After almost every major episode of violence in India, we witness a familiar cycle. There is shock, followed by protest, television debates, political statements, and an enormous volume of discussion on social media. For a few days, sometimes weeks, the country appears consumed by the question of justice. Then public attention moves elsewhere.

I remember the outrage after the rape and murder of the postgraduate trainee doctor at R.G. Kar Medical College and Hospital in Kolkata in August 2024. The protests were not simply about one crime; they reflected a much wider anxiety about whether an individual could expect safety inside an institution that was itself supposed to provide care. 

Two years later, however, the more consequential question is not whether R.G. Kar itself has become safer. It is whether the lessons of R.G. Kar has travelled beyond R.G. Kar.

The recent alleged gang rape of a 17-year-old at Aastha Kunj Park near the venerated Kalkaji Mandir in Delhi makes that question difficult to avoid. Investigators found that the large Delhi Development Authority (DDA)-managed park had serious security deficiencies, including poor lighting, non-functional or absent CCTV coverage, unmanned gates and damaged boundary infrastructure. The three accused allegedly posed as police personnel before intimidating the teenager and her friend. 

All three were subsequently arrested, according to Delhi Police. The details matter because some of these vulnerabilities were not unknowable. They were characteristics of the physical and administrative environment in which the alleged crime occurred.

Problem of Institutional Memory

This Problem is Not New.

In August 1978, Geeta and Sanjay Chopra, aged 16 and 14, were abducted in Delhi and later murdered. The case generated intense public outrage and scrutiny of the police response, becoming one of the most remembered criminal cases in the capital's modern history. 

Returning to the Chopra case is not an argument that Indian institutions have remained unchanged since 1978. They have not. Criminal law, forensic science, emergency response, public awareness and institutional mechanisms have developed considerably. The 2012 Delhi gang rape, for instance, resulted in the Justice Verma Committee and major changes to criminal law.

But recurrence raises a different question: what survives after the protest ends?

The R.G. Kar case resulted in nationwide demonstrations, Supreme Court intervention and the creation of a National Task Force on the safety and working conditions of healthcare professionals. The Supreme Court subsequently directed that the Task Force's recommendations be made available to states and implemented across healthcare institutions.

That is the institutional test of a tragedy. A crime in one hospital should lead to an examination of comparable vulnerabilities in other hospitals. A crime in one park should lead authorities to inspect other parks with similar characteristics. Otherwise, every tragedy remains an isolated administrative event rather than becoming institutional knowledge.

Violence is larger than the exceptional crime

The scale of registered violence makes this distinction important. NCRB data have consistently recorded hundreds of thousands of cases of crimes against women in India annually, while rape remains one component of a much larger spectrum of sexual, domestic and other forms of violence. 

Registered crime figures, however, should not be confused with actual incidence: reporting, registration practices, access to police and willingness to approach institutions all affect what appears in official statistics.

The data also complicate the popular image of sexual violence as principally an encounter with an unknown stranger in an isolated street. In the analysis of NCRB's 2024 data, CEDA found that in 96.8 per cent of registered rape cases the accused was known to the survivor; cruelty by husbands or relatives accounted for 1,20,227 registered cases, the largest category of crimes against women.

This is why a serious response cannot be reduced to CCTV cameras, streetlights or police vans. Those interventions matter, particularly in public spaces, but they address only one layer of the problem. Violence emerges through different combinations of individual agency, social norms, intimate relationships, opportunity structures, institutional weakness and physical environments. A credible policy framework must be able to hold these explanations together without reducing one to the other.

Agency, dignity and the psychology of violence

There is a tendency to explain extreme violence almost entirely through poverty, inequality or deprivation. Such conditions can shape vulnerability and exposure to crime, but they cannot become deterministic explanations. Millions experience deprivation without committing predatory violence. The opposite explanation is equally inadequate: treating every extreme offender as psychologically abnormal and therefore fundamentally different from the society around them. That can make violence appear inexplicable rather than asking what conditions allow it to occur.

Albert Bandura's concept of moral disengagement provides one useful framework. Individuals can distance themselves from the consequences of harmful conduct through mechanisms such as dehumanisation and displacement of responsibility. This does not remove individual responsibility; it helps explain how ordinary moral restraints can be suspended.

Michel Foucault's analysis of power offers another, though necessarily limited, lens. Violence can involve domination over another person's body and autonomy. Sexual violence, in particular, cannot always be reduced to the language of sexual desire. Yet theoretical explanations should not obscure the basic fact that responsibility rests with the person who chooses to commit the offence.

The State therefore has two distinct responsibilities: to hold the individual responsible for the act and to reduce institutional conditions that make such acts easier to commit or harder to detect.

Fear and Credibility of Law

This brings us to a less discussed dimension of criminal justice: fear.

