The phrase “encounter killing” suggests a chaotic, split-second firefight: police meet armed suspects, shots are exchanged, and someone dies. That is the official story. This piece attempts to argue that in a disturbing number of documented cases, the reality is closer to a staged execution. A person already in custody—or quietly abducted—is taken to a lonely spot, shot, and then turned into an “armed aggressor” after the fact with planted weapons, invented recovery memos, and a neatly standardized police narrative. The piece calls this “fake encounter” policing, and it is not uniquely Indian. It is a recurring state tactic used by regimes that want the appearance of lawful force without the inconvenience of trials.
In Indian administrative and journalistic language, an “encounter” is supposed to mean a sudden armed confrontation in which police fire in self-defence. In forensic reality, the article argues, many encounters are staged. The suspect is already secured or secretly picked up, transported to a desolate place, shot at close or middle range, and then retroactively made to look like a gunman. Evidence is planted. A “recovery memo” appears. The official story is that the suspect fired first. The scene is managed.
This is not an Indian invention. A global pattern can be traced. Colonial regimes used martial-law ordinances and “hot pursuit” to justify punitive killings. Nazi Germany and the Soviet Union used categories like auf der Flucht erschossen—“shot while attempting to escape”—to liquidate people without judicial review. Latin American military dictatorships used the ley de fuga, the “law of flight”: release a prisoner, shoot them from behind, and claim they were fleeing. In the Philippines’ drug war, police used nanlaban—“they fought back”—as a standard label for killings. In India, the same basic script has appeared in Punjab, Mumbai, Andhra Pradesh, Uttar Pradesh, Telangana, and Assam. The common thread is simple: the state wants the result of an execution without the paperwork, publicity, and legal risk of a trial. In democracies, the motive is often different from dictatorship. It is not usually a secret police hit list. It is a mix of public impatience with slow courts, low conviction rates, political demands for “tough” policing, and an institutional belief that due process is a luxury that lets criminals escape. Once that belief takes hold, extrajudicial violence gets rewarded—promotions, medals, political praise—and the line between policing and punishment disappears.
Legal history matters here because the “encounter” defence often rests on the idea that the state can kill first and explain later. The history says otherwise. The Code of Hammurabi (around 1753 BC to 1754 BC) is usually remembered for “an eye for an eye,” but something more important is clear: it set evidentiary and procedural rules before the state could impose death. The opening laws punished people who brought false capital charges or gave perjured testimony in capital cases. The state, not private vengeance, decided guilt—and only after witnesses and evidence were presented. Magna Carta (1215), Clause 39, later said no free person could be arrested, imprisoned, stripped of rights, or ruined except by the lawful judgment of equals or by the law of the land—per legem terrae. The 1354 statute (1354) turned that phrase into “due process of law.” Locke (1632–1704) and Beccaria (1738–1794) expanded the idea: life is not a privilege the executive can revoke at will; a government that kills without established law puts itself at war with its own people. After World War II, these ideas became international law. The Universal Declaration of Human Rights (1948) says everyone has a right to life, liberty, and security of person, and protects against arbitrary arrest and unfair trials. The International Covenant on Civil and Political Rights (1966) goes further: Article 6 protects the right to life, and Article 4(2) makes that right non-derogable—meaning even in a national emergency, the state cannot suspend it or permit arbitrary killings. The UN Human Rights Committee’s General Comment No. 36 (2018) says a killing is arbitrary if it fails to comply with law, lacks necessity and proportionality, or bypasses fair-trial guarantees. Intentional lethal force is allowed only as an extreme last resort against an imminent threat of death or serious injury. Then there are the operational rules. The UN’s Havana Principles on the Use of Force and Firearms (1990) say police should use non-violent means first, use force only when unavoidable, act proportionally, minimise harm, and preserve life. Firearms should not be used except in self-defence or defence of others against an imminent threat of death or serious injury, to prevent a particularly serious crime involving grave threat to life, to arrest a dangerous person resisting authority, or to prevent escape—and only when less extreme means are insufficient. Even then, intentional lethal force is allowed only when strictly unavoidable to protect life. The Minnesota Protocol (1991; revised 2016) adds the investigative side. When state agents are involved in a death, the state must conduct an investigation that is prompt, effective, thorough, independent, impartial, and transparent. It must secure the scene, preserve chain of custody, conduct proper autopsies, examine ballistics, identify command responsibility, and protect witnesses. In plain English: the state cannot shoot someone and then control the only evidence that would explain what happened.
