LLM Group 5 Criminal Law and Criminal Administration Collective Violence and Criminal Justice System 2025-26 - Set 2 Question Paper with Solutions

Force in Indian criminal law

Section 128 of the Bharatiya Nyaya Sanhita, 2023 defines force without any element of wrongfulness at all. A person uses force to another if he causes motion, change of motion or cessation of motion to that other, or if he causes to any substance such motion or change or cessation of motion as brings that substance into contact with any part of that other's body or with anything he is wearing or carrying, or in such a manner as to affect his sense of feeling. The definition is entirely physical and morally neutral: a doctor examining a patient and a nurse lifting a child both use force.

Wrongfulness enters at the next step. Section 129 defines criminal force as the intentional use of force to a person without his consent, in order to commit an offence, or with the intention or knowledge that the force will cause injury, fear or annoyance. Section 130 defines assault as making a gesture or preparation intending or knowing it likely that the other will apprehend the use of criminal force. The architecture is therefore a ladder: force is neutral, criminal force adds absence of consent and a culpable purpose, and assault punishes the apprehension without the contact.

Chikham Ammiraju v. Chikham Seshamma, ILR (1918) 41 Mad 33 is the classic illustration and remains a favourite with examiners. A husband threatened to commit suicide unless his wife and son released certain properties in favour of his brother. The Madras High Court, by majority, held that although suicide itself was beyond punishment, an attempt to commit suicide was then an offence under section 309 of the Indian Penal Code, so a threat to commit suicide was a threat to commit an act forbidden by the Code, and the release deed was obtained by coercion and was voidable. The case is a lesson in reading a definition literally, and it shows that coercion looks to the effect on the will rather than to physical contact.

The criminal law equivalent is not called coercion at all. It is extortion, where property is delivered under fear of injury intentionally put, and criminal intimidation under section 351 of the Bharatiya Nyaya Sanhita, where a threat of injury to person, reputation or property is made with intent to cause alarm or to compel an act. Coercion and criminal intimidation therefore overlap on the same facts but produce different consequences, one civil and one penal.

Institutionalised violence is narrower and more legal, and it is the one an Indian court can act upon directly. It means violence inflicted by the formal institutions of the state and their agents in the course of their functions: custodial torture and custodial death, encounter killings, prison conditions, the punitive demolition of homes.

Why the distinction is not merely academic

The three concepts have three different legal fates, and saying so is what turns a description into an argument.

Structural violence is not justiciable as such. It is addressed by Part IV of the Constitution, particularly Articles 38, 39 and 46, which are expressly non-enforceable by courts under Article 37. That is why a petition complaining of caste-correlated poverty must be reframed as a claim under Article 21 before a court will hear it.

Sukanya Shantha v. Union of India, 2024 INSC 753 shows all three concepts operating at once and is therefore the ideal closing illustration. State prison manuals assigned sweeping and scavenging to prisoners of particular castes, classified denotified tribes by a supposed "natural tendency to escape", and segregated barracks. The segregation was institutionalised violence because officials imposed it; the classification was symbolic violence because it was carried in categories treated as natural; and the resulting distribution of degrading work was structural violence. The Court struck the provisions down as violating Articles 14, 15, 17, 21 and 23 and directed that the caste column be deleted from prison registers.

Conclusion. Force, coercion and violence are three different legal ideas: force is defined neutrally by section 128 of the Bharatiya Nyaya Sanhita and becomes criminal only when section 129 adds absence of consent and a culpable purpose; coercion is a contractual vitiating factor under section 15 of the Contract Act whose sanction is voidability, as Chikham Ammiraju shows; and violence is deliberately left undefined by the penal law because it must be wide enough to cover the destruction that rioting under section 191 involves. The sociological trio adds what the legal vocabulary omits.

For full marks, cover: three distinct demands in one question, and the examiner will award marks limb by limb, so signpost them. Deal with the approaches briefly, spend the largest share on religiously sanctioned structural violence because that is where the criticism and the case law sit, and then give the five traditions accurately and without flattery, noting that each of them contains a doctrine of legitimate force as well as a doctrine of non-injury. The unifying argument, which is what lifts the answer, is that the same religious traditions supply both the sanction for structural violence and the vocabulary used to attack it.

