The RTI Act 2005 anti corruption politics of India began as a wage dispute on drought-relief works in Rajsamand district, Rajasthan, where labourers who had dug and measured earthworks demanded to see the muster rolls that recorded their own names, and it ended, fifteen years later, in a statute that let any citizen compel any public authority to open its files within thirty days. The Right to Information Act received the assent of President A. P. J. Abdul Kalam on 15 June 2005 and came fully into force on 12 October 2005, and the standard ways of describing that statute miss what it was built to do. The information-rights framing treats the Act as one more entry in a global family of freedom of information laws, a technical instrument for releasing records. The policy-history framing treats it as the endpoint of a legislative sequence running through committee drafts and state statutes. Both framings are accurate about the paperwork and wrong about the politics. This rereading defends the claim the record supports: the Act of 12 October 2005 operates as anti-corruption politics, a weapon designed by a movement that had spent a decade proving, village by village, that the file was where the theft was recorded. Five features carry that design. A movement in rural Rajasthan, the Mazdoor Kisan Shakti Sangathan founded at Devdungri on 1 May 1990, invented the public hearing as an audit of the state’s own paper. Tamil Nadu enacted the first state right to information law in 1997, and a run of states followed between 1997 and 2004, testing the instrument under live administrative conditions. The United Progressive Alliance government of Prime Minister Manmohan Singh, bound by the Common Minimum Programme adopted on 27 May 2004, carried a central statute through Parliament in May 2005 with a thirty-day answer clock, a duty of proactive disclosure in Section 4, a bounded exemption list in Section 8, and two tiers of information commissions armed with penalty powers. The years after 2005 then stress-tested that architecture in public: disclosure orders reached the funding of political parties in 2013 and the office of the Chief Justice of India in 2019, and the counter-attack arrived in the Right to Information (Amendment) Act of 2019, which cut the commissioners’ fixed tenure and salary parity out of the statute. Read this way, the Act belongs beside the economic-policy-as-politics framework that organises this pillar of the series: a statute is a political settlement frozen into clauses, and the RTI Act froze the settlement of a movement that had learned to read the state through its records.

A Wage Dispute at Devdungri: The Movement That Demanded the File

The Mazdoor Kisan Shakti Sangathan, the Organisation for the Empowerment of Workers and Peasants, was formally established on 1 May 1990 at a rally of roughly one thousand people from twenty-seven villages outside Bhim town in Rajsamand district, and its headquarters was a hut at Devdungri village that its founders had occupied since 1987. The three founders named in every account, including Aruna Roy’s own history, The RTI Story: Power to the People, published in 2018, were Roy herself, Nikhil Dey and Shankar Singh. Roy had resigned from the Indian Administrative Service in 1974 and had worked for years at the Social Work and Research Centre at Tilonia before the Devdungri team formed. The organisation’s founding demand was not information in the abstract. It was the minimum wage, fixed in law for drought-relief and public works, which workers in the Bhim area were routinely denied in part or in full. The mechanism of the denial was documentary. At relief sites such as Dadi Rapat, workers were given individual tasks and paid on collective measurements; officials entered the names of dead men and of men who had never attended the site into the muster rolls; and the difference between the wage the record disbursed and the wage the worker received stayed with the recording chain. When twenty workers at one site completed their measured task in full and were paid less than workers who had done less, the Devdungri team demanded that the government engineer measure the work, and a visiting engineer confirmed in 1989 that the workers had met their targets and ordered full payment. The payment still did not arrive in full. That sequence, recorded in Roy’s 2018 account and in the movement’s own pamphlet literature, fixed the organisation’s diagnosis: a petition about wages could be ignored because the evidence of the wage was locked inside the file, and the file belonged to the office accused of the theft.

The demand that followed was precise, and its precision is what makes the RTI Act’s later architecture intelligible. The MKSS asked for photocopies of muster rolls, bills and vouchers for named works: the school building, the dispensary, the small dam, the community centre that existed complete on paper and incomplete on the ground. Officials refused, citing the Official Secrets Act of 1923 and the civil service conduct rules that made disclosure itself an offence. The movement’s answer to that refusal became its slogan, recorded across its pamphlets and banners: Hamara Paisa, Hamara Hisab, our money, our accounts. The slogan is the whole theory of the later statute in six words. Public money was the premise; the account was the remedy; and the corruption the movement opposed was defined not as bribery in the shadows but as a false entry in a public record that no member of the public was allowed to inspect. Sohangarh village supplied the second founding proof. A local strongman, Hari Singh, held over fifteen hundred acres in defiance of the ceiling laws and fined villagers for grazing cattle and collecting firewood; the Devdungri team obtained the revenue records, established the illegal holding from the state’s own documents, and the land was resumed and part of it transferred to a women’s cooperative. Records, again, were the weapon, and the lesson generalised: any entitlement that existed on paper, a wage, a ration, a pension, a land title, could be stolen in the gap between the paper and the person, and only opening the paper closed the gap.

What did the MKSS ask for when it demanded information in 1990?

The organisation asked for certified copies of the primary records of named public works: the muster rolls listing who had been paid, the bills and vouchers recording what had been bought, and the measurement books recording what had been built, so that workers could compare the state’s paper against the wage in hand and the structure on the ground.

