LLM Group 5 Criminal Law and Criminal Administration Collective Violence and Criminal Justice System 2024-25 Question Paper with Solutions

For full marks, cover: the word the examiner has used in the second limb is "system", not "forces", and it is the key to a high mark. Do not describe the police and the paramilitary as two institutions; describe the chain that has to work for a riot to be prevented, from intelligence through deployment to investigation and prosecution, and show at which link it breaks. Cover causes and consequences first but keep them tight, because the second limb carries at least half the marks.

Causes, in three layers

The enabling structure. Religious identity became a unit of political counting under the colonial state through the decennial census from 1871 and, decisively, through the separate electorates introduced by the Indian Councils Act, 1909. Once seats and representation are allocated by community, community becomes a resource worth mobilising. Partition then supplied a template and a memory that every later episode draws on.

Consequences, in three horizons

Immediate. Death, injury, destruction of homes and businesses and the displacement of populations into camps, as at Muzaffarnagar in 2013.

Institutional. The criminal justice system is damaged by every episode. Investigations are conducted by a force that is itself accused, witnesses are neighbours who must go on living in the town, hundreds of accused are combined in unworkable single trials, and acquittals follow. Each acquittal teaches the victim community that the law is not available to it, which reduces reporting in the next episode.

Long term. Ghettoisation. After the Bombay violence of 1992-93 and after Muzaffarnagar, populations relocated permanently into segregated settlements. Since intercommunal civic contact is the strongest protective factor, segregation makes the next riot easier, and communal violence is therefore self-reproducing in a way that ordinary crime is not.

Tehseen S. Poonawalla v. Union of India, (2018) 9 SCC 501 converted these powers into duties with names attached: a nodal officer of at least Superintendent rank in every district, assisted by an officer of Deputy Superintendent rank, with a special task force to gather intelligence on those likely to commit such offences; identification of districts and villages where such violence has occurred; patrolling of vulnerable areas; and a direction that failure by a police officer or a district official to comply be treated as an act of deliberate negligence attracting departmental action. That last direction is the one that matters, because it converts inaction into a disciplinary offence.

Link three, deployment and the use of force. Section 148 allows an Executive Magistrate, an officer in charge of a police station or an officer of at least sub-inspector rank to command an unlawful assembly, or any assembly of five or more likely to disturb the peace, to disperse, and to disperse it by force. Section 149 allows the District Magistrate or an authorised Executive Magistrate to disperse an assembly by the armed forces where it cannot otherwise be dispersed.

Link five, investigation and prosecution. This is where the system fails most completely. The investigation is entrusted to the force whose conduct is in issue. Zahira Habibullah Sheikh v. State of Gujarat, (2004) 4 SCC 158 is the standing illustration: fourteen people were burnt alive at the Best Bakery in Vadodara, the witnesses turned hostile and all the accused were acquitted, and the Supreme Court set the acquittal aside, held that the criminal court is not a mute spectator and must use its powers to summon and examine witnesses, and transferred the retrial out of the State.

National Human Rights Commission v. State of Gujarat, (2009) 6 SCC 342 went further and appointed a Special Investigation Team to reinvestigate the major cases under the Court's continuing supervision. Both are corrections of a system that had failed, not evidence that it works.

Prakash Singh v. Union of India, (2006) 8 SCC 1 is the remedy the Supreme Court has already ordered and the States have not delivered. After the National Police Commission's eight reports between 1979 and 1981 had been ignored for a quarter of a century, the Court directed seven measures: a State Security Commission to insulate the force from extraneous influence, a minimum tenure of two years for the Director General and for officers in operational charge, separation of investigation from law and order in urban areas, a Police Establishment Board to decide transfers and postings, Police Complaints Authorities at State and district level, and a National Security Commission.

Two decades later compliance is partial in most States, and it is precisely political control over postings, which the second and fourth directions attack, that the Srikrishna Commission identified as the mechanism of partisan policing.

What the law itself has added since 2024

Two changes in the new codes bear directly on this question and are worth a short paragraph each.

