National Commission for Scheduled Castes

The National Commission for Scheduled Castes is the only constitutional body in India whose entire mandate is the enforcement gap between what the Constitution promises one community and what that community actually receives. It exists because the framers anticipated that written safeguards would be ignored, and it has spent seven decades demonstrating that they were right to worry and that the remedy they designed was insufficient to the problem. Its history is a story of successive upgrades — a single officer, then a joint commission, then a dedicated one — none of which closed the gap.

From a Special Officer to a Constitutional Commission

PSIR Optional Notes {PDF}

Complete Notes Package (Paper I & II) | 1500+ Pages | Instant Download

Discounted Price | Buy Now @ ₹1000 →

The Commission’s present shape is the product of three constitutional amendments spread across fifty years, each responding to the demonstrated inadequacy of the arrangement before it.

• Article 338 as originally enacted provided not for a commission but for a Special Officer for the Scheduled Castes and Scheduled Tribes, appointed by the President.
  • His duty was to investigate all matters relating to the safeguards provided for SCs and STs under the Constitution and to report to the President on their working, at such intervals as the President directed.
  • The President was required to lay all such reports before each House of Parliament.
  • The office came to be known as the Commissioner for Scheduled Castes and Scheduled Tribes, with a network of regional offices, and it was the sole institutional watchdog for both communities from 1950 to 1990.
• The design reflected the Constituent Assembly’s judgment that safeguards written into a document would not enforce themselves.
  • Ambedkar and others in the Assembly were explicit that reservation, the abolition of untouchability and the protective articles would be paper guarantees unless somebody was charged with watching them.
  • What the framers did not resolve was who would act when the watching produced findings — the Special Officer could report and nothing more.

“Political democracy cannot last unless there lies at the base of it social democracy.” — B. R. Ambedkar

The 65th Amendment and the multi-member commission

• By the late 1980s the government had concluded that a single officer with a reporting mandate was structurally too weak for the volume and gravity of the work.
  • The Union government initiated the 65th Constitutional Amendment in 1988 on the finding that institutional support for SCs and STs was insufficient.
  • The Constitution (Sixty-fifth Amendment) Act came into force in 1990, substituting Article 338 in its entirety.
• The amendment replaced the Special Officer with a high-level, multi-member National Commission for Scheduled Castes and Scheduled Tribes.
  • The change was qualitative, not merely numerical: a commission, unlike an officer, could hold hearings, take up complaints, sit in benches, and speak with the authority of a collegial constitutional body.
  • It acquired the powers of a civil court, a duty on governments to consult it, and an entrenched report-and-memorandum mechanism — none of which the Special Officer had.
  • A non-statutory multi-member commission had already been created by a government resolution of 1978, with a chairman and four members, and was renamed the National Commission for Scheduled Castes and Scheduled Tribes in 1987 as a national-level advisory body; the 65th Amendment gave that arrangement constitutional footing rather than inventing it.
  • The first constitutional commission was constituted on 12 March 1992, so the gap between the amendment and a functioning body was itself two years.

The 89th Amendment and bifurcation

• The Constitution (Eighty-ninth Amendment) Act, 2003 bifurcated the combined commission into two separate constitutional bodies.
  • Article 338 was recast to establish the National Commission for Scheduled Castes.
  • A new Article 338A was inserted to establish the National Commission for Scheduled Tribes.
  • The two commissions became functional as separate entities in 2004.
• The rationale for splitting was that the problems of the two communities are not the same problem, and a single body inevitably subordinated one agenda to the other.
  • Scheduled Caste disadvantage is organised around untouchability, ritual exclusion and occupational stigma within settled society.
  • Scheduled Tribe disadvantage is organised around land, forest, autonomy and displacement, and is governed by a distinct constitutional apparatus.
  • Merging them had produced a body whose attention was chronically divided and whose expertise was thin on both sides.

The 102nd Amendment and the shedding of the OBC mandate

• Until 2018 the NCSC was also required to discharge the same functions for the Other Backward Classes that it discharged for Scheduled Castes.
  • The Constitution (One Hundred and Second Amendment) Act, 2018 relieved it of that responsibility, having created the National Commission for Backward Classes as a constitutional body under Article 338B.
  • The Commission’s remaining extra-caste function is the Anglo-Indian community, for which it continues to investigate safeguards and report to the President.

Why the Office Was Upgraded Twice, and What the Upgrades Did Not Fix

The two upgrades were responses to a real diagnosis. Whether they treated the disease is a separate question, and the answer bears directly on every criticism the Commission attracts today.

• What the Special Officer could do was narrow but not trivial.
  • He could investigate, build a record, and produce annual reports of real evidentiary value on the non-implementation of safeguards.
  • Several of those early reports remain the best documentary record of how reservation, land-ceiling redistribution and the untouchability law were actually working in the districts.
• What he could not do defined the case for reform.
  • He had no power to summon a witness or compel a document — every fact depended on the cooperation of the very administration under scrutiny.
  • He could not inquire into an individual complaint as a matter of right; his mandate was systemic monitoring.
  • Governments were under no duty to consult him before framing policy that affected the communities.
  • He was a single individual in a country of half a million villages, and the office’s staffing never approached the scale of the mandate.
• What the 65th Amendment added was precisely the missing machinery.
  • Civil court powers to summon, examine on oath and requisition records.
  • An express function of inquiring into specific complaints of deprivation of rights.
  • A duty on the Union and the states to consult the Commission on all major policy matters affecting Scheduled Castes.
  • A memorandum requirement obliging the government to state, in Parliament, its reasons for rejecting any recommendation.
• What neither amendment supplied is the reason the criticism persists.
  • The Commission acquired the powers of a court of inquiry but not the powers of a court of decision — it can find a fact and it cannot order a remedy.
  • There is no independent investigative machinery: inquiries are conducted through the state administration, which is frequently the respondent.
  • No consequence attaches to a government that consults nobody, accepts nothing and explains itself late or not at all.
  • The appointment process was left entirely to executive discretion, with no qualifications prescribed and no security of tenure written into the Article.