By 1994 the MKSS had converted that demand into a repeatable institution, and the institution is the second half of the first feature. The instrument was the jan sunwai, the public hearing, and the design of the hearing explains the design of the Act better than any committee report. The first jan sunwai was held at Kot Kirana village in Pali district on 2 December 1994, and four more followed in the same winter and spring, at Vijaypura, Jawaja, Bhim and Thana, running to April 1995. The procedure was fixed. The movement obtained the official records of a panchayat’s development works for a stated period, through a sympathetic official, through the panchayat’s own disclosure, or after the Rajasthan Panchayati Raj Act was amended to let citizens take certified copies of panchayat documents. On the day of the hearing, in a public ground, with the sarpanch and the block officials invited and the press present, the records were read aloud entry by entry. Villagers then testified against the reading. A muster roll named a man as paid for twenty days of work on a school; the man stood and said he had worked six, or his widow stood and said he had been dead for three years. A bill recorded the purchase of cement; the mason testified that no cement had reached the site. The reading aloud was the audit. Literacy was not a barrier, because the record was spoken and the verification was lived experience, and the officials present were required to answer in the same public space rather than in a file noting six months later. Roy’s account records that at Kot Kirana the panchayat’s records, obtained after a campaign around the dismissal of a corrupt village functionary, disclosed payments to dead men and works that existed only in the vouchers, and that the hearing demanded three things: full transparency of public works accounts, a people’s audit as a standing institution, and restitution of the money siphoned off. Restitution is the detail the information-rights framing cannot explain. Officials and contractors returned money after jan sunwais, in some cases in cash at the hearing itself, because the hearing was not a request for information as a civic good. It was an accusation with the evidence read out loud.

The jan sunwai also carried the movement’s constitutional argument, which the national campaign later took to Delhi. The MKSS held that the right to information was not a favour Parliament might grant but a component of the rights the Constitution already guaranteed: the freedom of speech and expression in Article 19(1)(a), which is empty for a citizen forbidden to know what the state has done, and the right to life in Article 21, which for a relief worker meant the wage that bought food. That argument had judicial support before the movement existed, in State of Uttar Pradesh v Raj Narain, decided by the Supreme Court in 1975, where Justice K. K. Mathew wrote that the people of this country have a right to know every public act done in a public way by their public functionaries; the election case itself is treated in Indira Gandhi v Raj Narain and the 1975 verdict. The movement’s contribution was to take that doctrine out of the courtroom and give it a procedure a landless labourer could operate. When Prabhash Joshi, editor of the Hindi daily Jansatta, travelled to Rajasthan and titled his editorial on the hearings Hum Jaanenge, Hum Jiyenge, we shall know, we shall live, the phrase welded the wage to the record in the national press for the first time. Senior journalists including Nikhil Chakravartty and Kuldip Nayar made the same journey, and the movement that had begun with twenty workers and a measurement book acquired a national audience before it had a national law.

The hearings also won the movement its first statutory concession, and the concession’s form predicted the national Act’s Section 4. Under campaign pressure the Rajasthan legislature amended the Rajasthan Panchayati Raj Act in the mid-1990s to permit any person to obtain certified copies of panchayat documents, muster rolls, bills and vouchers among them, on payment of a fee, and to require panchayats to display their budgets, accounts, expenditures, policies and beneficiary lists on public boards and in newspapers. The amendment was narrow, confined to one tier of one state’s institutions, and it was evaded in practice by panchayat secretaries who pleaded missing registers and absent clerks. Its importance was doctrinal and tactical at once. Doctrinally, a legislature had accepted that a public record generated by public money was a public possession, inspectable as of right rather than released as grace. Tactically, the movement had its first experience of the gap between a disclosure right on paper and a disclosure practice in the office, the gap that the penalty clause and the independent commission were later invented to close. Nikhil Dey’s essays on the MKSS period return repeatedly to that gap: the organisation learned, between the amendment and Beawar, that every victory at the level of text had to be re-won at the level of the register room, and the national draft the campaign eventually carried to Delhi was written by people who assumed the register room would resist. Shankar Singh’s role in the same years completes the picture of the machine the movement built. Singh, who had given up the opportunity to become a government schoolteacher to join the movement, organised its communication arm, the puppet troupes, wall newspapers, songs and street theatre that carried the jan sunwai findings from the hearing ground into the surrounding villages, so that a falsified muster roll exposed at Kot Kirana became a story told at wells and bus stands across the block. The exposure had an audience engineered for it, which is why restitution followed exposure: a contractor who returned money at a hearing was buying back a reputation in a market the movement’s theatre had made.

Beawar, April 1996: Forty Days That Made a National Campaign

The Rajasthan government of Chief Minister Bhairon Singh Shekhawat tested the movement’s patience in the sequence the movement’s own histories record. On 5 April 1995 the Chief Minister announced in the Assembly that citizens would be given access to the records of development works, and the announcement was not implemented by the departments that held the records. The MKSS answered on the first anniversary of that unkept promise. On 5 April 1996 the organisation began a dharna, a continuous sit-in, at Chang Gate in Beawar town in Ajmer district, and held the site for forty days. The demand placed before the state government was legislative, not charitable: enact a right to information, and in the interim order the departments to supply the records the jan sunwais needed. The composition of the dharna is the fact its chroniclers stress. The largest contingent at Chang Gate was rural women from the Bhim and Raipur blocks, veterans of the wage struggles, sitting in a town square to demand not food, not work, not a wage increase, but the legal right to inspect the accounts of money already spent in their names. A demand for information from people whom the administration classified as its beneficiaries was a category error in the official mind, and that is why Beawar drew national editors to a provincial town and why the forty days converted a Rajasthan wage campaign into a national political question. The dharna ended without the state law it demanded; Rajasthan’s statute came only in 2000. Its product was organisational. In 1996 the National Campaign for People’s Right to Information, the NCPRI, was founded in Delhi, bringing the MKSS together with journalists, lawyers, retired civil servants and academics in a single platform whose stated twin purposes were to support grassroots struggles for records and to draft the legislation those struggles required.