Other courts have handled police killings, and the point is that modern constitutional law has increasingly rejected the old idea that a fleeing suspect can simply be shot. American common law once allowed police to use deadly force to stop a fleeing felony suspect. The old logic was grim: most felonies were capital crimes, so killing an escaping felon just sped up the inevitable. That changed in Tennessee v. Garner (471 U.S. 1 (1985)) in 1985. Memphis police shot and killed Edward Garner, a slight fifteen-year-old boy, as he tried to climb a fence after a nighttime burglary in which he had stolen ten dollars. The shooting was authorised by a Tennessee statute allowing “all necessary means” to effect an arrest. The Supreme Court struck that down. Justice Byron White said using deadly force to prevent the escape of all felony suspects, whatever the circumstances, is constitutionally unreasonable. “It is not better that all felony suspects die than that they escape,” he wrote. Deadly force is allowed only where the officer has probable cause to believe the suspect poses a threat of serious physical harm to the officer or others. Graham v. Connor (490 U.S. 386 (1989)) in 1989 added the “objective reasonableness” test: courts look at the severity of the crime, whether the suspect poses an immediate threat, and whether the suspect is actively resisting or fleeing. Being a “fleeing suspect” alone gives police no licence to kill.
In the United Kingdom, Section 3 of the Criminal Law Act 1967 allows reasonable force to prevent crime or effect a lawful arrest. But Article 2 of the European Convention on Human Rights permits deprivation of life only where force is “no more than absolutely necessary”—in defence against unlawful violence, to effect a lawful arrest or prevent escape of a lawfully detained person, or to quell a riot or insurrection. The key case is McCann and Others v. United Kingdom ((1995) 21 EHRR 97) in 1995, arising from Operation Flavius in Gibraltar. SAS soldiers shot and killed three unarmed IRA members suspected of carrying a remote detonator for a car bomb. The European Court of Human Rights said “absolute necessity” is stricter than ordinary reasonableness. Crucially, the court did not just look at the soldiers pulling the trigger. It looked at the whole operation. Senior commanders had failed to arrest the suspects when they entered Gibraltar, had misinformed the soldiers that a bomb was active, and had designed the mission in a way that made lethal shooting almost inevitable. The UK was found to have violated Article 2. In plain English: a state cannot plan a lethal outcome and then claim self-defence when the plan produces the killing. South Africa’s Constitutional Court made a similar move in Ex parte Minister of Safety and Security: In re S v Walters (2002 (4) SA 613 (CC)) in 2002. It struck down a law that allowed police to kill an escaping suspect if arrest could not be effected by other means. Justice Johann Kriegler, writing for a unanimous court, said the purpose of arrest is to bring a suspect before a court—not to inflict summary punishment. Force is allowed only when there are no reasonable means of arrest without it; the suspect must pose an immediate threat of serious bodily harm, or have committed a crime involving serious bodily harm; and lethal force is unconstitutional if the suspect can be apprehended later without exposing others to imminent danger. The Walters (2002 (4) SA 613 (CC)) ruling made clear that in modern Commonwealth constitutionalism, an escaping suspect does not lose the right to life.
In India, encounter policing moved from an emergency tool in border and insurgency areas into a normalised, sometimes openly celebrated, policing method. The modern template came from Punjab during the late 1980s and early 1990s counter-insurgency against Khalistani militants. Under Director General of Police K.P.S. Gill, police and central forces adopted a policy of targeted “neutralisation.” Courts were crippled by militant intimidation; judges, witnesses, and prosecutors were being assassinated. The security apparatus abandoned trials. Suspects were abducted, killed in simulated encounters, and secretly cremated as “unidentified” bodies. Civil rights defender Jaswant Singh Khalra documented this and was himself abducted and murdered by police in 1995. The Punjab model—secrecy, denial, bureaucratic impunity, and financial bonuses for “eliminated” terrorists—became a kind of operational manual. By the mid-1990s, the encounter moved to urban policing in Mumbai. Faced with organised crime syndicates like D-Company and the Chhota Rajan gang, Mumbai Police created specialised crime-intelligence squads. “Encounter specialists” became media celebrities. The script was remarkably consistent: police act on “confidential intelligence,” intercept gangsters in isolated suburban spots, the criminals allegedly fire country-made pistols, police return fire, suspects die, and no police officer suffers fatal wounds. Between 1993 and 2003, hundreds of alleged gang members were killed this way, creating an extralegal patronage system between police factions, real-estate cartels, and underworld figures. In undivided Andhra Pradesh, the state created the Greyhounds commando force to fight the People’s War Group and Maoist insurgencies. In the forested belts of Telangana and North Coastal Andhra, “combing operations” often ended in encounter deaths. Civil society groups, especially the Andhra Pradesh Civil Liberties Committee led by K. Balagopal, documented cases where captured Naxalite leaders were tortured in secret safehouses, driven into forests, shot, and photographed beside automatic weapons to make it look like an armed clash.