Every upgrade to Article 338 has strengthened the Commission’s capacity to find out what is happening and left untouched its capacity to change it.

Composition, Appointment and the Independence Architecture

• The Commission consists of a chairperson, a vice-chairperson and three other full-time members, of whom one is to be a woman member.
  • All five are appointed by the President by warrant under his hand and seal — the same formal instrument used for the highest constitutional offices.
  • The conditions of service and tenure of the chairperson, vice-chairperson and members are determined by rules made by the President under Article 338(2), and the term in practice is three years from the date of assumption of office.
• The warrant under hand and seal is a genuine mark of constitutional standing, but it is a form and not a safeguard.
  • The President acts on the aid and advice of the Council of Ministers, so the substantive choice is the government’s.
  • Unlike the Comptroller and Auditor General or the Chief Election Commissioner, the Commission’s members enjoy no constitutionally entrenched removal procedure — there is no requirement of an address by both Houses on proved misbehaviour or incapacity.
  • Article 338 prescribes no qualifications whatever — no requirement of judicial experience, of expertise in social justice, or even of membership of the Scheduled Castes.
• The consequence is an appointment process with no criteria, no shortlist and no scrutiny.
  • The absence of any institutionalised procedure has produced what critics call a spoils system: the offices are filled as political rewards and vacated with changes of political fortune.
  • The Commission has been chaired almost exclusively by serving or former politicians of the party in government, which shapes both the choice of cases it pursues and the vigour with which it pursues them against that government.
  • The absence of a fixed removal procedure means a member who displeases the executive has no constitutional shield.

• The Commission’s secretariat is staffed largely by officers on deputation from the central and state services, with a Secretary of Additional Secretary rank at its head.
  • Officers who will return to the parent cadre are structurally disinclined to press findings against the administration they will rejoin.
  • This is the same deputation problem that has drawn international criticism of India’s human rights machinery, and it applies with equal force here.
• The Commission maintains state offices across the country, but these are thinly staffed relative to caseload and have no independent investigative capacity of their own.

The Six Functions Under Article 338(5)

Article 338(5) sets out the Commission’s duties exhaustively. The list runs from systemic monitoring at one end to individual grievance at the other, and the balance between those two poles is where most of the criticism of the Commission’s actual practice lands.

• To investigate and monitor all matters relating to the safeguards provided for the Scheduled Castes under the Constitution, under any law in force, or under any order of the government, and to evaluate the working of such safeguards.
  • This is the systemic function inherited directly from the Special Officer, and it covers the entire field of protective legislation and executive policy.
• To inquire into specific complaints with respect to the deprivation of rights and safeguards of the Scheduled Castes.
  • This is the grievance function, and it is the one that brings the Commission into contact with individual citizens.
  • It is also the function that generates the Commission’s caseload, its public profile and its reputation, for good or ill.
• To participate and advise on the planning process of socio-economic development of the Scheduled Castes and to evaluate the progress of their development under the Union and any state.
  • This is the developmental function, and it is the constitutional hook for the Commission’s oversight of earmarked expenditure.
• To present to the President, annually and at such other times as the Commission may deem fit, reports upon the working of those safeguards.
  • The phrase “at such other times as the Commission may deem fit” is the basis of its special reports on particular episodes or sectors.
• To make recommendations in such reports as to the measures that should be taken by the Union or any state for the effective implementation of those safeguards and other measures for the protection, welfare and socio-economic development of the Scheduled Castes.
• To discharge such other functions in relation to the protection, welfare, development and advancement of the Scheduled Castes as the President may, subject to any law made by Parliament, by rule specify.
  • This residual clause makes the mandate expandable by executive rule without further constitutional amendment.

The Anglo-Indian community

• Article 338(10) directs that references to the Scheduled Castes in the Article be read as including the Anglo-Indian community, so the Commission must discharge the same functions for it as it does for Scheduled Castes.
  • It must therefore investigate the constitutional and legal safeguards for Anglo-Indians and report to the President on their working.
  • The function is an anomaly of drafting history — the Anglo-Indian community was one of the specially protected groups of 1950 and never acquired a body of its own — and it occupies a negligible share of the Commission’s attention.
  • It became more anomalous after the 104th Amendment (2019) ended Anglo-Indian nominated representation in the Lok Sabha and the state assemblies, leaving the Commission monitoring a shrinking set of safeguards.

Reports, Civil Court Powers and the Duty to Consult

The three clauses that follow the function list are where the Commission’s real leverage — such as it is — resides. They are also the clauses most often summarised away, and each contains an obligation on government rather than a power in the Commission.

The report mechanism

• The Commission presents an annual report to the President, and may submit a report as and when it thinks necessary.
• The President places all such reports before Parliament, accompanied by a memorandum explaining the action taken on the Commission’s recommendations.
  • The memorandum must also contain the reasons for the non-acceptance of any recommendation.
  • Where a report relates to a matter within a state’s competence, a copy is sent to the Governor, who lays it before the state legislature with a similar memorandum.
• This is a compelled-explanation mechanism, and it is the constitutional design’s answer to the absence of enforcement power.
  • The theory is that a government which must state its reasons on the floor of the House for rejecting a recommendation will either accept the recommendation or supply reasons that can be politically contested.
  • The mechanism collapses entirely if the reports are not tabled, and in practice tabling has been irregular for years.