The NCPRI’s drafting work ran on a track parallel to the government’s own. The Press Council of India, chaired by Justice P. B. Sawant, produced a model right to information bill in 1996, later revised with the National Institute of Rural Development into the Press Council NIRD draft. The United Front government appointed a working group under the former civil servant H. D. Shourie in 1997, and the Shourie draft, stronger than what governments were willing to enact, passed through two administrations without reaching Parliament’s floor. A third episode in that sequence shows the executive’s resistance in miniature: in 1999 the Union Minister for Urban Development, Ram Jethmalani, issued an administrative order opening his own ministry’s files to citizen inspection, and the Cabinet Secretary declined to let the order take effect. The lesson the campaign drew from the decade was the lesson of Devdungri in reverse. Voluntary disclosure, promised by a chief minister in 1995 or ordered by a minister in 1999, dissolved at the level of the office that held the file; only a statute with a duty, a deadline and a penalty could survive contact with the record-keeping chain. Every element of the 2005 Act that the information-rights framing treats as technical detail, the thirty-day clock, the deemed refusal, the personal penalty on the officer, is a scar from one of those failures.

How long did an applicant wait for an answer under the first state laws?

Most of the first state statutes allowed the administration thirty days or longer to answer, and several carried no enforceable deadline at all, which is why the MKSS model insisted on a short statutory clock backed by a penalty: time was the medium in which inconvenient applications were buried, and the state experiments proved the point repeatedly.

Tamil Nadu 1997: The First Statute, and Its Limits

The second feature of the architecture is a statute the movement did not write and barely influenced, and its weaknesses taught the campaign as much as Jan Sunwai practice had. Tamil Nadu became the first Indian state to enact a right to information law, passing the Tamil Nadu Right to Information Act in the first half of 1997 under the Dravida Munnetra Kazhagam government of Chief Minister M. Karunanidhi; the Act, numbered Act 24 of 1997, received the Governor’s assent on 4 May 1997. The seed of the statute, in the brief’s own phrasing, was an ordinance route debated in the state through 1996 and 1997, and the enactment placed Tamil Nadu ahead of every state where the movement itself was strong. The Act gave a citizen the right to seek information from government departments and public undertakings, and on paper it broke the principle of official secrecy for the first time in an Indian statute book. In operation it was built to fail politely. Its exemption clauses were drawn so widely that departments could refuse most of what mattered, its appeal lay to the government itself rather than to any independent body, it carried no penalty for an officer who simply did not answer, and its definition of information and its fee and procedure rules left the deciding discretion with the office holding the record. Roy’s RTI Story and the NCPRI’s assessments treat the Tamil Nadu Act as a cautionary exhibit: a right to information law written by the administration it was meant to police will reproduce the administration’s preferences, exemption first, deadline nowhere, appeal to oneself.

The Tamil Nadu enactment also fixed a political precedent the campaign exploited for a decade: the DMK had carried the promise of an information law in its manifesto for the 1996 state election, making Tamil Nadu the first case in which a governing party’s election promise, rather than a movement’s siege, produced an access statute, and demonstrating that the demand had acquired electoral value even where the movement itself had no organisation. The NCPRI’s assessments of the period treat the manifesto route and the movement route as the two roads to the same statute book, and note that the manifesto road produced the weaker law, written at the pace and to the tolerances of the secretariat that would administer it. The lesson was not lost on the drafters of the Common Minimum Programme eight years later, whose information commitment was negotiated with the movement looking over the secretariat’s shoulder.

Goa followed within the same year, enacting its Right to Information Act on 31 July 1997, and the Goa statute is generally rated the stronger of the two 1997 laws, with a wider coverage of bodies and a more workable appeal structure. The political meaning of the pair matters more than their drafting differences. By the end of 1997 the proposition that a legislature could enact citizen access to records at all had been demonstrated twice, by two different party governments, and the question in Delhi shifted from whether such a law was possible to whose version of it would be written. The MKSS read the 1997 statutes as a floor, not a model. The campaign’s own draft principles, tested at jan sunwais and in the panchayat records campaign that had already amended the Rajasthan Panchayati Raj Act to permit certified copies of panchayat documents and the posting of budgets and beneficiary lists on public boards, required four things the state laws lacked in full: a short answer clock, an independent appeal, a penalty that fell on the officer rather than the office, and a duty on the state to publish the routine record before anyone asked. That four-part test is the bridge between the movement sections of this rereading and the central statute, because the 2005 Act is legible as the first Indian information law that satisfied all four conditions at once, and the 2019 amendment is legible as the removal of the conditions that protected the second and third of them.

The State Laboratories, 1997 to 2004: Nine Statutes and a Pattern

Between Goa’s enactment in July 1997 and the central Act’s commencement in October 2005, the state-level experiment widened into the third feature of the architecture. Madhya Pradesh moved early and messily: the state government issued executive orders in 1997 opening the records of first thirty-six and then more than fifty departments, a Right to Information Bill passed by the state legislature in 1998 was denied the Governor’s assent, and a Madhya Pradesh statute followed in 2003. Rajasthan, the movement’s home state, enacted its Right to Information Act in 2000, five years after the Shekhawat announcement and four after Beawar, and Karnataka enacted its Act in the same year, bringing it into force in 2002. Delhi’s Legislative Assembly passed the Delhi Right to Information Act in 2001, and the Delhi statute earned a reputation among practitioners as the most heavily used of the state laws, its applicants filing in large numbers against the municipal and development authorities of the capital. Maharashtra’s path ran through an ordinance in 2000, a statute in 2002 that practitioners and the Anna Hazare campaign in that state criticised as weak, and a replacement Act in 2002 whose stronger form the state brought into force subsequently; the Maharashtra and Delhi enactments were the two most used state laws of the period. Assam enacted its law in 2002 and Jammu and Kashmir in 2004, completing a set of nine state statutes on the eve of the central Act. The Jammu and Kashmir statute had a second life that belongs to the record: because the central Act of 2005 did not extend to the state, the state legislature replaced its weaker 2004 law with the Jammu and Kashmir Right to Information Act 2009, modelled closely on the central Act’s machinery of clocks, commissions and penalties, so that the valley’s applicants worked for a decade under a state statute that was the central Act in all but name, until the reorganisation of 2019 unified the map.