A criminal law can prescribe severe punishment and still have a weak deterrent effect if the perceived probability of detection and punishment remains uncertain. Deterrence is not produced by the maximum sentence alone. It depends upon the credibility of the entire chain — detection, investigation, prosecution and adjudication.

The concern becomes sharper when citizens perceive that institutions do not operate with the same force for everyone. The perception that people with money, political connections or institutional influence can navigate bureaucratic systems differently from ordinary citizens can weaken confidence in the impartiality of law. That does not mean that every influential person escapes accountability, nor that the legal system is uniformly discriminatory. It means that perceived inequality before institutions itself has consequences for legitimacy.

A citizen who believes that the law is predictable has a different relationship with the State from one who believes that its consequences depend upon who is standing before it. This is why the question should not merely be whether India has made its criminal laws sufficiently stringent. It should also ask whether the State has made their enforcement sufficiently certain, timely and impartial.

The Geography of Impunity

The Kalkaji incident brings this question into the physical geography of the city. A broken streetlight does not cause rape. An absent CCTV camera does not cause rape. An unmanned gate does not cause rape. But such conditions can reduce visibility, delay intervention and lower the perceived probability of detection.

Reporting from Aastha Kunj found poor illumination, non-functional or absent CCTV coverage, damaged sections of the boundary and unmanned gates. The park's large physical size of about 200 acres made these deficiencies particularly consequential. The alleged impersonation of police officers adds another disturbing dimension. The accused reportedly used the symbolic authority of the State to intimidate the teenager and her friend.

This is why impunity is a more useful analytical concept than simply “crime prevention”. Impunity is not necessarily the certainty that one will never be caught. It can be the belief that detection is unlikely, institutional response will be slow or the system can be manipulated. The objective of public administration should therefore be to make that calculation increasingly difficult.

Justice Beyond Principal Offender

The R.G. Kar case demonstrates why justice cannot end with the conviction of an individual offender. Sanjoy Roy was convicted and sentenced to life imprisonment in January 2025. The trial court did not impose the death penalty, applying the legal framework governing the “rarest of rare” cases. 

But the criminal trial was only one dimension of the wider controversy. The Supreme Court's proceedings also addressed allegations concerning the initial handling of the crime, the conduct of hospital authorities and the broader safety of medical professionals. The Court transferred the investigation to the CBI and later constituted the National Task Force.

This distinction is important. Justice requires punishment where guilt is established, but it also requires institutional accountability where institutional wrongdoing is established. If the principal offender is convicted while failures in investigation, evidence preservation or administrative oversight remain insufficiently examined, the public may receive a conviction without receiving a complete account of how the system responded.

The question, therefore, is not simply “Was the perpetrator punished?” It is also “Did the institution learn to prevent the next failure?”

Violence does not end at the boundary of public space

The geography of violence is larger than the dark park or deserted road. If the underlying question is individual safety, the state must also confront violence that occurs inside spaces traditionally regarded as private. The home, the workplace, the hospital and marriage are not automatically spaces of consent simply because they are spaces of intimacy or institutional authority.

Marital rape presents one of the most difficult questions in this regard. Under Section 63 of the Bharatiya Nyaya Sanhita, 2023, the statutory rape provision retains an exception stating that sexual intercourse or sexual acts by a man with his own wife, where the wife is not under eighteen, do not constitute rape.

The legal debate surrounding that exception remains contested. The question is larger than the terminology of one offence. It concerns whether marriage should alter the legal understanding of bodily autonomy. Consent is meaningful precisely because it can be withheld. At the same time, any change in criminal law must confront questions of evidentiary standards, due process, family law and the appropriate scope of criminalisation.

But the underlying principle deserves clarity: intimacy cannot, by itself, extinguish agency. This is also why violence should not be analysed only through the figure of the stranger or predator. Violence can occur within relationships of familiarity and dependence, precisely where institutional intervention is often more complicated.

Social Environment Around Violence

The institutional problem exists alongside a social one. Digital platforms have created a public sphere in which individuals are routinely judged according to gender, appearance, sexuality, relationships and perceived respectability. It would be intellectually careless to speak of an “average user” online as though India possessed a single digital personality.

There is nevertheless empirical evidence of misogynistic discourse on Indian social media. A peer-reviewed study examining 30 million tweets posted from India between 2018 and 2021 estimated that around 2 per cent contained misogynistic content. In a manually coded sample, sexual objectification and sexist abuse were the most common categories.

The point is not that an offensive comment automatically produces a physical crime. Nor should the State criminalise every unpleasant opinion. The significance lies in the social norms being reproduced. After sexual violence, public discussion frequently shifts from the conduct of the perpetrator towards the behaviour of the victim: where she went, what she wore, whom she trusted or why she was outside at a particular hour.  That shift matters because it quietly transfers responsibility. The principle should be simpler: an individual's freedom should not become conditional upon successfully predicting another person's criminality. Reasonable precautions are part of ordinary life. They cannot become a substitute for institutional responsibility.