In recent years, the encounter doctrine became an explicit instrument of political governance. In Uttar Pradesh after 2017, the state adopted an aggressive law-and-order posture. Police conducted thousands of armed raids against alleged gangsters, repeat offenders, and illicit cattle transporters. “Operation Langra”—Operation Lame—became an administrative practice of shooting suspects below the knees, deliberately maiming them. The official rationale: the suspect tried to escape or fired during evidence recovery, so police had to fire back to incapacitate. When encounters turned fatal, police press releases followed the old templates. The most prominent contemporary urban case was in Telangana in December 2019. Cyberabad Police, led by Commissioner V.C. Sajjanar, shot dead four young men accused of the rape and murder of a veterinarian—the “Disha” case. The killings at Chatanpally triggered public celebrations, including flower petals showered on police. But the judicial investigation dismantled the official story. The Supreme Court appointed a three-member Commission of Inquiry headed by retired Justice V.S. Sirpurkar. The Commission’s 2022 report found that the “firearm snatching” story was fabricated: the suspects, two of them juveniles, were untrained and could not have operated the safety catches or fired the weapons. Safehouse records were fabricated. The alleged recovery of the victim’s mobile phone was staged. The blunt-force injuries to police were fabricated or self-inflicted. The bullet trajectories showed the accused were deliberately shot at close range while immobilised. The Commission concluded it was a calculated extrajudicial execution and recommended that all ten involved officers be tried for murder under Section 302 of the Indian Penal Code.
Assam has two distinct eras. Between 1998 and 2001, under the Asom Gana Parishad government led by Chief Minister Prafulla Kumar Mahanta, Assam experienced a systemic campaign of extrajudicial assassinations known as the Gupto Hoitya—“Secret Killings.” The target was the armed insurgency of the United Liberation Front of Asom, or ULFA. After AFSPA was imposed and a Unified Command Structure was set up in 1997—bringing the Army, central paramilitary forces, and Assam Police under the Army’s IV Corps commander—the security apparatus adopted an asymmetric strategy. Unable to easily find core ULFA insurgents operating from sanctuaries in Bhutan and Myanmar, state agencies used surrendered ULFA cadres—known as SULFA—as execution squads. The division of labour was unacknowledged but functional. Strategic directives and logistics flowed from the Unified Command Structure to district police. Local police stations withdrew night patrols from targeted areas. Covert strike squads of masked SULFA operatives carried out abductions and home invasions. The method was consistent: armed squads in civilian clothes, faces covered with black cloth, arrived late at night in unmarked vehicles without number plates, speaking local Assamese to gain entry. Victims were shot on their doorsteps or abducted and dumped on highways and in forests, hands bound, with torture marks and point-blank gunshot wounds to the head and neck. Police patrols were withdrawn before the attacks and redeployed afterwards to ensure safe passage. Victims included unarmed parents, siblings, and relatives of ULFA commanders, as well as human rights activists and journalists. Parag Kumar Das, executive editor of Asomiya Pratidin, was assassinated in broad daylight in Guwahati in May 1996. After public outrage, the Justice K.N. Saikia Commission of Inquiry was set up in August 2005. It reviewed 35 representative cases and submitted four reports in late 2007. It concluded that the Secret Killings were state-sponsored executions carried out by SULFA cadres under the protection and logistical guidance of Assam Police and the security establishment. It explicitly indicted former Chief Minister Mahanta, who held the Home portfolio, finding an institutional nexus between the political executive, police leadership, and surrendered militants. The Gauhati High Court later quashed the Commission’s report on technical procedural grounds in September 2018, but the evidentiary record remained.