The pattern those nine experiments established is evidence, and the anti-corruption reading of the central Act depends on reading the pattern correctly. First, usage concentrated where the grievance was material: ration cards, water connections, land records, housing allotments, works registers. The state laws were not used in the main by researchers or journalists; they were used by applicants pursuing a specific entitlement that a record could prove or restore, exactly as the MKSS had used muster rolls. Second, the administrative response concentrated on the same points in every state: delay past the deadline, transfer of the application between offices, exemption claims stretched past the text, and appeal forums that met rarely or lacked staff. Third, the statutes that lacked independent appellate commissions and personal penalties, which was most of them, decayed fastest, because nothing in their design raised the cost of refusal. The state decade therefore functioned as a controlled comparison for the drafters of 2004 and 2005. The NCPRI’s model provisions, the annual assessments by the Commonwealth Human Rights Initiative of the state laws in operation, and the campaign’s own litigation and jan sunwai experience all pointed the same way: the central Act’s anti-corruption force would live or die in its enforcement machinery, not in its preamble. That is the context in which the post-1991 institutional-political context matters to this article: the state that had liberalised its economy in 1991 while keeping its administrative secrecy intact was, by 2004, being presented with a statute drafted by the people its secrecy had injured.

The state decade had one more teacher, an assessing institution rather than a legislature. The Commonwealth Human Rights Initiative, whose access to information programme under its director Maja Daruwala tracked every state statute from enactment, published comparative assessments of the state laws in operation through the early 2000s, scoring them against the campaign’s tests and documenting, state by state, the same administrative repertoire the jan sunwais had met in the panchayat office: the application returned for a defect in form, the record declared untraceable, the exemption claimed under the clause nearest to hand, the appellate officer who was the deciding officer’s superior in the same corridor. Those assessments mattered to the central drafting because they converted anecdote into pattern at exactly the moment the National Advisory Council took up the bill. A drafter reading the state record in 2004 did not need to speculate about how departments would respond to a central Act; the response had been rehearsed in nine jurisdictions, and each rehearsal identified the clause that would be needed to defeat it. The thirty-day clock answered the delay documented in Tamil Nadu and Madhya Pradesh. The deemed refusal answered the silence documented everywhere. The independent commission answered the appeal-to-oneself documented in the 1997 statutes. The personal penalty answered the zero cost of refusal documented in all nine. The central Act’s text, read against the state record, is a set of replies, and the anti-corruption reading is simply the observation that the questions being replied to were all, at origin, questions about stolen wages, stolen rations and stolen land.

The Dead Letter of 2002 and the Opening of 2004

The central government’s first statute proved the campaign’s argument by enacting its opposite. The Freedom of Information Bill, introduced in Parliament in 2000 as a reworked and weakened descendant of the Shourie draft, was examined by the Parliamentary Standing Committee on Home Affairs, which reported in July 2001 and recommended that the flaws civil society had identified be corrected; the government did not adopt the corrections. The Bill passed both Houses in December 2002 and received presidential assent in January 2003 as the Freedom of Information Act 2002. It contained no independent appeal worth the name, no penalty, exemptions wide enough to hold the entire administrative record, and one defect that swallowed the rest: its commencement was left to a government notification, and the notification was never issued. The Freedom of Information Act 2002 therefore never operated for a single day, a statute on the book that no citizen ever used, and the NCPRI’s verdict on it, that it was a right to information law designed to disclose nothing, entered the drafting debates of 2004 as settled fact. When the new central bill was prepared, its first structural decision was negative and explicit: whatever else the statute did, it would commence on a date fixed in the Act itself, bind the officer personally, and answer refusal with an appeal outside the department.

The political opening came from the post-2004 UPA mandate. The Congress-led United Progressive Alliance formed the government on 22 May 2004 under Prime Minister Manmohan Singh, and the National Common Minimum Programme adopted on 27 May 2004 carried a commitment in terms the campaign had drafted toward for a decade: the government would enact a Right to Information Act that would be more progressive, participatory and meaningful than the dead letter of 2002. The mechanism that converted the commitment into a text was the National Advisory Council, constituted in June 2004 and chaired by Sonia Gandhi, in which Aruna Roy sat as a member. The Council considered the NCPRI’s draft in the second half of 2004 and recommended a set of amendments to the government’s bill that carried the movement’s four tests into the official draft: a thirty-day clock, an independent information commission at the Centre and in every state, penalties on defaulting officers, and a proactive disclosure chapter. The Right to Information Bill was introduced in the Lok Sabha on 23 December 2004, examined by the Standing Committee on Personnel, Public Grievances, Law and Justice, passed by the Lok Sabha on 11 May 2005 and by the Rajya Sabha on 12 May 2005, and received the assent of President A. P. J. Abdul Kalam on 15 June 2005. Its commencement clause fixed the 120th day, and on 12 October 2005 the Right to Information Act 2005 came fully into force across India except the then state of Jammu and Kashmir, which retained its own statute. The Act repealed the Freedom of Information Act 2002, which thus exited the statute book without ever having been law in operation, the only exhibit the information-rights framing needs and the exhibit the anti-corruption framing explains: the difference between the two statutes was not drafting skill. It was the presence, in 2004 and 2005, of a movement inside the drafting room and a governing coalition that had written its demand into its programme.