Section 172 of the Bharatiya Nagarik Suraksha Sanhita, 2023 has no counterpart in the Code of Criminal Procedure, 1973. It obliges all persons to conform to the lawful directions of a police officer given to prevent a cognizable offence or to preserve public order, and permits the officer to detain or remove a person resisting such a direction, producing him before a Magistrate or releasing him within twenty four hours. Whether this becomes a useful preventive tool or a wide new power of detention will depend entirely on how strictly the courts read the word "lawful", and it is the provision to watch in the next decade of communal-violence litigation.

Conclusion. The causes of communal violence in India are layered: a colonial inheritance that made religion the unit of political representation, an electoral incentive to polarise, a standing local apparatus that manufactures the riot, and a trigger that is almost incidental. Its consequences are immediate, institutional and long term, and the long term consequence of ghettoisation makes each episode a cause of the next. As for the system, the powers under sections 148, 163 and 168 to 172 of the Bharatiya Nagarik Suraksha Sanhita are adequate, the Rapid Action Force is a genuinely impartial instrument, and Tehseen Poonawalla has supplied a preventive framework with named officers and disciplinary consequences.

For full marks, cover: the second limb asks specifically about government measures, not about judicial decisions, so organise the answer by kind of measure, legislative, procedural, institutional and financial, and criticise each on its own terms. Case law belongs in the answer as the test of whether the measure works, not as the substance of it. Set out the forms of violence first, briefly and by setting, so that each measure can be matched to the harm it is supposed to address.

The forms, by setting

Within marriage and the family. Cruelty and dowry harassment, punished by sections 85 and 86 of the Bharatiya Nyaya Sanhita, 2023; dowry death within seven years of marriage, punished by section 80, with the presumption that follows from proof of cruelty soon before death; and the wide range of coercive conduct that the Protection of Women from Domestic Violence Act, 2005 addresses by civil remedy rather than by punishment.

Acid attack, honour crimes and trafficking, punished respectively by section 124, by the general homicide provisions with the additional framework laid down for khap and honour killings, and by the trafficking provisions of the Sanhita with the Immoral Traffic (Prevention) Act, 1956.

Violence before birth, which the criminal law reaches only indirectly, through the Pre-Conception and Pre-Natal Diagnostic Techniques Act, 1994, which criminalises the determination and communication of sex rather than the termination that follows.

Measure one: legislation, and what was left out

The Criminal Law (Amendment) Act, 2013 is the central government measure of the last two decades. It followed the Committee under Justice J.S. Verma, constituted on 23 December 2012 and reporting on 23 January 2013, and it widened rape beyond peno-vaginal intercourse, created discrete offences of acid attack, sexual harassment, disrobing, voyeurism and stalking, defined consent as unequivocal voluntary agreement, and provided that non-resistance does not by itself imply consent. The Criminal Law (Amendment) Act, 2018 added graded minimum sentences and capital punishment for rape of a girl below twelve.

Measure two: procedure

Procedural reform is the government measure most likely to change outcomes, and the Bharatiya Nagarik Suraksha Sanhita, 2023 carries a substantial set. Information given by a woman complaining of a sexual offence must be recorded by a woman police officer under the proviso to section 173(1). Her medical examination is regulated by section 184. Investigation of offences under sections 64 to 71 of the Sanhita must be completed within two months of the recording of information under section 193(2).

A forensic team must visit the scene of any offence punishable with seven years or more under section 176(3), from a date to be notified by the State Government within five years. A woman may not be arrested after sunset or before sunrise except in exceptional circumstances, with a written report by a woman police officer and the prior permission of a Magistrate, under section 43(5). Sections 396 to 398 require a victim compensation scheme, regulate the treatment of victims and require every State to notify a witness protection scheme.

Three criticisms are supported by the record. The Nirbhaya Fund has been persistently under-utilised, with successive parliamentary committee reports recording large unspent balances. Fast track courts are fast only in name where they are existing courts renotified without additional judges or staff. And the internal complaints committees under the 2013 Act are absent or non-functional in most of the unorganised sector, where the great majority of Indian women work; the Act's answer, the Local Complaints Committee at district level, is the least implemented part of the statute.