The approaches to the study of violence in India

Four approaches have dominated, and they ask different questions.

A serious answer uses all four. The Telangana insurrection is unintelligible without political economy; untouchability is unintelligible without the structural approach; and a prosecution requires the legal one.

Religiously sanctioned structural violence: caste

The proposition is not that religion causes violence but that a religious sanction converts a hierarchy into an order that its victims accept, which is precisely Bourdieu's symbolic violence. The classical texts supply that sanction. The varna scheme, the graded assignment of occupation by birth, the doctrine of pollution attaching to particular work and particular persons, and the punitive provisions of the Manusmriti, which prescribe punishments graded by the caste of the offender and of the victim, constitute a legal order in which unequal treatment is the norm and not the exception.

Ambedkar's analysis remains the sharpest available and should be used. Caste, he argued, is not merely division of labour but division of labourers, arranged in a graded hierarchy in which each stratum has someone below it and therefore an interest in the system's survival, which is why it cannot be dislodged by appeals to solidarity from below. Untouchability is its enforcement mechanism.

The statutory implementation is the Protection of Civil Rights Act, 1955, renamed from the Untouchability (Offences) Act in 1976, and the Scheduled Castes and the Scheduled Tribes (Prevention of Atrocities) Act, 1989, whose section 3(1) punishes precisely the religiously coded humiliations: dedicating a woman as a devadasi under clause (k), compelling manual scavenging under clause (j), compelling the carrying of carcasses or the digging of graves under clause (i), garlanding with footwear or parading naked under clause (d), and forcible tonsuring under clause (e).

State of Karnataka v. Appa Balu Ingale, 1995 Supp (4) SCC 469 is the judicial statement of the point. Dalits were prevented at gunpoint from drawing water from a newly dug borewell. Restoring the conviction under the 1955 Act, Ramaswamy J treated untouchability as a continuing social institution rather than an isolated act of obstruction, and read Article 17 as imposing a positive obligation on the state. Sukanya Shantha v. Union of India, 2024 INSC 753 shows the sanction surviving inside the state's own institutions: prison manuals in several States assigned scavenging and sweeping by caste, segregated barracks, and classified denotified tribes by supposed hereditary traits, and the Court struck them down under Articles 14, 15, 17, 21 and 23 as recently as October 2024.

Indian Young Lawyers Association v. State of Kerala, (2019) 11 SCC 1, the Sabarimala case, is the more direct authority on religiously sanctioned exclusion. The exclusion of women between the ages of ten and fifty from the temple was defended as an essential religious practice of a distinct denomination. The Court, by four to one, held that the devotees did not constitute a separate religious denomination, that the exclusion was not an essential practice, and that it violated Articles 14, 15, 21 and 25(1), with Chandrachud J holding additionally that the notion of pollution attaching to menstruation amounted to a form of untouchability within Article 17.

Indu Malhotra J dissented on the ground that courts should not ordinarily test the rationality of religious practice. The review petitions were referred to a nine-judge Bench in Kantaru Rajeevaru (Right to Religion, In re-9 J.), (2020) 2 SCC 1, and the larger reference remains undecided, so the decision stands but its future is open.

The pattern in both is the same: a practice is defended as religious, the Court asks whether it is essential to the religion and whether it survives Part III, and the sanction fails. The criticism, which a good answer states, is that the essential religious practices test requires judges to decide theological questions they are not equipped to decide, and that a straightforward constitutional morality test would be more honest.

Jainism holds the most rigorous position. Ahimsa is the first of the five vows, extended to thought and speech as well as act, and supported epistemically by anekantavada, the doctrine that reality is many-sided and no single standpoint captures it, which is an argument against the certainty that violence requires. Gandhi's debt to it was direct, through his Jain mentor Shrimad Rajchandra.