The Doctrine Before the Statute: What the Courts Had Already Held

The statute of 2005 did not create the right it enforced, and the adjudication between the framings turns partly on that priority. In State of Uttar Pradesh v Raj Narain, decided in 1975 in the litigation over Indira Gandhi’s Rae Bareli election, Justice K. K. Mathew held for the Supreme Court that the people of this country have a right to know every public act, everything that is done in a public way, by their public functionaries, grounding the right in the freedom of speech and expression, since a citizen cannot form or express an opinion about a government whose acts are hidden. The proposition was obiter in an election appeal and it lay in the reports for six years. Its ancestry ran through the press cases, above all Bennett Coleman and Company v Union of India in 1972, where the Supreme Court struck at the newsprint control policy in the name of Article 19(1)(a) and Justice Mathew’s opinion reasoned from the premise that a democratic citizenry cannot deliberate about a state it is forbidden to observe. The line from newsprint quotas to muster rolls is not straight, but it is continuous: in each decade the Court enlarged the class of public facts a citizen could claim as of right, and in each decade the enlargement stopped at the door of the office that kept the fact, for want of a procedure to fetch it out. In S. P. Gupta v Union of India, decided in December 1981 in the litigation over the transfer and appointment of judges, Justice P. N. Bhagwati took the proposition up and built the doctrine on it: the concept of an open government, he wrote, is the direct result of the right to know which is implicit in the right of free speech and expression guaranteed under Article 19(1)(a), and disclosure of information regarding the functioning of government must be the rule, secrecy the exception justified only where the strictest requirement of public interest demands it. The S. P. Gupta bench simultaneously narrowed the class of documents the executive could shield by class claim, holding that cabinet papers and the correspondence over judicial appointments were not immune from the court’s own inspection merely because of their category. The doctrine the movement inherited in the 1990s was therefore complete in outline: a fundamental right to know, a presumption of disclosure, secrecy as a justified exception. What the doctrine lacked was everything that makes a right usable by a person without a lawyer: a form, a fee, a deadline, a designated officer, an appeal that did not begin in a high court.

The voter cases of the following decade showed the doctrine reaching for machinery on its own. In Union of India v Association for Democratic Reforms, decided in May 2002, the Supreme Court directed the Election Commission to require every candidate for Parliament or a state legislature to declare criminal antecedents, assets, liabilities and educational qualifications, reasoning that the voter’s right to make an informed choice was part of the Article 19(1)(a) right to know; when Parliament responded with an amendment designed to dilute the direction, the Court struck the diluting provision down in People’s Union for Civil Liberties v Union of India in 2003, holding that the legislature could not curtail a fundamental right the Court had recognised. The episode is the doctrine’s biography in miniature. A right declared in 1975 and elaborated in 1981 was, by 2002, strong enough to defeat a statute and still not strong enough to tell a tehsil clerk that a muster roll must be photocopied by a date certain. Prashant Bhushan’s essays on the Act’s trajectory make the point from inside the litigation tradition: the courts had given India a right to know as a principle of constitutional interpretation, enforceable against the state in the state’s own courts at litigation prices, and the distance between that right and the jan sunwai was the distance the 2005 Act was drafted to close. The information-rights framing, which treats the Act as a specimen of a global legislative family, cannot see that distance, because in the family’s standard history a statute creates the right. The Indian sequence ran the other way: the right preceded the statute by thirty years, and the statute’s originality lies in its machinery, which is to say in its politics.

The Drafting Room: Council, Committee and the 2005 Text

The text that closed the distance was assembled in two rooms in 2004 and 2005, and the rooms’ membership is part of the evidence. The National Advisory Council, constituted in June 2004 to oversee the Common Minimum Programme’s implementation and chaired by Sonia Gandhi, took up the right to information in its first months, with Aruna Roy sitting as a member and the NCPRI’s draft before it. The Council’s recommendations to the government, conveyed in 2004, pressed for the state laws’ defects to be remedied in the central bill: coverage of the institutions the draft had omitted, a shortened response time, an independent appellate commission for the Centre and the states, penalties, and the proactive disclosure chapter. The government’s bill, when introduced in the Lok Sabha on 23 December 2004 by the Minister of State for Personnel, was widely assessed by the campaign as weaker than the Council’s recommendations in its coverage and exemptions, and it was referred to the Department-related Parliamentary Standing Committee on Personnel, Public Grievances, Law and Justice, chaired in that Parliament by E. M. Sudarsana Natchiappan. The committee heard the campaign, the press and the administration, and its report recommended strengthening the bill at the points where the state record had shown decay, including the commissions’ design and the disclosure duty; the government accepted the bulk of the committee’s direction, and the bill that passed in May 2005 carried the four tests in operative form. The parliamentary debates of 11 and 12 May 2005 show the settlement’s breadth: members across the treasury and opposition benches spoke for the bill, the objections recorded went to administrative burden and to the exemption clauses rather than to the right itself, and no division was forced on the bill’s passage in either House. A statute carried without a contested vote, drafted with a movement member inside the advisory room and a parliamentary committee repairing its weak points, is not the usual biography of an Indian regulatory law, and the contrast with the 2002 dead letter, drafted inside the administration and abandoned by it, is the political fact the policy-history framing records without explaining.