A sharp escalation began after May 2021, when Himanta Biswa Sarma became Chief Minister. Unlike the covert 1990s killings, the post-2021 wave was overt, officially publicised, and aimed mainly at ordinary criminal suspects. In July 2021, at a conference with police station officers, Sarma publicly said that shooting at criminals trying to escape custody or attack police “should be the pattern,” emphasising that if an accused flees or snatches a weapon, the law permits police to fire. A standardised sequence emerges. The suspect is in lawful custody, often remanded to police custody by a magistrate, then transported late at night or before dawn under the pretext of “crime scene reconstruction,” “recovering stolen property,” or “locating an arms cache.” The convoy stops in a desolate rural area or along an unlit highway because the handcuffed suspect allegedly needs to relieve himself. Upon exiting the vehicle, the suspect—often physically restrained—allegedly assaults an officer, throws soil, and tries to snatch a loaded service weapon. Police claim self-defence and fire, often aiming at the lower extremities—knees, thighs, calves—to maim rather than kill. But in dozens of cases, bullets hit the torso or head and the suspect dies. A prominent example is a Sivasagar case involving Ghanakanta Gogoi, prime accused in a commercial shooting at Hotel Bilas Inn in Guwahati that killed five people. He was arrested by Guwahati Police, remanded to judicial custody, then transferred to Sivasagar police custody in a separate bank fraud case. While in transit near Gaurisagar, the police vehicle halted after Gogoi asked to relieve himself. Senior Superintendent of Police Papori Chetia reported that Gogoi tried to snatch an investigating officer’s service weapon and attacked the escort team; police fired in self-defence, striking Gogoi, who was taken to Sivasagar Civil Hospital and pronounced dead. The case shows how the same script appears in both minor criminal cases and high-profile homicides.
According to official data presented by the Union Ministry of Home Affairs before Parliament, Assam recorded 18 police encounter deaths in 2021–22 alone—the third-highest in India, behind Jammu and Kashmir with 45 and Chhattisgarh with 30, and ahead of Uttar Pradesh with 11. Between May 2021 and August 2022, there were 171 documented police firing incidents, 56 people killed, and 139 suspects injured, mostly with lower-limb bullet injuries. Alleged offences included cattle theft, illicit narcotics peddling, armed dacoity, rape, and vehicle theft—showing an expansion beyond counter-insurgency into ordinary crimes. A demographic analysis of 30 sampled cases in a PIL found 14 Muslims and 10 indigenous tribals among victims, indicating disproportionate impact on religious minorities and tribal communities. The state defended the operations by citing deterrence. It claimed Assam’s overall crime rate per 100,000 people fell by 52%, from 387.5 in 2019 to 183.1 in 2023. Criminologists generally treat such a drop as likely reflecting changes in case registration, administrative diversion, and suppression of First Information Reports—rather than proof that summary police violence caused less crime.
The post-2021 wave faced legal resistance from Delhi-based advocate and human rights activist Arif Md Yeasin Jawader. In December 2021, he filed a Public Interest Litigation before the Gauhati High Court, citing investigative journalism and media records documenting over 80 custodial shootings in seven months. He alleged that Assam Police were conducting premeditated executions under the guise of fake encounters. He sought FIRs for murder under Section 302 IPC against all police personnel involved, an independent judicial inquiry by a retired judge or a CBI or Special Investigation Team investigation, strict enforcement of the Supreme Court’s 16-point guidelines in PUCL v. State of Maharashtra ((2014) 10 SCC 635) from 2014, and compensation for families of the dead and injured survivors. The Assam government contested the PIL, filing an affidavit saying all police actions were in the discharge of official duty, that accused persons had attacked officers or tried to flee, and that all 171 incidents had been subject to mandatory magisterial review. The state said no fake encounters had occurred. In early 2023, a Division Bench of the Gauhati High Court disposed of the PIL, declining to order a CBI or SIT investigation because it thought judicial intervention was premature and statutory remedies—such as magistrate-led inquiries—were already available. Jawader appealed to the Supreme Court. In 2024 hearings, a bench headed by Justice Surya Kant and Justice K.V. Viswanathan said 171 armed encounter incidents in a short time was “alarming” and “very, very serious,” and ruled that such petitions could not be dismissed as premature. The bench observed that the standardized police scripts did not align with the constitutional rule of law. On May 28, 2025, the Supreme Court disposed of the appeal by directing the Assam Human Rights Commission to conduct an independent, time-bound investigation into every encounter death and shooting since May 2021. If the AHRC found prima facie offences, it could order FIRs and recommend criminal prosecution. But human rights groups noted a structural problem: the AHRC largely relied on local police officers to gather field evidence against their own colleagues.