Measure four: the civil remedy, which is the most under-resourced

The Protection of Women from Domestic Violence Act, 2005 is the most innovative Indian measure in this field because it does not depend on prosecution. It defines domestic violence to include physical, sexual, verbal, emotional and economic abuse, gives the aggrieved woman a right to reside in the shared household, and empowers the Magistrate to make protection, residence, monetary relief, custody and compensation orders. Its enforcement depends on Protection Officers appointed under section 8.

State of Punjab v. Gurmit Singh, (1996) 2 SCC 384 shows the same for procedure. A schoolgirl was abducted and raped and the trial court acquitted, disbelieving her because of delay in complaint and because her conduct did not match its expectations. Setting the acquittal aside, the Supreme Court held that the evidence of the prosecutrix stands on the footing of an injured witness, that corroboration is not a rule of law, that delay must be assessed against social reality, and that trials should be in camera and conducted by a woman judge where practicable. Much of what the Court had to lay down in 1996 has since become statutory, which is itself the criticism: the government legislated only after the judiciary had done the work.

Aparna Bhat v. State of Madhya Pradesh, 2021 SCC OnLine SC 230 shows that the failure is not confined to the executive. A High Court had granted bail to a man accused of molestation on condition that he present himself at the complainant's house so that she could tie a rakhi on him. Setting the condition aside, the Supreme Court prohibited bail conditions requiring contact between the accused and the victim, forbade judicial language invoking a woman's dress, conduct or past, and directed gender sensitisation modules in judicial academies and in the syllabus for the Bar examination.

For full marks, cover: the question separates cleanly into an explanatory limb and an evaluative one, and the evaluative limb should be organised around what criminal law is being used for, because criminal law does four quite different jobs here and it performs them unequally well. Explain the notion in three registers, ordinary, statutory and constitutional, and be precise that the statutory notion turns on the identity of the parties. Then take the four uses one at a time and assess each on its own evidence.

The notion in three registers

In ordinary usage an atrocity is a shockingly cruel act, and severity is the whole of the idea.

In the constitutional register an atrocity is the enforcement mechanism of a prohibited social order. Article 17 abolishes untouchability and forbids its practice in any form; Article 15(2) opens shops, wells, tanks, roads and places of public resort; Article 23 prohibits forced labour; and Article 46 obliges the State to protect these communities from social injustice and all forms of exploitation. An atrocity is what happens when the social order that these Articles abolish is asserted against a person who has acted on the equality they confer.

Reading section 3 confirms this. It punishes forcing a person to eat or drink an inedible or obnoxious substance under clause (a), dumping excreta or carcasses in or at the entrance of their premises under clause (b), garlanding with footwear or parading naked under clause (d), forcible tonsuring or the removal of clothing under clause (e), wrongful occupation of land under clause (f), compelling the carrying of carcasses or the digging of graves under clause (i), compelling manual scavenging under clause (j), dedication as a devadasi under clause (k), and social or economic boycott under clause (n). Every one of these is a ritual of subordination, and almost every one is a response to an assertion of equality.

Use two: the deterrent use, and the evidence is poor

Deterrence requires a credible probability of conviction within a period the offender can foresee, and on this the record is weak. National Crime Records Bureau compilations show cases registered under the Act rising from 45,995 in 2020 to 53,372 in 2023, with crimes against Scheduled Castes as a whole crossing 57,000 in 2023, which is best read as improved reporting. Against that, of the cases before the Special Courts in 2023 only about 20,300 trials were completed while more than 93 per cent remained pending at the end of the year, notwithstanding that section 14(3) of the Act requires trial to be completed within two months from the filing of the charge sheet.

Two further legal features blunt the deterrent. Proof of caste motive defeats the enhanced offences. In Patan Jamal Vali v. State of Andhra Pradesh, (2021) 16 SCC 225, a blind Scheduled Caste woman was raped by a man known to the family. The Supreme Court upheld the conviction for rape but set aside the conviction under section 3(2)(v), holding that the prosecution had not established that the offence was committed on the ground that the victim belonged to a Scheduled Caste.