One boundary in the enacted text marked the limit of the 2005 settlement’s reach. Section 1(2) extended the Act to the whole of India except the State of Jammu and Kashmir, an exclusion that followed from the state’s special constitutional position and that legislators noted with some irony in debate, since the valley’s own access statute of 2004 was among the instruments the central Act elsewhere displaced. The exclusion stood for fourteen years, until the reorganisation of 2019 closed it, and it is a useful reminder that the settlement’s universality was a political achievement assembled in stages rather than a premise the statute could simply declare.

One drafting decision deserves isolation because it governed everything that followed. The bill’s authors placed the appellate power in new commissions rather than in the administrative tribunals or the departmental hierarchy, and they protected the commissioners with the two devices the Constitution uses for its own watchdogs: a fixed term written into the Act, and a salary equated by the Act to a named constitutional officeholder’s, removable only by the same prescription that protected the Election Commission. The model in the drafters’ minds, visible in the choice of the Election Commission as the salary anchor, was the one Indian institution that had demonstrated, under T. N. Seshan in the 1990s, what a statutory officeholder with security of tenure could do to the behaviour of the political executive. The 2019 amendment’s significance, assessed below, is measurable only against that original choice: the amendment did not misread the Act. It read the Act exactly as its drafters had written it, identified the clauses that made the referees independent, and replaced prescription by statute with prescription by the government of the day.

The Act by Its Text: A Clock, a Duty, a Bounded Refusal

Section 3 of the Act states the right in one line: subject to the provisions of the Act, all citizens shall have the right to information. Section 2(f) defines information to mean any material in any form, including records, documents, memos, emails, opinions, advices, press releases, circulars, orders, logbooks, contracts, reports, papers, samples, models and data held in electronic form, and Section 2(j) defines the right itself to include inspecting works, documents and records, taking notes and certified copies, and taking certified samples of material. The sample clause is the jan sunwai written into law: the applicant may test the road against the road’s record by taking a sample of the road. Section 6 permits an application in writing or electronically, in English, Hindi or the official language of the area, to a designated Public Information Officer, and it provides that the applicant need give no reason for the request and no personal particulars beyond a contact address. The fee prescribed for central public authorities was ten rupees, with applicants below the poverty line exempted from the fee entirely. Section 7 fixes the clock that the whole design hangs on: the officer must supply the information or reject the request within thirty days, within forty-eight hours where the information concerns the life or liberty of a person, and a failure to decide within the period is a deemed refusal that entitles the applicant to the information free of charge on eventual supply. Section 5 obliges every public authority to designate Public Information Officers across its offices, and applicants examining disclosed files in spreadsheet or document form routinely open them with free browser tools such as the Office File Viewer, which reads Excel, Word and PowerPoint records without proprietary software.

Who is a Public Information Officer, and what must the officer do?

The Public Information Officer is the official each public authority must designate in every administrative unit to receive applications and supply information. The officer must assist applicants, decide each request within thirty days, or forty-eight hours where life or liberty is at stake, and answer personally for default, because the penalty falls on the officer, not the office.

Section 4 is the chapter the movement valued most after the clock, because it reverses the default direction of disclosure. Section 4(1)(b) obliges every public authority to publish, on its own initiative and within one hundred and twenty days of the Act’s commencement, seventeen categories of information about itself: its functions and duties, the powers of its officers, its decision-making procedure, the norms it follows, its rules and manuals, the categories of documents it holds, its consultation arrangements, its boards and committees, a directory of its officers, their monthly remuneration, its budget and disbursements, the particulars of subsidy programmes and beneficiaries, the recipients of concessions and permits, and the particulars of its information officers. Section 4(2) directs authorities to provide as much information suo motu to the public at regular intervals through notice boards, newspapers, public announcements and the internet so that the public have minimum resort to the Act itself. The design ambition is explicit in the text: the application is meant to be the exception, the published record the rule. That the ambition was never met is a failure of implementation recorded below; that it stands in the statute at all is the MKSS panchayat-board campaign of the 1990s, the painted wall of budgets and beneficiary lists, promoted into national law.

Section 8 states the exemptions, and the anti-corruption reading turns on their bounds. Section 8(1) lists ten clauses, (a) to (j): information whose disclosure would prejudicially affect the sovereignty and integrity of India, its security, strategic, scientific or economic interests, or its relations with a foreign state; information expressly forbidden by a court or whose disclosure would constitute contempt; information causing a breach of parliamentary privilege; commercial confidence, trade secrets and intellectual property whose disclosure would harm a third party’s competitive position; information held in a fiduciary relationship; information received in confidence from a foreign government; information whose disclosure would endanger life or physical safety or identify a confidential law-enforcement source; information that would impede investigation, apprehension or prosecution; cabinet papers, including the records of deliberations of the Council of Ministers, until the decision is complete; and personal information unconnected with any public activity or interest, or whose disclosure would cause an unwarranted invasion of privacy. Three structural limits cut those clauses down. Section 8(2) lets a public authority disclose even exempt information where the public interest in disclosure outweighs the harm to the protected interest. Section 8(3) sunsets most exemptions at twenty years, so that records of events more than twenty years old are disclosable notwithstanding clauses (a), (c) and (i). And Section 22 gives the Act overriding effect over the Official Secrets Act 1923 and every other inconsistent law, answering the ground on which the Devdungri officials had refused the muster rolls a decade and a half earlier. Section 10 adds the severability rule: where a record contains both exempt and non-exempt matter, the part that can reasonably be severed and disclosed must be disclosed, with the officer recording the reasons for the partial release, so that exemption operates on entries and not on files. Section 6(3) requires an officer who receives an application meant for another authority to transfer it within five days and inform the applicant, answering the state-law practice of the application shuttled between offices until its deadline died of travel. Section 9 protects third-party copyright, Section 11 requires notice to a third party before its confidential information is released, and Section 24 removes the intelligence and security organisations listed in the Second Schedule from the Act’s reach, with one exception that belongs at the centre of the anti-corruption reading: even those organisations must supply information pertaining to allegations of corruption and human rights violations, the first with the approval of the Central Information Commission and the second after a forty-five day process. The exemption architecture is a boundary, and the boundary was drawn by drafters who wrote corruption into the exception to the exception.