Powers of a civil court

• Under Article 338(8), the Commission, while investigating any matter or inquiring into any complaint, has all the powers of a civil court trying a suit, and in particular in respect of:
  • Summoning and enforcing the attendance of any person from any part of India and examining him on oath.
  • Requiring the discovery and production of any document.
  • Receiving evidence on affidavits.
  • Requisitioning any public record from any court or office.
  • Issuing commissions to examine witnesses and documents.
  • Any other matter which the President may determine.
• Separately, Article 338(4) gives the Commission the power to regulate its own procedure — it frames its own rules for hearings, benches and case allocation.
• These are powers of inquiry and not powers of decision, a distinction the Supreme Court has recently insisted upon.
  • A civil court’s powers of summons are only useful to a body that can do something with the evidence they produce.
  • The Commission can compel a district magistrate to appear and answer; it cannot compel him to act on the answer.

The duty to consult

• Article 338(9) requires that the Union and every state government shall consult the Commission on all major policy matters affecting the Scheduled Castes.
• This is, on paper, the strongest thing in the Article, and it deserves more attention than it usually gets.
  • It is not a power the Commission exercises but an obligation binding on governments, and it operates before a policy is made rather than after harm has occurred.
  • It reaches every major policy — reservation rules, land legislation, education policy, welfare scheme design, budget earmarking — not merely matters of atrocity or discrimination.
  • It makes the Commission, in principle, a standing participant in policy formulation rather than a post-hoc complaints tribunal.
• In practice the clause has been the least enforced provision of Article 338, for three reasons that are structural rather than accidental.
  • “Major policy matter” is nowhere defined, so a government that does not wish to consult can characterise almost anything as routine administration.
  • There is no prescribed form or timing of consultation — a file sent for comment a week before a Cabinet note is technically compliance.
  • No consequence follows from non-consultation: the Constitution attaches no invalidity to a policy made without consulting the Commission, and no court has struck one down on that ground.
• The result is that the clause is frequently invoked in the Commission’s annual reports and rarely honoured in the making of policy, which is the design-versus-practice gap of the whole institution in miniature.

The Field the Commission Polices: Safeguards for Scheduled Castes

The Commission’s mandate is defined by reference to “safeguards”, and it has itself organised that field into a five-fold classification that structures its annual reports and its internal wings.

• Social safeguards — the abolition of untouchability and the prohibition of forced labour and trafficking.
• Economic safeguards — land, debt bondage, earmarked expenditure and livelihood protection.
• Educational and cultural safeguards — special provision for advancement in education and the promotion of educational interests.
• Political safeguards — reservation of seats in legislatures and local bodies.
• Service safeguards — reservation in appointment and promotion, and relaxation of standards.

Constitutional safeguards

• Article 17 is the provision the Commission most often invokes and the one least amenable to institutional enforcement.
  • It is one of the few fundamental rights enforceable against private persons, not merely against the state.
  • It contains no definition of untouchability, which was a deliberate choice — the Constituent Assembly declined to freeze the practice into a fixed set of acts.
  • Its second limb — that enforcing any disability arising out of untouchability shall be an offence — is the constitutional command that produced the penal legislation.
• Article 335 is a double-edged safeguard.
  • Its efficiency proviso has repeatedly been used to resist reservation, particularly in promotions and in technical and higher services.
  • The 82nd Amendment (2000) inserted a proviso to Article 335 permitting the state to make provision in favour of SCs and STs for relaxation in qualifying marks or lowering of standards of evaluation for reservation in matters of promotion, reversing the effect of judicial decisions that had barred such relaxation.
• Article 341 is the source of the Presidential list, and its second clause — that only Parliament may amend the list — is precisely the provision at the centre of the sub-classification litigation.

Statutory safeguards

• The Untouchability (Offences) Act, 1955 was the first legislative implementation of Article 17.
  • It was renamed the Protection of Civil Rights Act, 1955 by an amendment of 1976, which also stiffened the penalties, made all offences cognisable and non-compoundable, and cast a duty on state governments to take measures for effective implementation.
  • The rename mattered symbolically: the statute moved from prohibiting a practice to protecting a right.
  • Its record is poor. Registrations under it have collapsed to a few dozen cases a year nationally, and the near-total displacement of the 1955 Act by the 1989 atrocities law is itself evidence that proving “untouchability” as such in a criminal court is close to impossible.
• The Employment of Manual Scavengers and Construction of Dry Latrines (Prohibition) Act, 1993 was the first statutory attack on the occupation most closely tied to caste stigma.
  • It required states to adopt it by resolution, and many did so late or not at all, which is why its enforcement record is negligible.
• The Prohibition of Employment as Manual Scavengers and their Rehabilitation Act, 2013 replaced it with a far wider instrument.
  • It outlaws all manual scavenging and hazardous cleaning of sewers and septic tanks without protective gear, not merely dry latrines.
  • It creates a duty of survey, identification and rehabilitation — cash assistance, scholarships, housing, skill training and concessional loans.
  • Penalties run to two years’ imprisonment or a fine of up to one lakh rupees, and offences are cognisable and non-bailable.
  • It is rights-based rather than merely prohibitory, and it does not require state adoption.
• Reservation in services rests on executive orders and departmental instructions rather than a general statute — a striking gap for a safeguard of this importance.
  • The Central Educational Institutions (Reservation in Admission) Act, 2006 covers admissions.
  • Promotion reservation operates under Articles 16(4A) and 16(4B) and the 77th, 81st, 82nd and 85th Amendments, and remains subject to the requirements of quantifiable data laid down by the courts.
• The Scheduled Caste Sub-Plan, now the Development Action Plan for Scheduled Castes (DAPSC), requires ministries to earmark a share of plan expenditure proportionate to the SC share of population.
  • Around 38 ministries and departments currently earmark funds under it.
  • The Commission’s developmental function under Article 338(5) gives it standing to evaluate this earmarking, and it is one of the few places where the Commission produces genuinely useful quantitative oversight.
  • Utilisation has consistently fallen short of revised estimates, with the rate slipping from above 93 per cent in 2023-24 to around 86 per cent in 2025-26 — and the deeper criticism is that funds are booked to DAPSC for general schemes that happen to have SC beneficiaries, inflating the notional allocation.