Use three: the protective and procedural use, which is where the Act is strongest

Most of the 1989 Act is not about defining crime at all. It is about controlling the enforcement chain, on the assumption that the ordinary law already prohibited the conduct and that what failed was the police. Rule 7 of the 1995 Rules confines investigation to an officer not below the rank of Deputy Superintendent of Police. Section 4 punishes wilful neglect of duty by a public servant who is not a member of these communities. Section 8 creates presumptions, including of common intention where the accused acted in concert. Section 14 creates Special Courts and Exclusive Special Courts. Section 15A, inserted in 2016, gives the victim and witnesses a right to be heard at every stage, to be given notice, and to appeal against an order granting bail. Section 18 excludes anticipatory bail.

The directions produced nationwide protests on 2 April 2018 in which lives were lost, and Parliament responded within months by inserting section 18A, which provides that no preliminary enquiry is required for registration of an FIR and no approval is required for arrest, and that section 438 of the Code shall not apply. In Prathvi Raj Chauhan v. Union of India, (2020) 4 SCC 727 the Supreme Court upheld section 18A, while clarifying that in a case where the complaint on its face does not make out an offence under the Act anticipatory bail is not wholly excluded, and in Union of India v. State of Maharashtra, (2020) 4 SCC 761 it recalled the Mahajan directions on review. The sequence is a complete example of the legislature correcting a judicial gloss and the Court accepting the correction.

Three structural criticisms remain. Section 4 is nearly a dead letter, because prosecuting an officer for neglect requires the cooperation of the force to which he belongs.

Criminal law arrives after the humiliation, which is fatal where the wrong is a continuing condition rather than an event: social and economic boycott, exclusion from common resources, and the caste classifications that Sukanya Shantha v. Union of India, 2024 INSC 753 found written into State prison manuals as recently as October 2024, assigning sweeping and scavenging by caste and classifying denotified tribes by supposed hereditary traits, are conditions, and the Court had to strike them down under Articles 14, 15, 17, 21 and 23 rather than prosecute anyone. And criminal law cannot reach the economy of caste, which operates through endogamy, land holding and the informal labour market, none of which is an offence.

For full marks, cover: this question rewards organisation more than detail. Divide the two centuries into three phases, because the character of the violence changes decisively between them, and give each phase its representative risings and its distinctive cause. Then deal with "nature" analytically, using the historiography, because naming Ranajit Guha, Eric Stokes and the moral economy argument is what marks out a postgraduate answer from an undergraduate one. Keep a short section for what the violence achieved, and close with its continuing legal life.

Phase one, roughly 1763 to 1830: resistance to the new revenue state

The earliest disturbances were reactions to the arrival of a revenue administration that demanded more, demanded it in cash and demanded it regardless of the season.

The Kol insurrection of 1831-32 in Chotanagpur, the Santhal hool of 1855-56 in the Rajmahal hills led by Sido and Kanhu Murmu, and the Bhil and Koli risings in western India belong here. The Santhal hool was the largest rising of the period, involved tens of thousands, was suppressed by regular troops over roughly a year with very heavy Santhal casualties, and produced a distinctive administrative answer: the Santhal Parganas were constituted as a separate non-regulation district in which the ordinary law and the ordinary civil courts did not run, an early admission that the general legal order was itself the problem.

The character of this phase is millenarian and total. The Santhal proclamation that the Thakur had commanded them to rule themselves, and Birsa Munda's later ulgulan of 1899-1900, are not demands for the adjustment of rent; they are demands for the removal of an entire alien presence.

Phase three, roughly 1860 to 1900: commercial agriculture, debt and tenancy

The third phase is the most legally interesting because its grievances are generated almost entirely by the operation of the courts.

The nature of the violence: what the historiography establishes

It was restitutive rather than revolutionary. Ranajit Guha's Elementary Aspects of Peasant Insurgency in Colonial India argues that the insurgency of the period had a consciousness of its own, structured around negation, that is, the inversion of the signs of authority rather than the design of a new order. The insurgent did not want a different society; he wanted the restoration of a customary entitlement.

Its targets were documents. The bond, the ledger and the decree were burnt before anything else and often instead of anything else. In the Deccan in 1875 the destruction was overwhelmingly of paper, and the number of deaths was very small in proportion to the scale of the disturbance. This tells us that the peasant had correctly located his subordination in an enforceable legal instrument.