Can an applicant inspect a public work itself, or only its papers?

The Act permits both. Section 2(j) defines the right to information to include inspection of works as well as documents and records, and it extends to taking certified samples of material used in a work, so a road, a culvert or a school building can be tested against its own measurement book by the people the work was built for.

Two Tiers of Commissions: The Architecture Above the Officer

The fourth feature closes with the enforcement machine, because a right with a clock needs a court of appeal that is not the office accused. Sections 12 to 17 constitute a Central Information Commission and a State Information Commission in every state, each headed by a Chief Information Commissioner with up to ten Information Commissioners. Commissioners are appointed by the President or the Governor on the recommendation of a committee combining the Prime Minister or Chief Minister, the Leader of the Opposition in the Lok Sabha or the state assembly, and a cabinet minister, a composition designed to put the appointment beyond single-party gift. The first Chief Information Commissioner of India was Wajahat Habibullah, a retired civil servant of the Jammu and Kashmir cadre, who took office when the Central Commission was constituted in October 2005. The Commission’s first benches worked out of borrowed rooms with a registry built from nothing, and their early procedural choices, public hearings as the default, reasoned orders published for the parties and the press, the applicant heard in person or by telephone rather than confined to the paper, set the template the state commissions copied at their constitution across 2005 and 2006. Habibullah’s commission treated its own practice as part of the statute’s meaning, a posture the movement recognised, since the jan sunwai had likewise been a procedure before it was a doctrine. In hearing appeals and complaints the commissions hold the powers of a civil court: they may summon and enforce attendance, compel the production of documents, receive evidence on affidavit and requisition public records. Section 19 builds the ladder the applicant climbs: a first appeal within thirty days to an officer senior to the Public Information Officer inside the same authority, and a second appeal within ninety days to the Commission, whose decision is binding. Section 18 lets a citizen complain directly to the Commission where an authority has failed to appoint its officers, refused an application, missed the deadline, demanded an unreasonable fee or supplied misleading information.

Section 20 is the provision that distinguishes the 2005 Act from every state statute and from the dead letter of 2002, and it deserves statement in its own terms. Where the Commission finds that a Public Information Officer has, without reasonable cause, refused to receive an application, failed to furnish information within the statutory time, denied a request mala fide, knowingly given incorrect, incomplete or misleading information, destroyed information that was the subject of a request, or obstructed its supply in any manner, the Commission shall impose a penalty of two hundred and fifty rupees for each day of default, capped at twenty-five thousand rupees, recoverable from the officer personally, and it may in addition recommend disciplinary action. The penalty is small money with a large design: it converts delay from an institutional tactic into a personal cost, and it is the statutory form of the jan sunwai’s central move, which was to make a named official answer in person for a written record. The original tenure settlement protected the machine’s independence. Under Section 13 as enacted, the Chief Information Commissioner and Information Commissioners held office for five years or until the age of sixty-five, whichever came first, with no reappointment, and their salaries were equated by statute to those of the Chief Election Commissioner and Election Commissioners respectively; Section 16 made the same settlement for the state commissions. Two further provisions complete the machine’s design and are habitually under-read. Section 19(8) arms the commissions, in deciding an appeal, to require the public authority to publish specified information, to change its record-keeping and training practices, and to compensate the complainant for any loss or detriment suffered, which makes the commission a reforming authority over the office’s systems and not only a tribunal over its answers. Section 14 protects the central commissioners in office: removal requires proved misbehaviour or incapacity established after an inquiry by the Supreme Court on a presidential reference, and the intervening sections bar the courts’ jurisdiction over commission orders except by constitutional writ, oblige governments to run public awareness and officer-training programmes under Section 26, and repeal the Freedom of Information Act 2002 in Section 31. The five-feature map below, the findable artifact of this rereading, sets the whole architecture out in citable form: it is labelled in the text here as the Five-Feature Map of the 2005 RTI Architecture, and each row names the feature, its date or instrument, and the institutional content the feature contributed.

Feature	Date or instrument	Institutional content contributed
One: the MKSS movement and the jan sunwai	MKSS founded at Devdungri, 1 May 1990; first jan sunwai, Kot Kirana, 2 December 1994; Beawar dharna, April to May 1996	The public hearing as a working audit of muster rolls, bills and vouchers; the demand that records be opened as a condition of honest wages
Two: the first state statute	Tamil Nadu Right to Information Act (Act 24 of 1997), assent 4 May 1997; Goa Act, 31 July 1997	Proof that a legislature could enact citizen access, and a cautionary model: wide exemptions, appeal to the government itself, no penalty
Three: the state laboratories	Rajasthan and Karnataka 2000; Delhi 2001; Assam and Maharashtra 2002; Madhya Pradesh 2003; Jammu and Kashmir 2004	Nine state statutes in live use; usage concentrated on material entitlements; delay and exemption-stretching documented as the standard administrative response
Four: the central Act	Right to Information Act 2005; passed 11 and 12 May 2005; assent 15 June 2005; in force 12 October 2005	Thirty-day clock and forty-eight-hour rule in Section 7; proactive disclosure in Section 4; bounded exemptions in Section 8; commissions with personal penalty powers in Sections 12 to 20
Five: the trajectory and the counter-attack	CIC order on political parties, 3 June 2013; Supreme Court on the Chief Justice’s office, 13 November 2019; Right to Information (Amendment) Act, 2019	Disclosure extended to party funding and the judiciary’s administrative office; the 2019 amendment replaced fixed five-year tenure and Election Commission salary parity with terms prescribed by the central government