Safe Urban Living is a Question of Liberty

This is where the discussion needs to move beyond the narrower vocabulary of “women's safety”. The problem is ultimately about individual safety and the practical exercise of citizenship.

A person may formally possess the right to walk through a public park, travel home after dark, attend a late class, visit a hospital, use public transport or participate in public life. But if the practical conditions surrounding those rights require constant calculation of whether the route is illuminated, whether a police patrol is nearby, or whether someone should accompany them, the formal right is being exercised under constraint. 

Gender-based violence is an especially serious manifestation of this problem, but the broader question concerns the ability of individuals to inhabit public and private spaces without organising their lives around fear. A city can possess modern infrastructure and sophisticated technology while still failing to provide the elementary conditions of secure public life. India's unfinished task is not simply to build safer spaces after crime occurs. It is to make safety an ordinary characteristic of those spaces.

Visible and Measurable Policing

One practical response is greater emphasis on visible, hotspot-based policing. The idea is not to place a police van on every street. It is to identify locations and periods where evidence indicates elevated risk and maintain predictable police presence there. 

Delhi Police's 2026 rollout of PCR motorcycles across 173 police stations provides an existing domestic model, with the force also identifying crime-prone hotspots for intensified patrolling. This can be developed further.

Parks, hospitals, universities, transport interchanges and poorly connected residential areas should have time-bound hotspot patrol plans, particularly during periods identified through local crime and emergency-call data. The performance indicator should not be the number of police vehicles deployed. It should be response time, repeat incidents, infrastructure deficiencies corrected and citizen complaints resolved.

The relevant international lesson is similarly not that India should mechanically copy another country's policing system. It is that visible, local and sustained policing can complement investigative capacity when deployment is based on evidence rather than merely on political attention after a high-profile crime.

Reactive Measures to Institutional Standards

The first reform should be institutional safety audits. Every major public institution and high-footfall public space should undergo periodic independent assessments covering lighting, access points, surveillance, emergency communication, staffing, policing and transport connectivity. Every deficiency should have a responsible authority and a deadline for correction.

Second, existing Safe City programmes should be evaluated on functionality rather than installation. A CCTV camera that stops working, a streetlight that remains unrepaired or a police booth that is permanently unmanned should count as an institutional failure, not completed infrastructure.

Third, the R.G. Kar National Task Force recommendations should become a national institutional benchmark rather than a Kolkata-specific reform exercise. The Supreme Court expressly directed that the recommendations on prevention of violence and safe working conditions be implemented across healthcare institutions and states.

Fourth, serious investigations require stronger insulation from institutional conflicts of interest. Crime-scene preservation, forensic collection and documentation should follow enforceable protocols, with clear responsibility for every stage of evidence handling. Where deliberate obstruction, evidence destruction or institutional interference is established through due process, accountability should not end with the frontline offender.

Finally, accountability must extend upwards as well as downwards. If an ordinary offender is punished while established institutional negligence or deliberate interference produces little consequence, the system communicates a deeply problematic lesson: institutional power can become a buffer against accountability. The answer is not arbitrary State power. It is equal institutional treatment.

From Outrage to Institutional Memory

The Chopra siblings' murder in 1978, the Nirbhaya case of 2012, R.G. Kar Kolkata in 2024 and Kalkaji in 2026 belong to different historical contexts. They should not be artificially presented as identical events. But they reveal a recurring difficulty in Indian governance: we are capable of extraordinary mobilisation after extraordinary violence, yet less consistent at converting that mobilisation into permanent institutional practice.

The purpose of remembering R.G. Kar should therefore not simply be to remember the victim. It should be to ask what every government hospital changed because of R.G. Kar. The purpose of remembering Kalkaji should not simply be to prosecute the accused. It should also be to ask how many parks, public institutions and urban spaces possess similar vulnerabilities today. And the purpose of public outrage should not be to demonstrate, once again, that society can be horrified. It should be to make the next tragedy less likely.

A State cannot guarantee the complete absence of violence. It can, however, make violence harder to commit, detection more probable, investigation more credible, and institutional failure more difficult to ignore. That is the meaning of safe urban living. 

It is not a promise that nothing terrible will ever happen. It is the more modest, but constitutionally significant, promise that an individual should be able to exercise ordinary freedom without living in perennial anticipation of violence.

(The author is a Research Fellow at the Society for Policy Studies, New Delhi and Editorial Consultant, South Asia Monitor. The views expressed are personal. He can be contacted at piyushchaudhary2125@gmail.com )