The Scheduled Castes and Scheduled Tribes (Prevention of Atrocities) Act, 1989

The atrocities law is the single most important statute the Commission monitors, and the area in which it is judged most harshly. It was enacted because the Protection of Civil Rights Act had proved incapable of reaching the gravest forms of caste violence, which were being prosecuted, if at all, under ordinary penal provisions that carried no recognition of their caste character.

• The Act prohibits the commission of offences against members of Scheduled Castes and Scheduled Tribes, establishes special courts for the trial of such offences, and provides for the relief and rehabilitation of victims.
• Its structural innovation is that it treats caste violence as a distinct category of wrong, not as an aggravating circumstance in an ordinary crime.

What counts as an atrocity

• The Act enumerates a long list of offences. Among the most illustrative are being forced to:
  • Drink or eat any inedible or obnoxious substance.
  • Being paraded naked or with painted face or body, or otherwise subjected to an act derogatory to human dignity.
  • Sexual assault upon a woman belonging to a Scheduled Caste or Scheduled Tribe.
  • Being forced to leave one’s house, village or residence by wrongful dispossession, intimidation or social boycott.
• The 2015 amendment substantially widened the catalogue, adding among others:
  • Garlanding with footwear, compelling to dispose of or carry human or animal carcasses, compelling to do manual scavenging, and abetting or forcing dedication as a devadasi.
  • Tonsuring of head, removal of moustache, and other acts derogatory to dignity.
  • Obstructing access to public spaces, water sources, burial grounds and irrigation facilities.
  • Social or economic boycott as a defined offence.
• The offences are made out only when committed by a person who is not a member of a Scheduled Caste or Scheduled Tribe against a person who is.
  • This asymmetry is deliberate and is the Act’s defining feature: it is not a general law against violence but a law against a specific structure of domination.
  • It is also the feature most frequently mistaken for arbitrariness by critics, who read it as a law that punishes one group and not another.
  • The constitutional justification is Article 15(4) read with Article 17: the state may make special provision, and untouchability is a wrong whose direction is intrinsic to its definition.

Procedure, punishment and the burden on the state

• Offences under the Act are cognisable — the police may register a case and arrest without a warrant — and non-bailable, so bail is not available as of right.
• The general punishment ranges from a minimum of six months to five years with fine, with enhanced and specified punishments for the graver offences, extending to life imprisonment and, for certain offences causing death, the death penalty.
• Section 4 makes dereliction of duty by a public servant an offence in itself.
  • A public servant who is not a member of a Scheduled Caste or Scheduled Tribe and who wilfully neglects his duties under the Act is punishable with imprisonment of six months to one year.
  • The 2015 amendment specified those duties — registering an FIR, recording the victim’s statement, reading it over, correctly applying the Act’s sections, protecting the victim and witnesses — so that neglect became provable rather than merely alleged.
  • This provision is the Act’s answer to the fact that the most common failure is not the absence of a law but the refusal of the police to apply it, and it is very rarely invoked.
• Section 14 requires the establishment of Special Courts, and after 2015 Exclusive Special Courts, to try offences under the Act, with the object of completing trial within two months of the filing of the charge sheet.
• Section 15A, inserted in 2015, creates a set of rights of victims and witnesses — the right to be treated with fairness and dignity, to be heard at every stage, to be given notice of proceedings, and to protection and travel and maintenance expenses.
• Section 21 casts a duty on the state to take such measures as may be necessary for the effective implementation of the Act.
  • The Rules of 1995 flesh this out: relief and compensation on a graded scale, immediate relief within seven days, rehabilitation including employment and land where the victim is displaced, and the constitution of State and District Level Vigilance and Monitoring Committees to review implementation.
  • The rehabilitation duty is not incidental. Caste violence characteristically destroys the victim’s livelihood and residence, so a criminal conviction without rehabilitation leaves the harm intact.
• The Commission’s role runs across all of this: it monitors registration, chargesheeting, the constitution of special courts, the payment of relief and the functioning of the vigilance committees, and reports on them to the President.

The Misuse Debate and the Judicial Round Trip of 2018 to 2020

No debate about the Commission’s work is more politically charged than the claim that the atrocities law is systematically abused. The claim, the judicial response to it, and the legislative reversal of that response together form the sharpest illustration in Indian constitutional practice of the tension between individual liberty and group protection.