It observed a moral economy. The crowds distinguished carefully between the oppressive creditor or landlord and others, which is the Indian counterpart of E.P. Thompson's account of the English food riot, and it is the reason these events look disciplined rather than anarchic when the records are read closely.

What it achieved, and why that is the answer to the question

Not one rising of the period succeeded in the field. Yet the Indigo Commission followed the indigo revolt in 1860; the Bengal Tenancy Act, 1885 followed Pabna in 1873; the Deccan Agriculturists' Relief Act, 1879 followed the Deccan riots of 1875; the Santhal Parganas were removed from the ordinary legal regime after the hool; and the Chotanagpur Tenancy Act, 1908 followed the Munda ulgulan. A state that concedes a tenancy statute after each defeat of the insurgents is admitting that its own tenurial law produced them.

The continuing legal life of the period

Banwasi Seva Ashram v. State of U.P., (1986) 4 SCC 753 shows the forest question unresolved a century later. Adivasis who had cultivated and lived in the Mirzapur forests for generations were treated as encroachers on notification under forest law descended from the Acts of 1865 and 1878. The Supreme Court refused to allow dispossession until each claim had been individually inquired into, appointed judicial officers for the purpose, and permitted acquisition for a power project only on terms of rehabilitation.

What the two centuries did not settle

The period closes in 1900 only by convention. Its unfinished business runs straight into the twentieth century and then into the Republic, and a short account of that is the best test of whether a candidate has understood the nature of the violence rather than memorised its episodes.

Gandhi's Champaran satyagraha of 1917 was an agrarian dispute in the same idiom as the indigo revolt of 1859, over the tinkathia obligation to plant indigo on three kathas in every bigha, and it produced the Champaran Agrarian Act, 1918. The Bardoli satyagraha of 1928 was a revenue dispute over an enhanced assessment, settled by an inquiry that reduced it. The Tebhaga movement of 1946-47 in Bengal demanded that the sharecropper's share rise from one half to two thirds, and the Telangana insurrection of 1946 to 1951 attacked forced labour and debt bondage. Every one of these is a phase three grievance, that is, a grievance about tenure and about the enforcement of obligation, appearing in a new political setting.

Conclusion. Agrarian violence between the eighteenth and the twentieth century falls into three phases with different characters: resistance to the new revenue state to about 1830, millenarian tribal risings against forest law and outsider settlement to about 1860, and, after 1860, revolts generated by commercial agriculture, debt and the civil decree. Its nature was restitutive rather than revolutionary, disciplined by a moral economy, led through religious or tribal authority, seasonal in timing, and directed first at bonds, ledgers and decrees, which is the clearest evidence that the peasant understood the instrument of his dispossession to have changed from custom to law.

Its scope was pan-Indian and continuous, with well over a hundred distinct risings between 1757 and 1900, none militarily successful and a striking number followed by remedial tenancy statutes. That the same questions were decided again in Banwasi Seva Ashram, Samatha and Orissa Mining Corporation, and that Parliament described the position in 2006 as a historical injustice, shows that the period bequeathed a legal problem rather than a closed chapter of history.

The prehistory in short

Between 1946 and 1951 the Telangana districts of the princely State of Hyderabad saw an armed peasant insurrection against the deshmukhs and jagirdars, known locally as doras, directed at two institutions: vetti, the unpaid forced labour owed by the labouring castes, and bhagela, hereditary debt bondage. Organised through the village sanghams of the Andhra Mahasabha under communist leadership, it seized grain, burnt the doras' account books, occupied land, killed particularly hated landlords and finally fielded guerrilla units. It was fought first against the Nizam's irregulars, the Razakars, then against the Indian Army after Operation Polo of 13 to 18 September 1948, and it was called off on 21 October 1951.

Its relevance to the statehood struggle is threefold. It destroyed vetti as a social institution and forced the Hyderabad Tenancy and Agricultural Lands Act, 1950. It left a regional political memory in which collective action against the state is remembered as having worked. And it created the organisational networks from which the later left wing extremist movement in the same districts grew, which is why the state read every later Telangana agitation partly as a security problem.