Buddhism makes abstention from taking life the first of the five precepts and grounds it in compassion rather than in prohibition. Its greatest political expression is Ashoka's Thirteenth Rock Edict, which records remorse for the Kalinga war and announces conquest by dhamma in place of conquest by arms, while candidly retaining the machinery of the state.

Hinduism is genuinely divided, and pretending otherwise loses marks. Ahimsa is a yama in the Yoga Sutra and the Mahabharata calls it the highest dharma, yet the same epic is framed by a war fought at divine urging, and the Arthashastra is a manual of statecraft in which force is an instrument of policy. The tradition contains both a renunciatory ethic and a ksatriya ethic, and Gandhi's achievement was to read the Gita allegorically so that the battle became an inner one.

Conclusion. The four approaches to violence in India answer different questions, and the structural and political economy approaches are indispensable because the legal approach can see only events. Religiously sanctioned structural violence is real and it works by supplying legitimacy rather than force, which is why the humiliations listed in section 3 of the 1989 Act are punished for their symbolic content, and why Article 17 alone among fundamental rights operates directly against private persons.

Indian courts have consistently refused the religious defence, in Sri Venkataramana Devaru on temple entry, in Shayara Bano on instantaneous talaq and in Indian Young Lawyers Association on Sabarimala, although the last is under reference to a nine-judge Bench and the essential religious practices test used in all three is open to the criticism that it makes judges into theologians. As to the five traditions, each contains a doctrine of non-injury and each contains a licence for legitimate force, so no tradition is simply a religion of peace or of the sword; the Indian contribution, from Ashoka's edict to Ghaffar Khan's Khudai Khidmatgar and Gandhi's satyagraha, is the attempt to turn a personal ethic of ahimsa into a public method, and that attempt is what the modern law of protest, dissent and public order is still negotiating.

The nature of agrarian violence: five features

It was restitutive, not revolutionary. With the partial exception of the Santhal hool, the risings did not demand a new social order. They demanded the restoration of a customary entitlement: the rent that had always been paid, the forest that had always been used, the share that had always been left. Ranajit Guha's study of peasant insurgency identifies exactly this negative and restorative consciousness.

Its targets were documents and their holders. The peasant burnt bonds, account books and decrees before he burnt anything else, and he attacked the moneylender's house rather than the moneylender. This is the single most revealing fact in the whole field, because it shows the insurgent had correctly identified where his subordination now lived: not in custom or in force but in an enforceable legal instrument.

It was local and short. Risings were district-scale and rarely lasted a full season, because the peasantry could not leave the land and because the colonial state's military superiority was decisive once concentrated.

The colonial legal order, instrument by instrument

Property law: the Permanent Settlement of 1793. Cornwallis's settlement created in the zamindar a heritable, saleable proprietary right, fixed the revenue in perpetuity, and said nothing about the cultivator. The occupancy right that custom had given the resident cultivator survived only as a matter of usage, and Regulation VII of 1799 and Regulation V of 1812 gave the zamindar summary powers to distrain a tenant's crops and property for arrears without recourse to a court. The Pabna agrarian leagues of 1873 in eastern Bengal were formed to resist enhancement of rent enforced through precisely this machinery, and the state's answer, the Bengal Tenancy Act, 1885, at last conferred occupancy rights on cultivators of twelve years' standing.

Revenue law: the ryotwari cash demand. In the Madras and Bombay Presidencies the state settled directly with the cultivator, which sounds more benign and was in one respect worse: the demand was in cash, fixed in advance, and payable irrespective of the harvest. It forced the cultivator into the credit market and made land the security.

Criminal law enforcing a commercial contract: indigo. The indigo planters of Bengal maintained their supply through advances that were structurally irredeemable and through legislation which made breach of an indigo contract a criminal matter, so that a ryot who refused to sow faced prosecution rather than a civil suit. The Indigo Revolt of 1859-60 was in form a strike, a mass refusal to sow, and the Indigo Commission of 1860 substantially vindicated the ryots. The comparable device in Bihar was the tinkathia system, under which a tenant was obliged to plant indigo on three kathas in every bigha; Gandhi's Champaran satyagraha of 1917 produced the Champaran Agrarian Act, 1918, which abolished it.