The three grounds of alleged misuse

• False cases are filed to settle personal disputes, to threaten adversaries, or to extort money, exploiting the Act’s severe consequences to generate leverage.
• Minor arguments and ordinary disagreements are recast as caste-based offences, which is said to overburden the criminal justice system and divert investigative and judicial resources from genuine cases.
• The provisions permitting arrest without preliminary verification are used to harass public servants, particularly those administering schemes for Scheduled Caste communities, deterring officials from taking legitimate but unpopular decisions.

The March 2018 judgment

• In Dr Subhash Kashinath Mahajan v. State of Maharashtra, decided on 20 March 2018, a bench of Justices A. K. Goel and U. U. Lalit issued directions intended to prevent misuse.
  • A preliminary inquiry by a Deputy Superintendent of Police was required before registering a case, to establish that the allegations were not frivolous or motivated.
  • Arrest of a public servant required the written permission of his appointing authority; arrest of a non-public servant required the approval of the Senior Superintendent of Police, with reasons recorded and supplied to the accused and the court.
  • The magistrate was required to examine those reasons before permitting further detention.
  • Anticipatory bail was held available despite Section 18 of the Act, where no prima facie case was made out or the complaint appeared motivated and false.
• The reasoning rested on acquittal and case-disposal statistics treated as a proxy for false complaints, and on the proposition that the Act could not be allowed to become an instrument of blackmail or private vengeance.

The reaction and the legislative reversal

• The judgment provoked immediate and large-scale protest by Dalit and Adivasi organisations, culminating in a Bharat Bandh on 2 April 2018 in which several people died in clashes with police.
  • The objection was not primarily legal. It was that the Court had read a general presumption of falsity into complaints from one class of citizens, and had done so on statistics that measure conviction failure, not fabrication.
• Parliament responded with the Scheduled Castes and Scheduled Tribes (Prevention of Atrocities) Amendment Act, 2018, passed in August 2018, which inserted Section 18A.
  • No preliminary inquiry is required for registration of an FIR.
  • The investigating officer needs no approval for arrest.
  • Section 438 of the Code of Criminal Procedure — anticipatory bail — shall not apply, notwithstanding any judgment or order of any court.
  • The amendment therefore legislatively nullified each of the directions, restoring the Act to its pre-judgment operation.

The Court’s reversal

• On 1 October 2019, in Union of India v. State of Maharashtra, a bench of Justices Arun Mishra, M. R. Shah and B. R. Gavai allowed the review petitions and recalled the directions of March 2018.
  • The Court held that the directions had imposed fetters on registration and arrest more onerous than those under the general criminal law, without any warrant in the statute.
  • It held that there was no ambiguity in the Act, and therefore no legislative vacuum justifying the exercise of the Court’s extraordinary power under Article 142 — an express acknowledgement that it had strayed into legislative territory.
  • It rejected the premise of the earlier reasoning outright.

“There is no presumption that the members of the Scheduled Castes and Scheduled Tribes may misuse the provisions of law as a class.” — Supreme Court, Union of India v. State of Maharashtra

• In Prathvi Raj Chauhan v. Union of India, decided on 10 February 2020, a bench of Justices Arun Mishra, Vineet Saran and S. Ravindra Bhatupheld the constitutional validity of Section 18A and of the 2018 amendment.
  • The Court held that the safeguards struck down had placed an unnecessary burden on the very people the Act protects.
  • Justice Bhat, concurring, held that the bar on anticipatory bail is not absolute: where the complaint on its face makes out no prima facie case, or is manifestly motivated, the court retains jurisdiction to grant it in exceptional cases.
  • This concurrence is the operative law today, and it means the ultimate position is a narrow judicial discretion inside a statutory bar rather than either extreme.

What the evidence actually shows

• The misuse argument leans almost entirely on low conviction rates and high acquittals, which is a weak inference.
  • Acquittal in a criminal trial reflects failure of proof, and in atrocity cases proof fails for identifiable reasons: hostile witnesses under social and economic pressure, delayed and defective investigation, non-appearance in court, and gaps in evidence collection.
  • Those are symptoms of the difficulty of prosecuting caste crime, not evidence that the complaints were fabricated.
• The counter-evidence is equally worth stating.
  • Under-reporting is the dominant distortion, not over-reporting: the Court itself observed that victims from these communities “hardly muster the courage to lodge even a first information report, much less, a false one”.
  • Refusal to register, registration under the wrong sections, and dilution of the caste element are documented failures that the misuse framing entirely obscures.
• A defensible position acknowledges that individual false complaints certainly occur, as they do under every criminal statute, while denying that they justify a class-wide presumption or a special procedural filter applied to no other offence.

Atrocity Data and the Distance Between Registration and Conviction

The Commission’s own reports and the annual crime statistics of the National Crime Records Bureau together describe an enforcement system that registers cases in growing numbers and converts very few of them into convictions.