Encounter killings have no formal legal status in India. An “encounter” is not a legal category. An encounter death is, on its face, a homicide by a public servant. It is illegal unless the officers can prove a recognised legal defence in court. Article 21 says no person shall be deprived of life or personal liberty except according to procedure established by law. After Maneka Gandhi v. Union of India ((1978) 1 SCC 248) in 1978, “procedure established by law” does not mean any procedure the legislature happens to pass. It must be right, just, and fair—not arbitrary, oppressive, or fanciful. Article 21 protects citizens, foreigners, and suspected criminals alike. The right to life does not shrink because someone is accused of a heinous crime. An extrajudicial execution bypasses the entire judicial machinery and violates the separation of powers. Article 14 guarantees equality before the law and equal protection of the laws. In E.P. Royappa v. State of Tamil Nadu ((1974) 4 SCC 3) in 1974 and in Maneka Gandhi ((1978) 1 SCC 248), the Supreme Court developed the anti-arbitrariness doctrine: arbitrariness and the rule of law are sworn enemies. State action that lets executive agents execute people based on unproven allegations is inherently arbitrary.
Police officers usually invoke the right of private defence, codified in Sections 96 to 106 of the Indian Penal Code, 1860, and now mirrored in Sections 34 to 44 of the Bharatiya Nyaya Sanhita, 2023. The key limits are these. Section 99 IPC, now Section 37 BNS, says the right of private defence never extends to inflicting more harm than necessary for defence. Section 100 IPC, now Section 38 BNS, says causing death in defence of the body is allowed only in specific circumstances—primarily where there is a reasonable apprehension that death or grievous hurt will otherwise result. Section 105 of the Indian Evidence Act, 1872, now Section 108 of the Bharatiya Sakshya Adhiniyam, 2023, creates a strict statutory presumption against the existence of circumstances bringing an act within a general exception. In plain English: the court presumes the defence does not exist unless the police officer proves it. The burden is on the officer. Self-defence cannot be finally validated during a police investigation or a magisterial inquiry to prevent an FIR. It is a defence for a trial judge to evaluate on sworn evidence. Section 46(2) CrPC—now Section 43(2) BNSS—says that if a person forcibly resists arrest or tries to evade it, the police officer may use “all means necessary to effect the arrest.” But Sub-section (3) expressly says nothing in the section gives a right to cause the death of a person who is not accused of an offence punishable with death or imprisonment for life. Police often build post-encounter narratives around this exception: they allege the suspect was involved in a capital offence—armed robbery under Section 396 IPC, murder under Section 302, large-scale narcotics—and was actively resisting. But Section 46(3), now Section 43(3) BNSS, is an exclusionary clause, not an authorising grant. It does not give police a statutory power to kill. It only says death can never be caused if the offence carries lesser penalties. Even when the suspect is accused of a capital offence, the officer’s force must still satisfy Article 21’s requirements of necessity and proportionality.
Two provisions help staged encounters persist. Section 197 CrPC, now Section 218 BNSS, says no court can take cognizance of an offence alleged against a public servant acting or purporting to act in the discharge of official duty, except with prior sanction from the appropriate government. In practice, state executives use this as an administrative shield. When magistrates or private complainants try to start murder prosecutions against encounter squads, the state denies sanction, saying the officers were acting within law-enforcement duties. The Supreme Court has clarified that extrajudicial killings and torture can never be considered acts in the discharge of official duty, but trial courts frequently dismiss private complaints because sanction is missing. Section 6 of AFSPA, 1958, adds another layer in conflict zones. It says no prosecution, suit, or legal proceeding can be instituted against any person for anything done or purported to be done under the Act, except with the previous sanction of the Central Government. Over six decades of AFSPA in Assam, Nagaland, and Manipur, the Central Government has granted sanction in only a negligible fraction of cases—creating de facto legal immunity for military and paramilitary personnel.
In People’s Union for Civil Liberties v. State of Maharashtra ((2014) 10 SCC 635) in 2014, a two-judge bench of Chief Justice R.M. Lodha and Justice Rohinton F. Nariman laid down 16 mandatory directives for encounter deaths. In plain terms, any tip-off about criminal movements must be written in the case diary or electronically recorded before an operation. If firearms are used and someone dies, an FIR must be registered immediately, including under Section 302 IPC for murder. The investigation must be done by the CID or a police station other than the one involved, headed by an officer senior to the encounter team leader. A magisterial inquiry under Section 176 CrPC, now Section 196 BNSS, must be conducted promptly, with a report to the Judicial Magistrate. Fingerprints, weapons, and shell casings must be recovered, sealed, and sent for chemical and ballistic analysis without delay. Post-mortems must be done by two doctors in a district hospital, with videography. All encounter deaths must be reported to the National Human Rights Commission or State Human Rights Commission immediately. No out-of-turn promotion or instant gallantry award can be given until the officers are exonerated after full judicial inquiry. If the investigation shows unlawful force, criminal proceedings under IPC or BNS must be initiated without waiting for administrative sanction. The idea was a chain: written intelligence before deployment, an independent FIR under Section 302 after any fatal shooting, a CID investigation and magisterial inquiry, ballistic and medical evidence preservation, NHRC reporting, and no rewards before exoneration.
