The Prison That Trusts You Back

Steadfirst: Our jails are not full because there is too much crime. They are full because we have never seriously asked what punishment is actually supposed to accomplish.

Thirty-five kilometres from Thiruvananthapuram, in the foothills of the Western Ghats, there is a prison that has produced only one repeat offender in six decades of operation. The men inside wake before dawn, tap rubber trees, tend cattle, earn wages credited to their own accounts, and return home for one month in every six. Their children attend the neighbourhood school. The facility earns two crore rupees a year for the Kerala government. It has no walls. Nettukaltheri has been answering the central question of Indian penology since 1962, and the answer is so clear, and so damning of everything we do elsewhere, that the only honest question remaining is why we have spent sixty years refusing to act on what we already know.

A constitution deferred

The Supreme Court's judgment in Suhas Chakma v. Union of India, delivered on 26 February 2026, was not primarily about prisons. It was about what the State owes a person it has chosen to punish. The numbers framing the judgment are familiar to anyone who has followed India's prison crisis: 120.8 percent national occupancy, Delhi and Uttar Pradesh and Madhya Pradesh exceeding 150 percent, nearly three-quarters of all inmates undertrial, many held longer than the maximum sentence their alleged offence could have carried. These are not statistics about overcrowding. They are statistics about a system that has quietly abandoned its own stated purpose and replaced it with warehousing, and that has done so most aggressively against people who are poor, Dalit, Muslim, or tribal.

The constitutional doctrine the Court drew upon spans half a century, moving in a single direction. Francis Coralie Mullin v. The Administrator, Union Territory of Delhi & Ors. read Article 21 to include the right to live with human dignity even while incarcerated. Earlier decisions had established that fundamental rights survive conviction, that rehabilitation is State obligation rather than State generosity, and that meaningful work, wages and family contact are goals the sentence must actively pursue. These were aspirations. What Suhas Chakma judgement did was convert them into obligation. The State's duty is not discharged by confinement alone. It is discharged only when confinement is oriented toward returning a person capable of living freely. A State that delivers people back more damaged than when it received them is not administering justice. It is, in the Court's own phrasing, running correctional houses that cruelly ache the soul.

What the evidence has suggested

Michel Foucault's argument in Discipline and Punish was not simply that prisons fail at rehabilitation. It was that they succeeded at something else entirely: producing institutionally dependent, socially stigmatised and economically unplaceable people, which are precisely the conditions that make return to prison almost inevitable. Applied to India, the insight is devastating. It explains not an aberration, but the design logic of the system we continue to defend. A person who enters a closed prison loses her family contact, her income, her vocational skills, and eventually her sense of independent agency. She exits with a criminal record and into a labour market that will not receive her. We have, over decades of institutional inertia, built a system that manufactures the very failures it claims to punish, and then treats their return as surprise.

Alexander Paterson put the governing principle in a sentence no Indian prison administrator has successfully refuted: a person is sent to prison as punishment, not for punishment. Loss of liberty is the penalty. Everything the State inflicts beyond that is either rehabilitation or gratuitous harm. The undertrial crisis makes this indictment sharper still. Three-quarters of our prison population have not been convicted of anything. They are inside because the bail system failed them, and it failed them most reliably because they were poor. For them, the closed prison is not even punishment in the legal sense. It is harm administered in advance of any finding of guilt.

Numbers which should embarrass everyone

The Sanganer open-air camp in Rajasthan costs the State Rs 49.60 per prisoner per day. A closed prison costs Rs 333.12. One seventh of the expenditure, with recidivism rates that are negligible by comparison. That number deserves a moment of stillness before the argument moves on, because it means we have been spending seven times more money, for decades, to produce worse outcomes, and calling the result a correctional system. It should also provoke embarrassment.

Prisoners at Sanganer do not escape despite the absence of walls, because escape means surrendering the earned liberty they have built for themselves, which proves a more powerful deterrent than any physical barrier. What keeps people from reoffending is not the severity of the cage but the value of what lies outside it. That is not a reformer's sentiment. It is sixty years of operational data from facilities the State itself runs, sitting in government reports the State itself commissioned, waiting for a political will that has not arrived.

Machinery absorbing every instruction

The Suhas Chakma directions are specific: mandatory Open Correctional Institution expansion across all states and union territories, inclusion of women prisoners currently excluded from almost every existing facility, and eligibility redesign focused on reform potential rather than time served. That last directive matters most. Most current frameworks identify OCI candidates only after the most damaging years of the sentence have already passed. A person transferred to an open setting in the final months has experienced the full weight of institutional harm with only a residual benefit. Redesigning eligibility is not administrative adjustment. It is a structural acknowledgment that the sentence should be designed around what it is trying to accomplish.

India has had reform committees since 1957, model manuals since 2016, a Model Prisons and Correction Act since 2023, and a sequence of Supreme Court directions that states have treated as optional reading. Thirty-five states and union territories still had no open correctional infrastructure when this judgment arrived. This is not a failure of legislation. It is a failure of political will sustained by identifiable incentives. Prisoners cannot vote. Their families carry stigma. The bureaucratic instinct always favours the defensible custody of high walls over the accountable openness of a Sanganer, because when something goes wrong behind walls, the institution absorbs the blame. Every successful reintegration is a data point that appears nowhere in any minister's political calculus, and the federal structure ensures the Union can issue frameworks but cannot compel the states to honour them.

The Suhas Chakma judgment has given constitutional obligation a deadline, a committee, and the Court's explicit refusal to defer further. Whether it survives contact with the machinery that has absorbed every previous instruction is the only question that remains, and the answer will say something precise and uncomfortable not only about the State, but about what we, as a society, are willing to accept in the name of justice.

The author is an Assistant Manager at Crashfree India and an alumnus of the University of Delhi and NALSAR University of Law.

The views expressed here are personal.