The 1969 agitation: the forms of violence

The agitation of January to September 1969 was led by students and government employees, beginning with a hunger strike over the non-implementation of the safeguards and spreading into a general strike across the Telangana districts. Its violence had three components: crowd action against public property, especially the burning of buses and attacks on government offices; police firing, which was the principal cause of death and which killed several hundred people over the year, most of them young; and preventive detention and mass arrest under the security legislation of the day.

Two features should be noticed. First, the movement was overwhelmingly urban and student-led, unlike the insurrection that preceded it. Second, the state's answer was almost entirely coercive, and no judicial inquiry ever fixed responsibility for the firings, which is the origin of the sense of impunity that shaped the later phase.

Prolonged general strikes, of which the sakala janula samme of 2011, a general strike of over forty days involving government employees, coal miners and transport workers, was the largest, and whose economic damage was the movement's principal instrument of pressure.

Violence turned inward. The most distinctive and disturbing feature of this phase was a sustained wave of suicides and self-immolations, overwhelmingly by young men, presented and commemorated as martyrdom for the cause. This is properly analysed as a form of political violence, because it was public, symbolic and intended to compel a political outcome, and it is the feature that most sharply distinguishes Telangana from other statehood movements.

Institutional non-cooperation, including mass resignations by legislators and the paralysis of the State administration.

The legal framework the episode illustrates

Babulal Parate v. State of Bombay, AIR 1960 SC 51 is the decision that made the creation of Telangana constitutionally straightforward. A challenge to the bifurcation of Bombay argued that the proviso to Article 3 required a fresh reference to the State Legislature when Parliament materially altered the scheme placed before it. The Supreme Court rejected this, holding that the proviso requires only that the State's views be ascertained and placed before Parliament within the time allowed, and that Parliament is free to depart from them. The Andhra Pradesh Legislative Assembly rejected the reorganisation Bill in 2014, and under Babulal Parate that rejection had no legal effect.

What the episode teaches

Collective violence in a constitutional democracy takes the forms the law leaves available. Where armed insurrection is impossible, the repertoire becomes property destruction, the general strike and self-harm, because these impose costs without inviting the response that violence against persons would attract.

The legal order absorbed the claim twice, and each absorption was slower and costlier than the last. In 1973 it produced Article 371D; in 2014 it produced a State. Between the two lay four decades in which the grievance was neither remedied nor extinguished.

The costs fell on those with no part in the dispute. The burnt buses and the interrupted year of schooling were borne by the population the movement claimed to represent, and the suicides were of young men with the least to gain personally from the outcome.

For full marks, cover: two limbs, and the second is where most answers become vague. Give the concept precisely, using Gandhi's own distinctions, that ahimsa is not passivity, that it is inseparable from truth, and that means and ends are convertible. Then organise the impact by domain, with dated examples in each, rather than writing general praise. A first class answer also states where the doctrine failed, because "impact" includes the impact it did not have.

The concept, in Gandhi's own terms

Ahimsa is positive, not negative. Gandhi rejected the reading of the word as mere abstention from injury and described it as the largest love and the greatest charity, an active force he called soul force. The person who abstains from injury because he is afraid is not practising ahimsa at all, and Gandhi said explicitly that if the choice were only between cowardice and violence he would counsel violence.

The vows underpin it. The disciplines of the ashram, truth, non-violence, chastity, non-stealing, non-possession, control of the palate, fearlessness, the removal of untouchability, bread labour, equality of religions and swadeshi, are not decoration. They are the argument that a person who is not disciplined in private cannot be non-violent in public under provocation.

The record of the method in India

Champaran in 1917, where the grievance of indigo cultivators subject to the tinkathia obligation was met by an inquiry and then by the Champaran Agrarian Act, 1918. Kheda in 1918, a revenue satyagraha in a year of failed harvest. The Rowlatt satyagraha of 1919 and its suspension after the violence that followed. Non-cooperation from 1920 to 1922. The Salt Satyagraha of 1930, whose genius lay in choosing a tax on a universal necessity so that any person anywhere could break the law and invite the penalty. Individual satyagraha in 1940 and 1941, and Quit India in 1942.

