind it. Earlier expansions of Article 21 — in cases like Navtej Singh Johar v. Union of India (2018)[4] and Justice K.S. Puttaswamy v. Union of India (2017)[5] — tended to focus on individual autonomy and liberty. Sukanya Shantha does something different: it situates the expansion within a shared historical experience, treating the harm of caste not as a personal affront to one prisoner but as a structural injury reproduced by institutional design. Chief Justice Chandrachud has taken this kind of approach before — treating the Constitution as a document meant to actively transform existing social arrangements, not just reflect them. Whether this newly recognised right will extend beyond the prison context is genuinely uncertain, and that uncertainty itself raises interesting questions for future constitutional litigation.

The Court’s treatment of indirect discrimination under Articles 14 and 15 deserves attention. The holding that facially neutral criteria — terms like ‘habit,’ ‘custom,’ and ‘superior mode of living’ — can be unconstitutional when their practical effect consistently disadvantages marginalised communities is a meaningful step. Lt. Col. Nitisha v. Union of India (2021)[6] had moved in this direction; Sukanya Shantha goes further by applying the framework to a context where the discriminatory pattern is unusually well-documented. The implications are potentially wide: similar neutral-looking criteria operate in housing allocation, civic services, and educational institutions. Whether courts will extend this reasoning beyond prisons is a question likely to be litigated in the next few years.

B. Consistency with Precedent

The judgment sits comfortably within existing doctrine, even as it pushes it further. The E.P. Royappa principle[7] — that Article 14 prohibits manifestly arbitrary classification, not just formal discrimination — does the foundational work on the prison manual criteria. The reading of Article 17 as an active directive binding on State institutions picks up from People’s Union for Democratic Rights (1982)[8] in a way that is clearly reasoned rather than invented. And the substantive content given to ‘procedure established by law’ under Article 21 via Maneka Gandhi (1978)[9] is, at this point, settled ground. What feels genuinely new is not any single principle but the way the Court assembles them here — each reinforcing the others into an argument that is stronger than any one of them would produce on its own.

The alignment with the United Nations Standard Minimum Rules for the Treatment of Prisoners (the Nelson Mandela Rules) is worth flagging, even though the Court does not lean on it. The Rules prohibit discrimination in prison administration on grounds including social origin. The Court’s restraint here — keeping the argument on constitutional ground rather than reaching for international instruments — looks deliberate. Bringing in international law would have opened a debate about domestic legal status that simply was not necessary. The consonance matters anyway, not as authority, but as confirmation that India’s obligations sit within a broader developing norm around prison dignity.

C. Limitations and Implementation Concerns

The doctrinal analysis in the judgment is strong. The implementation side is where the greatest concern lies, and it is arguably the most serious limitation of the decision as a practical matter. The three-month deadline for all States and Union Territories to revise their prison manuals is strikingly optimistic. India’s prisons are among the most overcrowded and under-resourced public institutions in the country. Caste hierarchies within them do not stop at the prisoner population — they extend to prison staff, to procurement, to informal authority structures that a revised manual will not automatically displace. Rewriting a rulebook does not change who sweeps the floor. That requires resources, training, monitoring, and accountability mechanisms that a Supreme Court order cannot conjure.

The suo motu registration of ‘In Re: Discrimination Inside Prisons in India’ shows the Court is aware that a single judgment is not enough. But past experience with continuous mandamus proceedings gives reason for scepticism. The forest rights litigation and the Vishaka framework on sexual harassment at work both show the same pattern: structural orders that produce genuine progress in some States and stagnation in others, with the Court lacking effective tools to distinguish between the two. What is missing here — and what the judgment does not create — is an independent body with real institutional capacity and enforcement authority to monitor compliance with prison reform. Without that, compliance will depend almost entirely on how seriously individual State governments choose to treat this judgment. That is not a minor implementation detail. It may be the difference between this case becoming a landmark and becoming a citation.

D. Broader Significance

Perhaps the most striking thing about this judgment is what it forces us to sit with. Article 17 has been part of the Constitution since 1950. The abolition of untouchability is not a recent judicial innovation — it is founding constitutional law, written into the text by people who understood exactly what they were doing. The fact that, seventy-four years after ratification, the Supreme Court still needed to instruct State governments that Dalit prisoners cannot be made to clean latrines by virtue of their caste is not a routine doctrinal step forward. It is an indictment. It tells us that the State’s most coercive institutions — the places where the Constitution’s guarantees matter most — were allowed to operate largely outside constitutional scrutiny for the better part of a century. NCRB data on the disproportionate incarceration of SC/ST communities sits in the background of the case, but the significance of the judgment is not statistical. It is a declaration that this impunity is over.

Something else about this case deserves attention: what actually started it. A 2020 article in The Wire, by a journalist without institutional legal backing, became the factual foundation for a Supreme Court judgment that prompted reform across multiple State prison systems. That is not a minor detail. It says something about how constitutional accountability really works — that the Court’s ability to enforce rights often depends on people doing the slower, less visible work of documenting what happens inside State institutions when no one from a legal chamber is looking.

Conclusion

Sukanya Shantha v. Union of India (2024) is doing more than one thing, and the parts are worth separating out. It strikes down caste-based practices embedded in the administrative machinery of prisons across India. It expands Article 21 to include a right to be free from the institutional reproduction of caste. And it brings a more demanding indirect discrimination framework into the equality jurisprudence of Articles 14 and 15. These are genuine contributions. Whether they produce change on the ground is a different question, and one the judgment itself cannot answer. What it does settle is simpler and more important: a State that abolished untouchability in 1950 cannot go on sheltering its prisons from that abolition by calling discrimination ‘habit,’ ‘custom,’ or administrative security. The Constitution’s guarantees of equality and dignity are not aspirations. They apply inside the walls the State itself builds and runs.

References

[1] Sukanya Shantha, ‘From Segregation to Labour, Manu’s Caste Law Governs the Indian Prison System’ (The Wire, 10 December 2020).
[2] Model Prison Manual 2016 (Ministry of Home Affairs, Government of India).
[3] Model Prisons and Correctional Services Act 2023.
[4] Navtej Singh Johar v. Union of India (2018) 10 SCC 1.
[5] Justice K.S. Puttaswamy v. Union of India (2017) 10 SCC 1.
[6] Lt. Col. Nitisha v. Union of India (2021) 15 SCC 533.
[7] E.P. Royappa v. State of Tamil Nadu (1974) 4 SCC 3.
[8] People’s Union for Democratic Rights v. Union of India (1982) 3 SCC 235.
[9] Maneka Gandhi v. Union of India (1978) 1 SCC 248.

Bibliography

Constitution of India 1950, Arts 14, 15, 17, 21, 23.
SCC Online Blog, ‘Landmark Constitutional Law Judgments in 2024 by the Supreme Court of India’ (April 2025) <https://www.scconline.com>.
IACL-AIDC Blog, ‘The Case of Caste-Based Discrimination in Indian Prisons: A New Constitutional Right to Overcome Prejudice?’ (December 2024) <https://blog-iacl-aidc.org>.
Oxford Human Rights Hub, ‘Upholding the Dignity of Prisoners: The Indian Supreme Court’s Stand Against Caste-based Discrimination in Prisons’ (2024) <https://ohrh.law.ox.ac.uk>.