Forest law: the criminalisation of subsistence. The Indian Forest Act, 1865 and its much wider successor of 1878 created reserved and protected forests and converted shifting cultivation, grazing and the collection of minor forest produce into offences. The tribal risings, the Kol insurrection of 1831-32, the Santhal hool of 1855-56, and Birsa Munda's ulgulan of 1899-1900, are unintelligible without this, and the settlements that followed were again tenurial: the Santhal Parganas were placed under a special non-regulation regime and the Chotanagpur Tenancy Act, 1908 restricted the transfer of tribal land.

To what extent: the honest answer

The legal order's contribution was very large but not exclusive, and three qualifications are needed.

First, the law converted misfortune into dispossession, but it did not create the misfortune. The proximate cause of the Deccan riots was a price collapse, and the proximate cause of much unrest was famine, of which the Bengal famine of 1770 and the famines of 1876-78 and 1896-97 are the largest. What the legal order did was ensure that a bad season ended in the permanent loss of the holding rather than in a remission, which under Mughal practice it usually had.

The same questions, litigated in the Republic's courts

The strongest proof that the colonial legal order created a durable structure rather than a series of episodes is that its central questions are still being decided, and decided in the same terms.

Banwasi Seva Ashram v. State of U.P., (1986) 4 SCC 753 concerned adivasis of the Mirzapur forests who had lived and cultivated there for generations and were treated as encroachers when the tracts were notified as reserved forest under the Indian Forest Act, 1927, the direct descendant of the 1878 Act. The Supreme Court held that they could not be dispossessed until their claims had been individually inquired into, appointed judicial officers to determine the claims, and permitted the acquisition of part of the land for a power project only on terms of rehabilitation. It is a nineteenth century forest settlement dispute decided under Article 21.

Orissa Mining Corporation v. Ministry of Environment and Forests, (2013) 6 SCC 476, the Niyamgiri case, completed the transfer of decision-making. Bauxite mining was proposed on hills held sacred by the Dongria Kondh. The Court held that the community and religious rights recognised by the Scheduled Tribes and Other Traditional Forest Dwellers (Recognition of Forest Rights) Act, 2006 had to be settled first, and that it was for the gram sabhas, not for the executive, to decide whether the proposed mining affected those rights. Twelve gram sabhas were held and all of them refused. Two centuries after the Forest Acts made customary use an offence, the customary user was given the power of decision.

Conclusion. Agrarian violence in the eighteenth and nineteenth centuries was restitutive, local, organised through religious or tribal authority, and directed at documents, and that last feature is the clue to its cause. The colonial legal order contributed to it very substantially and in an identifiable way: it created alienable property in land, made the moneylender's bond enforceable by decree against the holding, enforced plantation contracts through the criminal law, turned customary forest use into an offence, and classified whole communities as criminal by birth.

For full marks, cover: four limbs are named and the examiner will look for all four, but the one that separates a good answer from an ordinary one is the third. Most candidates can list riots and causes; very few can say what a named commission actually found and what happened to its report. Build the answer around the commissions, use their findings as the evidence for the criticism of the police, and then explain the legal reason their findings led nowhere, which is the status of a report under the Commissions of Inquiry Act, 1952.

Causes, sorted by depth

Structural. The colonial construction of enumerable religious communities through the census and, from the Indian Councils Act, 1909, through separate electorates, made religion the unit of political representation; Partition supplied a template of mass violence; and the economic competition between communities in particular trades supplies the local incentive.

Political. Steven Wilkinson's account, that state governments prevent riots when their electoral coalition needs minority votes and permit them when polarisation pays, explains the variation between towns and States better than any other single factor. Paul Brass's institutionalised riot system explains the mechanism: rumour specialists, conversion specialists who recode an ordinary crime as communal, and crowd recruiters, standing ready in riot-prone towns.