On Indian constitutional law. The Constitution does not use the word ahimsa but carries its vocabulary. Article 51 directs the promotion of international peace and security and the settlement of international disputes by arbitration. Article 51A(b) requires citizens to cherish the ideals that inspired the freedom struggle. Articles 40, 43, 47 and 48 are Gandhian in origin. More importantly, the method itself is constitutionally protected, and the courts have marked out the space in which it may be practised.

Himat Lal K. Shah v. Commissioner of Police, (1973) 1 SCC 227 struck down a rule that allowed the police to refuse permission for a public meeting on any ground they thought fit, holding the discretion uncanalised, while upholding a requirement of prior notice and reasonable regulation of time and place. That is a legal statement of Gandhi's own rule that the satyagrahi announces his act in advance.

Ramlila Maidan Incident, In re, (2012) 5 SCC 1 held that the power to issue prohibitory orders, now section 163 of the Bharatiya Nagarik Suraksha Sanhita, 2023, is preventive and not punitive, that it may not be used against an assembly which has caused no apprehension of disorder, and awarded compensation for the midnight clearance of a sleeping crowd.

On the ethics of dissent. The requirement that the satyagrahi accept the penalty rather than evade it remains the working test that separates principled civil disobedience from ordinary lawbreaking, and it is the standard against which contemporary movements are judged.

Where it failed, and why the failure matters

It requires an opponent capable of shame. Non-violence works by imposing a moral cost on a ruler who has a public and some residual scruple. Gandhi's advice to the Jews of Germany in 1938 to meet Nazism with civil resistance is the least defensible thing he wrote, and its indefensibility identifies the precondition exactly.

It did not prevent Partition. The most successful non-violent movement in history coexisted with communal killing on a scale of hundreds of thousands and the displacement of millions in 1947, and Gandhi was himself assassinated on 30 January 1948. Any claim about impact must be measured against that.

Its members took an oath of non-violence, and they held it under conditions considerably harsher than those faced elsewhere in India, most notably at the Qissa Khwani Bazaar in Peshawar on 23 April 1930, where troops fired repeatedly on an unarmed crowd that did not disperse and did not retaliate. That a mass non-violent organisation was built out of Islamic sources in the most martial region of British India is the single best answer to the objection that satyagraha depended on a particular religious temperament, and a candidate who can name it has said something most answers do not.

Ambedkar's objection, which a serious answer must face

The most important Indian criticism of Gandhian ahimsa is not that it was ineffective against the British but that it was applied unevenly at home. Ambedkar argued that Gandhi's method was directed outward at the colonial state while the violence of caste, which was structural, everyday and internal, was addressed by appeals to the conscience of the dominant rather than by the transfer of political power to the subordinated.

Conclusion. Ahimsa as Gandhi preached it is an active discipline, inseparable from truth, resting on the proposition that the means used determine the character of the end achieved, and operationalised as satyagraha with a strict technique that includes acceptance of the penalty and suspension of the campaign when one's own side turns violent. Its impact on today's world is measurable rather than sentimental: it supplied the method of the American civil rights movement, the anti-apartheid campaign and the People Power movements, and the comparative evidence now indicates that non-violent campaigns succeed about twice as often as violent ones.

In Indian law it survives less as a value than as a protected method, mapped by Himat Lal K. Shah, Ramlila Maidan and Amit Sahni, and its vocabulary is written into Articles 51 and 51A(b). Its limits are equally clear: it requires an opponent who can be shamed, it did not prevent the worst violence in modern Indian history, and no state can adopt it wholesale. For a student of collective violence its enduring value is diagnostic, because it insists that a movement becomes what its methods make it, and the histories of agrarian, communal and terrorist violence in this syllabus are the evidence for that claim.

(a) Increase in institutional violence: causes and remedies

What it means. Institutional violence is harm inflicted by the formal institutions of the state and their agents in the course of and under colour of their functions: torture and death in police custody, encounter killings, conditions and practices in prisons, and the punitive use of administrative power such as demolition. It differs from ordinary crime in that the perpetrator is exercising public power, and from structural violence in that there is an identifiable actor.