• Registrations have risen steadily, from 45,935 cases of crimes against Scheduled Castes in 2019 to 57,789 in 2023, with the sharpest jump — over 13 per cent — between 2021 and 2022.
  • Roughly 262,000 cases were registered over the five years to 2023.
  • Rising registration is ambiguous evidence: it may indicate more crime, or better reporting and less police refusal, or both.
• Geographic concentration is extreme. Uttar Pradesh, Rajasthan, Madhya Pradesh and Bihar together account for well over half of all cases nationally, and six states account for roughly four-fifths.
  • This concentration tracks the states with the largest Scheduled Caste populations and the most entrenched agrarian caste hierarchies, and it makes state-level political will the decisive variable.
• Police performance has improved marginally. The chargesheeting rate rose from about 78.5 per cent in 2019 to 81.2 per cent in 2023, though over 16,000 cases remained pending investigation at the end of 2023.
• Court performance has not improved at all, and this is where the system fails.
  • The conviction rate stood at about 32 per cent in 2023, against an acquittal rate of around 63 per cent — down from roughly 40 per cent in 2020.
  • Pendency is the defining figure: about 94 per cent of cases before the courts remained pending at the end of 2023, with well over three lakh cases awaiting trial.
  • Conviction rates fall further for the gravest offences — around 30 per cent for rape and lower still for the rape of children — which inverts the expected relationship between seriousness and outcome.
• State variation is large and instructive. Conviction rates above 40 per cent in Tamil Nadu and Karnataka sit against rates in the low twenties in Madhya Pradesh and Rajasthan, which shows that the outcome is a function of prosecutorial capacity and administrative seriousness, not of the statute.
• The special courts machinery is the weakest link. The Act contemplates trial completed within two months of the charge sheet; a 94 per cent pendency rate means that timeline is not merely missed but structurally irrelevant.
  • Many states have designated existing sessions courts as special courts rather than establishing exclusive special courts, so the caseload competes with the ordinary docket and the statutory timeline is unattainable.
• Sanitation deaths are the other index the Commission is expected to track, and the figures are stubborn.
  • 471 workers died in hazardous cleaning of sewers and septic tanks between 2019 and October 2025, including 52 in the twelve months to October 2025.
  • The government’s formal position is that manual scavenging as defined by the 2013 Act has been eliminated, so these are recorded as hazardous cleaning deaths rather than manual scavenging deaths — a classification that keeps the statutory prohibition technically satisfied while the deaths continue.
  • In Dr Balram Singh v. Union of India (2023), a bench of Justices S. Ravindra Bhat and Aravind Kumar issued fourteen directions to give the 2013 Act effect.
    • Compensation for a sewer death was raised to Rs 30 lakh, with scholarships for dependants and a national survey of affected workers.
    • The Court required the phased elimination of manual entry into sewers altogether.
    • In December 2024 it found compliance inadequate, and in 2025 directed a complete halt to manual sewer entry in Delhi, Mumbai, Chennai, Kolkata, Bengaluru and Hyderabad.

The Commission monitors a system in which the probability of a case being registered has risen and the probability of it ending in conviction has fallen, and it has no power over either variable.

Judgments That Have Shaped the Commission’s Field

• State of Karnataka v. Appa Balu Ingale (1993) was the Supreme Court’s first substantive decision under the Protection of Civil Rights Act, 1955.
  • Members of a dominant caste had prevented Dalits from drawing water from a newly dug borewell, at gunpoint. The High Court had overturned the convictions on discrepancies in witness testimony.
  • The Supreme Court restored the convictions, holding that the High Court had lost sight of the fact that the disability was enforced on a threat of using a gun.
  • Justice K. Ramaswamy, concurring, situated untouchability within Article 17 as a wrong not founded on mens rea, and urged courts to approach evidence in such cases without the deep-seated prejudices that shape ordinary credibility assessment.
  • The case matters institutionally because it demonstrates the evidentiary point at which prosecutions collapse — the appellate reassessment of testimony from socially vulnerable witnesses.
• Dr Subhash Kashinath Mahajan (2018), Union of India v. State of Maharashtra (2019) and Prathvi Raj Chauhan (2020) together settled the procedural architecture of the atrocities law, and are discussed above.
• Mumbai Port Authority v. National Commission for Scheduled Castes (July 2026) is the most consequential recent ruling on the Commission’s own powers.
  • A bench of Justices Sanjay Karol and Augustine George Masih held that the Commission’s role under Article 338 is recommendatory and advisory, not adjudicatory.
  • It set aside a Bombay High Court judgment that had upheld an NCSC direction requiring the Port Authority to grant promotion benefits and salary arrears to a Scheduled Caste employee.
  • The Court distinguished the civil court powers under Article 338(8), which are powers of inquiry, from a power of binding determination, which the Article does not confer.
  • The practical effect is significant: a substantial share of the Commission’s caseload is service matters, and this judgment tells every employer in the country that its orders in such matters need not be obeyed.

Sub-Classification of Scheduled Castes and the Creamy Layer Question

The most important recent development in the Commission’s field is not about atrocity at all. It concerns the internal composition of the Scheduled Castes as a category, and it changes the terrain on which every safeguard the Commission monitors operates.

• In State of Punjab v. Davinder Singh, decided on 1 August 2024, a seven-judge Constitution Bench headed by Chief Justice D. Y. Chandrachud held by 6:1 that states may sub-classify Scheduled Castes for the purpose of reservation.
  • It overruled E. V. Chinnaiah v. State of Andhra Pradesh (2004), which had held that Scheduled Castes constitute a homogeneous class incapable of sub-division.
  • The majority reasoned that Article 341 governs identification — who is a Scheduled Caste — while Articles 15 and 16 govern the apportionment of benefits among those identified. A state may distribute a quota internally; it may not touch the Presidential list.
  • Sub-classification must rest on quantifiable and demonstrable data of relative backwardness and inadequate representation, and is subject to judicial review.
  • A state may not reserve the entire quota for one sub-group; preferential treatment must stop short of exclusion.
• Justice Bela M. Trivedi dissented, holding that the Presidential list under Article 341 creates a single homogeneous class by constitutional fiat, that only Parliament may vary it, and that state sub-classification is an impermissible tinkering with a Union list that will invite competitive political fragmentation.
• The judgment’s logic is that untouchability is not experienced uniformly, and that within the Scheduled Castes some communities have captured a disproportionate share of reserved posts and seats while others have captured almost none.
  • This has been the demand of movements such as the Madiga mobilisation in Andhra Pradesh and Telangana for three decades.
  • The counter-argument is that sub-classification converts a struggle against the caste order into a competition among its victims for a quota that has never been filled, and that the real deficit is unfilled reserved posts and backlog vacancies, not maldistribution.