Proximate. Processions past places of worship, cow slaughter rumours, desecration, inter-religious relationships, and since about 2013 the manufactured video and the forwarded message.

The consequences follow inexorably. A commission's finding cannot convict. Its material cannot be used as evidence in a subsequent trial. Under section 3(4) the government need only lay the report before the legislature with a memorandum of the action taken, and it may reject every recommendation. That is why the Srikrishna Commission could name 31 police officers in 1998 and produce almost no prosecutions, and why the Liberhan report of 2009 had no legal consequence at all.

The criminal justice system

The system's failures in communal cases are structural and repeat in every episode. The investigation is conducted by the same local force the commissions have found to be partisan. The witnesses are neighbours who must go on living in the same town. Charge sheets combine hundreds of accused in unmanageable single trials. And the delay is such that by the time of trial the identification evidence has decayed.

The Bharatiya Nagarik Suraksha Sanhita, 2023 has added statutory support that did not exist when these cases were decided: section 396 requires every State to have a victim compensation scheme, section 397 deals with the treatment of victims, and section 398 requires every State Government to prepare and notify a witness protection scheme. Whether these change outcomes will depend on implementation, and the record of the equivalent guarantees in earlier law is not encouraging.

In re Destruction of Public and Private Properties v. State of A.P., (2009) 5 SCC 212 arose from destruction during bandhs and agitations. The Court accepted the recommendations of the Justice K.T. Thomas Committee and the F.S. Nariman Committee and laid down that where such destruction occurs the High Court or the Supreme Court may appoint a Claims Commissioner to assess damages, that the organisers of the event may be made prima facie liable and put to proof that they were not responsible, that damages may be exemplary, and that the police should record the events on video as a matter of routine. Read with the Prevention of Damage to Public Property Act, 1984, this shifts the burden from the state to those who called the crowd out.

Kodungallur Film Society v. Union of India, (2018) 10 SCC 713 extended the framework to mob violence generally. The Court held that the state must anticipate and prevent such violence, directed the appointment of a nodal officer in every district, required video recording of protests and prompt registration of FIRs, and reiterated that the leaders of a group that causes damage are liable for it. Both decisions matter for communal violence because they operate on the organiser rather than on the individual rioter, which is where the evidence usually is.

For full marks, cover: the entire question turns on one distinction, between advocacy and incitement, and the answer should be organised around it. Set out Article 19(1)(a) and the eight grounds in Article 19(2), note that public order was added by the First Amendment in 1951 and why, then run the case line that develops the proximity requirement from Ram Manohar Lohia through Rangarajan to Shreya Singhal. Only then list the penal provisions, because a list of sections without the constitutional test is worth very little. Finish with the current position on sedition, which is unusual and which examiners are watching.

Third, "public order" and "incitement to an offence" are separate grounds, and they were not in the original text. Both were inserted by the Constitution (First Amendment) Act, 1951, and the reason is the pair of decisions handed down in 1950.

The founding cases and the amendment they provoked

Romesh Thappar v. State of Madras, AIR 1950 SC 124 is where the story begins. The Government of Madras banned the entry and circulation of the journal Cross Roads under a public safety statute, invoking public safety and the maintenance of public order.

The Supreme Court struck the ban down, holding that Article 19(2) as it then stood permitted restriction only for undermining the security of the State or tending to overthrow it, and that public order was a much wider concept which the framers had deliberately not included; a law drawn so widely that it could restrict speech falling short of that threshold was void in its entirety. Its companion, Brij Bhushan v. State of Delhi, AIR 1950 SC 129, struck down pre-censorship of Organiser on the same reasoning. Parliament's response was the First Amendment, which added public order and incitement to an offence to Article 19(2).

S. Rangarajan v. P. Jagjivan Ram, (1989) 2 SCC 574 applied the test to a film whose exhibition was said to risk agitation over reservation policy. The Court restored the certificate and held that the expression of thought must be intrinsically dangerous to the public interest and that the anticipated danger must not be remote, conjectural or far-fetched: it should have a proximate and direct nexus with the expression, like the equivalent of a spark in a powder keg. Its more important holding is the refusal of the heckler's veto: the State cannot plead its inability to handle the hostile audience as a reason to suppress the speech, because that is a plea of incompetence and it would put the freedom at the mercy of the most intolerant.