Is it increasing? The honest answer is that reported incidence is high and stable rather than clearly rising, and that accountability has fallen close to zero. The National Human Rights Commission registered an average of over 1,700 custodial death cases a year between 2010-11 and 2021-22. Against that volume, government data placed before Parliament records that of 2,373 cases registered against police personnel for human rights violations between 1999 and 2023, only three ended in conviction, all of them in 2017, and that there were no convictions at all in the six years from 2018 to 2023. What has increased, therefore, is the gap between complaint and consequence.

Causes.

The absence of a torture statute. India signed the Convention against Torture on 14 October 1997 and has never ratified it. The Law Commission's 273rd Report of 2017 recommended ratification and annexed a draft Prevention of Torture Bill; it has not been acted upon. India therefore has no offence of torture as such, and prosecutions must be brought as hurt, grievous hurt or culpable homicide.

Remedies, and what the courts have already ordered.

D.K. Basu v. State of West Bengal, (1997) 1 SCC 416 is the foundation. Acting on a letter about custodial deaths, the Supreme Court held that custodial torture violates Article 21 even where the detention is lawful, and issued eleven requirements: accurate identification of the arresting officers, a memo of arrest attested by a family member or a respectable local person and countersigned by the arrestee, intimation to a friend or relative, entry in a diary, examination of injuries at the time of arrest, medical examination every forty eight hours by an approved doctor, and the arrestee's right to meet a lawyer during interrogation.

To these the statute has added Arnesh Kumar v. State of Bihar, (2014) 8 SCC 273, which required police to justify the necessity of arrest for offences punishable up to seven years and is now reflected in the notice procedure of section 35(3) and the sub-section (7) requirement of prior permission of an officer not below Deputy Superintendent for arrest in such cases, along with the videography and forensic requirements in section 176(3) of the Bharatiya Nagarik Suraksha Sanhita. The remaining remedies are institutional: implementation of Prakash Singh v. Union of India, (2006) 8 SCC 1, ratification of the Convention against Torture with a substantive offence of torture, and removal or narrowing of the sanction requirement, without which no set of guidelines will change the conviction figures quoted above.

(b) Collective political violence and the legal order

The relationship is not one of simple opposition and it runs in four directions.

Law as respondent, and the paradox. The state answers with proscription, preventive detention, special courts and reversed burdens, all of which have been upheld: Kartar Singh v. State of Punjab, (1994) 3 SCC 569 upheld the Terrorist and Disruptive Activities (Prevention) Act, 1985 including the admissibility of a confession to a senior police officer, and People's Union for Civil Liberties v. Union of India, (2004) 9 SCC 580 upheld the Prevention of Terrorism Act, 2002 on the reasoning that the possibility of abuse is not a ground of invalidity. The paradox is that a movement complaining that the law is not neutral is answered by suspending the ordinary law for it alone, and both statutes were in fact allowed to lapse or were repealed after their use bore out the complaint.

Criminal force under section 129 adds the elements that make it an offence: the force must be intentional, without the other's consent, and used in order to commit an offence or with the intention or knowledge that it will cause injury, fear or annoyance. Section 130 then defines assault as a gesture or preparation intending or knowing it likely that the other will apprehend the use of criminal force, so that the apprehension is punished without any contact.

Coercion belongs to the law of contract, not to the criminal law. Section 15 of the Indian Contract Act, 1872 defines it as committing or threatening to commit any act forbidden by the penal law, or unlawfully detaining or threatening to detain any property, to the prejudice of any person, with the intention of causing any person to enter into an agreement. Its consequence is not punishment but voidability under section 19.

Conclusion. The three notes are connected by a single question, which is where the law locates responsibility. Institutional violence is increasing in the only sense that matters, the gap between roughly 1,700 custodial death complaints a year and three convictions in twenty four years, and the remedies already exist on paper in D.K. Basu, Nilabati Behera and Paramvir Singh Saini and fail for want of a torture statute, of police reform under Prakash Singh, and of prosecutions unblocked by the sanction requirement.