What states have done since

• Telangana became the first state to operationalise sub-categorisation, enacting the Telangana Scheduled Castes (Rationalisation of Reservations) Act, 2025.
  • Acting on the report of the Shamim Akhtar Commission, which examined more than 8,600 representations, the state divided the SC quota of 15 per cent into three groups.
  • Group I — 15 sub-castes, about 3.3 per cent of the SC population — receives 1 per cent; Group II — 18 sub-castes, about 62.7 per cent — receives 9 per cent; Group III — 26 sub-castes, about 34 per cent — receives 5 per cent.
• Andhra Pradesh followed by ordinance, and Karnataka and Haryana have moved in the same direction, each on the basis of a commission of inquiry into intra-SC representation.
• For the NCSC this creates a new and unfamiliar workload: it must now evaluate the empirical adequacy of state sub-classification exercises, a technical task requiring statistical capacity the Commission does not currently possess.

The creamy layer question

• Four of the seven judges, most explicitly Justice B. R. Gavai, observed in concurring opinions that the state should identify and exclude a creamy layer within the Scheduled Castes and Scheduled Tribes, as it does for the Other Backward Classes under Indra Sawhney.
  • The argument is that a child of parents already in the IAS or an equivalent position competes with a first-generation aspirant from a landless family, and that treating them as equally disadvantaged defeats the purpose of the reservation.
• The contrary argument, and it is the stronger one on the constitutional text, is that SC disadvantage is not economic in origin.
  • Scheduled Caste status derives from untouchability, a ritual and social disability that income does not dissolve — a wealthy Dalit officer’s children still face discrimination in housing, marriage, education and workplace networks.
  • Indra Sawhney itself confined the creamy layer doctrine to OBCs and expressly did not extend it to SCs and STs.
  • Representation, not affluence, is the constitutional test under Article 16(4), and Scheduled Castes remain under-represented in Group A services even after seven decades.
  • Practically, an income-based exclusion would remove precisely the households with the social capital to use reservation effectively, without expanding the pool of qualified candidates.
• The observations were obiter and not part of the operative holding, and the political response was immediate.
  • The Union Cabinet resolved on 9 August 2024 that the Constitution as framed by Ambedkar contains no provision for a creamy layer among Scheduled Castes and Scheduled Tribes, and that reservation will continue as it stands.
  • The Union has since maintained in affidavits that SC and ST backwardness is historical and social rather than economic, and that any change is for Parliament and the executive after empirical study, not for the courts.
• The question is unresolved and will return. It matters to the Commission because the Commission is the body constitutionally required to be consulted on exactly this kind of major policy matter, and it has been almost invisible in the debate.

The Record: What the Commission Has Actually Done

• Complaint handling is the bulk of its work. The Commission registered over 47,000 complaints between 2020 and late 2024 — roughly 11,900 in 2020-21, 13,960 in 2021-22, 12,400 in 2022-23 and 9,550 in 2024.
  • The largest category is atrocities, followed by land disputes and service matters in the government sector.
  • Uttar Pradesh, Bihar and Odisha generate the largest volumes, tracking the atrocity data closely.
  • Officials have acknowledged a standing backlog of roughly 50,000 to 55,000 active pending complaints, which is more than a year’s intake and indicates that disposal has not kept pace with registration.
• Its own reports are its most durable product, and their circulation is erratic.
  • The Annual Report for 2023-24 was submitted to the President on 9 July 2025, more than a year after the period it covers.
  • It recommended a National Atrocities Tracking System for real-time data, dedicated Scheduled Caste nodal officers in every central and state department, district-level social justice committees, a dedicated grievance portal, and a fund for Scheduled Caste women and youth entrepreneurship.
  • More than a dozen annual reports of the commissions for Scheduled Castes, Scheduled Tribes and Backward Classes have not been placed in the public domain for years, which defeats the entire logic of Article 338(6).
• The Commission’s five wings do not work at equal intensity.
  • The service safeguards wing is by far the most active — reservation rosters, promotion disputes, backlog vacancies, seniority.
  • Atrocities, by contrast, are the area of greatest concern and the weakest performance, marked by delays in inquiry and a tendency to affirm the position of the government whose conduct is in question.
  • The Commission has conducted no detailed study of the atrocities problem commensurate with its scale, despite having the constitutional mandate to do so.
• The imbalance has drawn the charge of elitist bias.
  • Service matters overwhelmingly concern Scheduled Caste government employees, who are among the most secure members of the community.
  • Atrocities overwhelmingly concern the landless rural poor, who have no access to the Commission and often no knowledge that it exists.
  • The Commission has, in effect, become most responsive to the constituency that needs it least, and this is a choice of priorities rather than a constraint of law.
• The structural flaw in its complaint procedure compounds this. Complaints are routinely referred back to the very authorities accused of complicity or of covering up — the district magistrate, the superintendent of police, the department head.
  • The Commission has no independent investigative machinery, so a referral is not a stage in an inquiry; it is frequently the whole of it.
  • The response that comes back is the accused administration’s own account, and the Commission is institutionally disposed to accept it.
• Hathras (2020) is the case against which the Commission is now routinely measured. A nineteen-year-old Dalit woman died after an alleged gang rape in Uttar Pradesh; the district administration cremated her body at around 2.30 a.m. without the family’s consent while they were confined to their house.
  • The Allahabad High Court took suo motu cognisance, recording that the events after the death had “shocked our conscience“.
  • In March 2023, a special SC/ST court acquitted three of the four accused and convicted the fourth only of culpable homicide not amounting to murder along with offences under the Act; no rape or murder charge was proved. Appeals are pending before the High Court.
  • Through all of this the Commission appeared as a toothless body — visible in statements, absent in the machinery that determined the outcome.
• Vacancies have been chronic and are current. The office of chairperson lay vacant between August 2023 and March 2024, when Kishor Makwana took charge; the vice-chairpersonship has been vacant since February 2024, and one member’s post is also unfilled.
  • The Commission has therefore been running on three of its five constitutional offices, with the chairperson and two members carrying the whole caseload.
  • The chairperson’s public position is that the vacancies have had no effect on functioning, which sits awkwardly against the acknowledged backlog.
  • A petition has been filed in the Supreme Court on precisely this — that with the offices of chairman, vice-chairman and members vacant, the Commission is run by its bureaucracy, which leaves it toothless and the victim without remedy.
  • The problem is not confined to this Commission: several national commissions have functioned without chairpersons or with a fraction of their sanctioned membership, which suggests a pattern rather than an oversight.