Shreya Singhal v. Union of India, (2015) 5 SCC 1 stated the distinction in its modern form and is the case to build the answer around. Section 66A of the Information Technology Act, 2000 punished the sending of information that was grossly offensive or had menacing character or that caused annoyance or inconvenience. Striking it down in its entirety, the Court distinguished three stages: discussion, advocacy and incitement, and held that only at the third stage does Article 19(2) permit criminalisation.

The penal provisions, and how they are read

The Bharatiya Nyaya Sanhita, 2023 carries the operative offences. Section 196 punishes promoting enmity between different groups on grounds of religion, race, place of birth, residence, language, caste or community, and doing acts prejudicial to the maintenance of harmony. Section 197 punishes imputations and assertions prejudicial to national integration. Section 299 punishes deliberate and malicious acts intended to outrage religious feelings by insulting religion or religious beliefs, the successor of section 295A of the Indian Penal Code. Section 152 punishes exciting or attempting to excite secession, armed rebellion or subversive activities, or encouraging feelings of separatist activity, or endangering the sovereignty, unity and integrity of India. Section 353 punishes statements conducing to public mischief.

Two glosses are essential. Ramji Lal Modi v. State of U.P., AIR 1957 SC 620 upheld section 295A on the reasoning that it does not punish every insult to religion but only aggravated forms committed with deliberate and malicious intention, and that the words "in the interests of public order" in Article 19(2) are wider than "for the maintenance of public order".

Sedition: the position a candidate must state correctly

Kedar Nath Singh v. State of Bihar, AIR 1962 SC 955 saved section 124A of the Indian Penal Code by reading it down: a Constitution Bench held that the section is constitutional only if confined to acts involving intention or tendency to create disorder or disturbance of law and order by incitement to violence, and that strongly worded criticism of government without such incitement is not sedition.

In S.G. Vombatkere v. Union of India, (2022) 7 SCC 433 the Supreme Court took the extraordinary step of directing that section 124A be kept in abeyance, that no fresh FIR be registered under it and that pending proceedings be stayed, while the Union reconsidered the provision. Section 124A has since been replaced by section 152 of the Bharatiya Nyaya Sanhita with effect from 1 July 2024, and the constitutional validity of section 152 is itself under challenge before the Supreme Court, notice having been issued. A candidate should state this as an open question rather than as settled law.

Conclusion. The relationship between free speech and incitement is a question of proximity, not of content. Indian constitutional law protects discussion and advocacy however unpopular and permits criminalisation only where the speech is directed to producing imminent disorder, and the line was drawn by Ram Manohar Lohia's requirement of a proximate and not remote nexus, sharpened by Rangarajan's spark in a powder keg and its refusal of the heckler's veto, and stated definitively in Shreya Singhal's three stages.

The penal provisions in sections 196, 197, 299, 353 and 152 of the Bharatiya Nyaya Sanhita must be read subject to that test, with the mens rea requirement of Bilal Ahmed Kaloo and the reasonable reader standard of Manzar Sayeed Khan, and Amish Devgan has added a context-sensitive framework for hate speech that takes account of the speaker's reach. The unresolved question is sedition: Kedar Nath Singh saved section 124A by confining it to incitement, S.G. Vombatkere has held it in abeyance since 2022, and its successor section 152 is now itself before the Court, so the doctrinal line is settled while its most contested application is not.

The constitutional allocation, which explains most of the difficulty

Public order is Entry 1 and police is Entry 2 of the State List, so the primary responsibility is the State's and the primary instrument is the State police. The Union's forces enter by two routes. Entry 2A of the Union List covers the deployment of any armed force of the Union in aid of the civil power of a State, and provides that the powers and jurisdiction of such a force while so deployed are a Union subject. Article 355 imposes on the Union the duty to protect every State against external aggression and internal disturbance.