The Supreme Court reinforced this in Extra Judicial Execution Victim Families Association v. Union of India ((2016) 14 SCC 536) in 2016, a case documenting 1,528 alleged encounter killings in Manipur by Assam Rifles and Manipur Police. A bench of Justice Madan B. Lokur and Justice U.U. Lalit rejected the Union Government’s argument that internal conflict authorised security forces to kill without judicial accountability. The Court affirmed findings of the Justice N. Santosh Hegde Commission, which had found that in every sampled case, the alleged encounter was a staged execution of an unarmed detainee. Justice Lokur’s words are worth quoting in plain English: “It does not matter whether the victim was a common person or a militant or a terrorist, nor does it matter whether the aggressor was a common person or the State. The law is the same for both and is equally applicable to both… There is no concept of absolute immunity for our armed forces acting within the country. An internal disturbance does not equate to an armed war granting a license to kill.” In Om Prakash v. State of Jharkhand ((2012) 12 SCC 72) in 2012 and Rohtash Kumar v. State of Haryana ((2013) 14 SCC 290) in 2013, the Supreme Court held that police cannot kill suspects merely because they are dreaded criminals. Extrajudicial executions are cold-blooded murders, and state officers who carry them out are liable as common criminals.
So why do encounters continue? Several mechanisms of evasion stand out. In most encounter cases, police do not register an FIR for murder against the shooting officers, as PUCL ((2014) 10 SCC 635) requires. Instead, they register an FIR against the deceased suspect under Section 307 IPC or BNS for attempt to murder and Section 353 for assault on a public servant. The dead suspect becomes the sole accused; the police become the victims. When the suspect dies, the case is closed as “abated due to death.” The legality of the police shooting is never examined. Section 176(1) CrPC allows inquiries by Executive Magistrates—revenue and administrative officials under the District Magistrate and state home department—rather than Judicial Magistrates. These officials depend on local police for daily administration, election logistics, and executive operations. So many magisterial inquiry reports simply transcribe the police version, clear the officers, and never interview independent witnesses or scrutinise ballistic angles or call records. The CID or external police station teams required by PUCL ((2014) 10 SCC 635) are often staffed by officers from the same district cadre, sharing command relationships, social ties, and institutional loyalties with the people under investigation. That undermines the Minnesota Protocol’s independence requirement. State governments often delay filing encounter reports with the NHRC and SHRCs for months. Under Section 19 of the Protection of Human Rights Act, 1993, the NHRC lacks investigative jurisdiction over central military and paramilitary forces and can only request factual reports from the Central Government. So AFSPA deployments remain largely outside human rights commission oversight.
Three big drivers stand out. Indian trial courts have a backlog exceeding 40 million matters. Serious violent-crime cases can run one to two decades through trial and appeal. Witness intimidation, weak forensic collection, under-resourced prosecution offices, and the absence of a comprehensive witness protection framework produce low conviction rates. Faced with public frustration, the political executive treats encounter policing as a way to deliver immediate, visible, retributive outcomes. In contemporary campaigns, physical coercion and summary justice are often celebrated as evidence of decisiveness. When senior political figures publicly tell police to shoot escaping criminals, field officers hear operational immunity. An officer with an aggressive reputation is often protected from disciplinary oversight, given preferential postings, and rewarded through formal and informal career incentives. And encounter shootings fall disproportionately on landless agricultural workers, ethnic and linguistic minorities, undocumented migrants, and historical minority communities. These groups have limited legal representation, little structural leverage in local administration, and restricted access to media—so the state can stage encounter narratives without sustained political fallout.
The conclusion is blunt. The evolution of encounter killings—from Assam’s Gupto Hoitya and Mumbai’s gangland liquidations to the post-2021 custodial shooting wave—is a serious challenge to Indian constitutionalism. The practice flips the presumption of innocence. It strips citizens of Article 21 due process. And it concentrates the roles of investigator, prosecutor, judge, and executioner in the same local police squad. The law on the books is clear; the problem is the gap between the law and what actually happens on the ground.