“Political will to end the scourge of caste discrimination is needed at all levels of government to alter traditional attitudes and turn well-meaning laws into reality.” — Human Rights Watch

Structural Weaknesses

• Recommendations bind nobody. The Commission’s decisions are recommendatory, not mandatory, a position now authoritatively confirmed for service matters.
  • It can therefore be effective only where the responsible authority chooses to act, which makes its output a function of the goodwill of the body it is scrutinising.
• The consultation duty under Article 338(9) is unenforced, so the Commission’s intended role in shaping policy before harm occurs has never materialised.
• Appointments are unstructured. With no prescribed qualifications, no search procedure and no entrenched removal protection, the offices operate as political patronage, and the Commission’s willingness to confront the government of the day varies with the party in power.
• The reporting chain is broken at both ends — reports are submitted late, tabled irregularly, and when tabled are not discussed, so the memorandum requirement produces a document nobody debates.
• Capacity is thin. Staff are on deputation, state offices are under-resourced, and the Commission lacks the statistical and research capability to evaluate sub-classification data, DAPSC utilisation or atrocity trends independently of the ministries that generate the numbers.
• There is an inadequate understanding of the institution’s significance within the executive itself, which treats consultation as a formality and the annual report as a compliance document.

The Commission was designed to be the conscience of the state on caste and has been permitted to become one of its filing systems.

The Reform Agenda

• The 2nd ARC’s parliamentary committee proposal is the most economical of these because it changes nothing about the Commission and everything about its audience.
  • A report that a committee must examine acquires a constituency with an institutional interest in its findings, which is exactly what the memorandum mechanism assumed and never obtained.
• Making recommendations binding is the most-demanded and least-defensible reform. A body with no adversarial procedure, no evidentiary standard and no appeal cannot be given the power to determine rights.
  • The workable version is binding process, advisory substance: a duty to respond within a fixed period, to give reasons, and to place non-compliance before Parliament.
• Should the equality commissions be merged? The argument for a single Equal Opportunity Commission covering caste, tribe, religion, gender and disability is that India’s protections are fragmented across bodies with identical powers and no coordination, producing duplication, jurisdictional confusion and uniform weakness; a consolidated body would command greater resources and political weight.
  • The Sachar Committee recommended an Equal Opportunity Commission on precisely this reasoning, and the idea has been in circulation since.
  • The argument against is that caste disadvantage is constitutionally distinct: it is the only form of disadvantage the Constitution names as an abolished practice under Article 17, and it is entrenched in Articles 330, 332, 335 and 341.
  • Merging would dissolve that specificity into a general anti-discrimination mandate and remove the constitutional status that took forty years to obtain.
  • The stronger position is coordination without merger — a common secretariat for research and data, a shared parliamentary committee, and retained constitutional identity.

Design and Practice

The National Commission for Scheduled Castes is not a weak institution because its founders were careless. It is weak because the Constitution gave it the architecture of a supervisory body and withheld the architecture of an enforcing one, and because every subsequent reform has strengthened the first without touching the second. It can summon a district magistrate, and it cannot make him act.

• The gap between design and practice has three distinct components, and conflating them produces bad diagnosis.
  • A design gap: recommendations that bind nobody and a consultation duty with no sanction — these are defects in Article 338 itself.
  • An appointment gap: an unstructured, politically driven selection process and prolonged vacancies — these are executive failures that require no amendment to fix.
  • A priorities gap: the Commission’s own choice to concentrate on service matters and to treat atrocities as secondary — this is internal, and it is the one thing the Commission could correct tomorrow.
• What the Commission does possess is constitutional status, the power to compel evidence, a direct channel to the President, and a claim on Parliament’s attention, and these are not trivial.
  • The Special Officer of 1950 had none of them, and the difference between his reports and the Commission’s caseload is real.
• The institution’s ultimate limitation is the one the Human Rights Watch observation identifies: a commission cannot supply the political will that its own existence was meant to compel.
  • Where that will exists — in the states with high conviction rates and functioning special courts — the safeguards work.
  • Where it does not, no amount of constitutional status has substituted for it.