The consequences are practical. The Union cannot deploy on its own assessment; the State must ordinarily requisition. The deployed force does not displace the district magistrate; it acts in aid of him. And the force's members remain governed by their own service law while their powers on the ground are those of the criminal procedure code.

The powers, with sections

The paramilitary: what it is and what it is for

The forces used in communal disturbances are the Central Armed Police Forces, chiefly the Central Reserve Police Force constituted under the Central Reserve Police Force Act, 1949, and within it the Rapid Action Force, raised in 1992 specifically for riot control. The RAF's design is instructive because it is an explicit institutional answer to the commissions' findings: battalions of mixed composition, trained in non-lethal crowd control, equipped for it, and deliberately rotated so that they have no local attachments. Its stated operating principle is impartiality by unfamiliarity.

Its advantages over the local police are exactly the qualities the commissions found the local police to lack: it has no stake in local property or politics, it is not subject to the district's political control, and it is trained for crowd management rather than for investigation.

The pattern the commissions describe is not principally corruption. It is a combination of delayed deployment, selective application of force and selective registration of cases, and its cause is that the local force is recruited from and lives in the society whose divisions it is policing, and that its postings and transfers are controlled by the political executive.

The courts have supplied the legal consequences. Mohd. Haroon v. Union of India, (2014) 5 SCC 252 arose from the Muzaffarnagar violence of 2013, which killed more than sixty people and displaced tens of thousands into relief camps.

The Supreme Court held that the State Government had failed in its constitutional obligation to maintain law and order and to protect life, directed arrests irrespective of political affiliation, ordered compensation and rehabilitation, and made clear that once the administration has notice of impending violence it cannot plead helplessness. Anita Thakur v. State of J&K, (2016) 15 SCC 525 addressed the opposite failure, that of excessive force: participants in a procession were beaten after being taken into custody, and the Court held that while the police may use reasonable force to disperse an unlawful assembly, force after the assembly is dispersed and against persons already in custody violates Article 21, and it awarded compensation.

Composition and legitimacy. Minority representation in most State police forces is well below population share, which does not by itself prove bias but makes the perception of bias unanswerable, and perception determines whether victims report and witnesses testify.

The law on the use of force is unclear at the point of use. Sections 148 and 149 authorise dispersal by force but supply no graduated standard, and the constable on the line is left with departmental standing orders. Section 151's sanction requirement then insulates the decision from review.

Investigation by the accused institution. Where the force is implicated, it also investigates, which is the failure Zahira Habibullah Sheikh and National Human Rights Commission v. State of Gujarat had to correct by transferring the trial and appointing a Special Investigation Team.

The local force possesses the intelligence and arrives first but is compromised by its embeddedness and by political control of its postings, which five commissions from Raghubar Dayal in 1967 to Srikrishna in 1998 have documented in nearly identical terms; the central force is impartial but arrives after the decisive hours and without local knowledge. The judicial answer has been to fix responsibility on the district administration in Mohd. Haroon, to penalise excess in Anita Thakur, and to impose a preventive framework in Tehseen Poonawalla, but the structural remedy was ordered in Prakash Singh in 2006 and has still not been implemented, and until it is, the effectiveness of the police in communal violence will continue to depend on the political will of the government of the day rather than on the law.

(a) Caste system as a black spot on India

What the description means. The phrase is not rhetorical excess. India's constitutional order rests on the equal moral standing of every person, and caste is the one institution that denies it as a matter of birth and does so with religious sanction. It is a black spot because it is the internal contradiction of an otherwise egalitarian constitutional settlement, and because it is invisible to the criminal law until it produces an incident.

Its structure. Ambedkar's analysis remains the sharpest: caste is not division of labour but division of labourers, arranged in a graded hierarchy in which every stratum has one below it and therefore an interest in the survival of the whole. That is why it cannot be dislodged by solidarity from below, and why untouchability is not an excess of the system but its enforcement mechanism.

Arumugam Servai v. State of Tamil Nadu, (2011) 6 SCC 405 dealt with khap and caste panchayats and the use of caste-based abuse; the Court directed that such panchayats and honour killings be "ruthlessly stamped out" and that officials who fail to prevent them face proceedings, a direction developed into a full framework in Shakti Vahini v. Union of India, (2018) 7 SCC 192. Sukanya Shantha v. Union of India, 2024 INSC 753 found caste written into State prison manuals as recently as 2024, assigning sweeping and scavenging by caste and classifying denotified tribes by hereditary traits, and struck the provisions down.

The honest assessment. Law has abolished the sanction and criminalised the humiliation, and reservation has produced a substantial Dalit presence in education, employment and legislatures. What law has not touched is endogamy, residential segregation and the informal economy of deference, and those are the mechanisms by which the system reproduces. The National Crime Records Bureau records more than fifty seven thousand crimes against Scheduled Castes in 2023 alone. A black spot is an apt description precisely because it is localised in one institution and has proved resistant to every general remedy.

The first is bail. Section 43D(5) bars release on bail where the court, on a perusal of the case diary or the report, is of opinion that there are reasonable grounds for believing the accusation is prima facie true. In NIA v. Zahoor Ahmad Shah Watali, (2019) 5 SCC 1 the Supreme Court held that at this stage the court must take the material at face value and must not weigh its admissibility or evaluate its probative worth. The practical result is that the accusation decides the bail application. The corrective in Union of India v. K.A. Najeeb, (2021) 3 SCC 713, that a constitutional court may grant bail under Article 21 where trial within a reasonable time is not possible, arrives only after years of custody.

The second is the width of the offence. In Arup Bhuyan v. State of Assam, (2023) 8 SCC 745 a three-judge Bench overruled the Court's own decisions of 2011 and held that mere continued membership of an association declared unlawful is an offence under section 10(a)(i), rejecting the earlier requirement of active incitement borrowed from American doctrine.

The third is executive designation without a hearing. The 2019 amendment allows the Central Government to designate an individual as a terrorist under section 35 on its own satisfaction; the only remedy is an application for denotification to the same Government, and thereafter a review committee headed by a serving or retired judge.

(c) The report on the Arwal massacre

The name first. No body called the Indian Human Rights Commission conducted this inquiry. The National Human Rights Commission did not exist in 1986, having been created by the Protection of Human Rights Act, 1993. The report the question refers to is the Report of the Indian People's Human Rights Tribunal on the Arwal Massacre, an unofficial tribunal convened under the auspices of the Indian People's Human Rights Commission and staffed by retired judges, among them Justice T.U. Mehta, and published in 1987. Getting this right is worth marks, because the whole significance of the report lies in its being unofficial.

It is the clearest Indian example of institutionalised violence in the strict sense, that is, lethal force applied by the state's own agents against an assembly asserting an economic claim, and it should be assessed against the law that governs such force: what are now sections 148 to 151 of the Bharatiya Nagarik Suraksha Sanhita, 2023, which permit dispersal by civil force only after a command to disperse and permit the use of armed force only where the assembly cannot otherwise be dispersed, coupled with section 151, which requires government sanction before any prosecution of the officers concerned. That sanction requirement is why nothing followed.

It shows what unofficial fact-finding is for. A commission under the Commissions of Inquiry Act, 1952 produces a report that binds nobody, as Ram Krishna Dalmia v. Justice S.R. Tendolkar, AIR 1958 SC 538 holds, and that the government may decline to publish. A people's tribunal has even less legal force, but it has the one thing the official machinery lacked here, which is publication. In the absence of a National Human Rights Commission, which arrived only in 1993 with powers under sections 12 and 18 of the 1993 Act to inquire into violations and to recommend compensation, unofficial tribunals were the only mechanism that recorded the event at all.

And the Arwal report is the clearest instance of the state's own violence going unexamined, because the official inquiry was never laid before the legislature, the sanction requirement in what is now section 151 of the Bharatiya Nagarik Suraksha Sanhita blocked prosecution, and the only public record was made by an unofficial tribunal with no legal power at all.