Section 8 in the Hearing Room: The Exemptions Argued Case by Case

A statute’s exemption clauses mean whatever its appellate forum repeatedly holds them to mean, and the commissions’ first decade of hearings gave Section 8 a working shape that neither the information-rights celebration nor the administrative complaint predicted. The fiduciary clause, Section 8(1)(e), was tested earliest and hardest by the government itself in the file notings controversy of 2006: the Department of Personnel and Training’s official guidance for applicants asserted that file notings, the recorded opinions of officials on a file, stood outside the Act, and the Central Information Commission under Habibullah ruled that notings form part of the record of a decision and are disclosable unless a Section 8 clause covers them, after which the department’s guidance was corrected. The episode fixed the commission’s interpretive posture: exemptions were to be read from the text’s ten clauses, not expanded by departmental circular, and the burden of bringing a record inside a clause lay on the authority resisting disclosure. The personal information clause, Section 8(1)(j), received its restrictive landmark in Girish Ramchandra Deshpande v Central Information Commission, where the Supreme Court in 2012 held that the service details of a government employee, increments, postings, assets and liabilities among them, were personal information whose disclosure had no relationship to public activity, absent a demonstrated larger public interest, a holding the commissions applied widely and the privacy bench’s proportionality method in the Puttaswamy judgment of 2017 later reframed without displacing. The commercial confidence clause was read against the discloser of last resort in the spectrum and mining matters, where authorities’ claims that bids, pricing notes and correspondence were commercially confidential failed wherever the record showed the disposal of a public resource. The cabinet papers clause was held to its textual limit: the exemption covers the deliberations while a decision is incomplete, and the records of a completed decision, with the material on which it rested, open to the applicant. The hearing room’s cumulative doctrine, visible across the Commission’s annual reports and its published decisions, is that Section 8 operates as a set of pleaded defences in an adversarial process, each to be proved by the office, and not as a standing classification of topics. That is the jan sunwai’s burden of proof promoted into administrative law: the record is public unless the custodian shows cause, in the open, why this entry must stay shut.

Disclosure in Action: What the Files Showed, 2005 to 2019

The fifth feature is the trajectory, and it opens with the uses that made the Act’s reputation. The employment guarantee programme enacted in 2005, the Mahatma Gandhi National Rural Employment Guarantee Act, wrote the jan sunwai into its own statute as the social audit, and social audit units and village assemblies across Andhra Pradesh, Rajasthan and other states read muster rolls aloud in public hearings on the MKSS pattern, recovering wages and striking ghost names from the rolls; the genealogy from Kot Kirana to a statutory audit inside the world’s largest public works programme is direct and documented in the programme’s own audit manuals. The social audit line deserves its own tracing, because it is the point at which the movement’s method became the state’s machinery. In October 2006 the first large social audits of the employment guarantee programme were conducted in Dungarpur district, Rajasthan, with MKSS veterans and the economist Jean Dreze among the participants, reading job cards and muster rolls aloud in village assemblies on the jan sunwai pattern and recovering wages that the rolls had diverted. Andhra Pradesh institutionalised the method from 2009 in a dedicated Society for Social Audit, Accountability and Transparency, which fielded district audit teams and public hearings across the state, and the programme’s audit of accounts rules later wrote the public hearing into the national framework. The genealogy is documented in the programme’s own manuals, and it closes the first feature’s arc: an audit form invented to confront a panchayat secretary in Pali district in 1994 became, within fifteen years, a funded line in the administrative budget of the largest public works programme on earth.

Beyond the worksite, applicants used the Act against the paper of large scandals. Records obtained under the Act fed the exposure of the Adarsh Housing Society allotments in Mumbai, in which flats in a Colaba tower sanctioned for war widows and defence personnel were allotted to politicians, officials and their relatives, a disclosure chain that contributed to the resignation of Maharashtra Chief Minister Ashok Chavan in November 2010. Information obtained under the Act on the allocation of second-generation spectrum licences and on the Commonwealth Games contracts of 2010 entered parliamentary debate, court proceedings and the reports of the Comptroller and Auditor General; the attribution in each case belongs to named applicants and named files, and journalists including the applicants who pursued the Prime Minister’s Office records on spectrum pricing documented their applications in print. Usage volumes establish the scale of the instrument: applications to central public authorities grew from the tens of thousands recorded in the Act’s first partial year to several lakhs a year within the decade, and the Central Information Commission’s annual reports record well over a million applications a year reaching central public authorities alone in the Act’s second decade, a figure no state statute approached, and the Commission’s own reporting attributes the bulk of applications to grievances about entitlements, land, employment and pensions rather than to research or journalism, replicating at national scale the pattern the state laboratories had shown.

The trajectory’s other direction of travel ran through the Second Schedule, and its central exhibit is the notification of June 2011 by which the central government added the Central Bureau of Investigation, the National Investigation Agency and the National Intelligence Grid to the Schedule, removing the country’s premier investigative agency from the Act’s ordinary reach at the height of the spectrum and Commonwealth Games investigations. The transparency campaign’s response, led by the NCPRI and taken up in Parliament, turned on the proviso the drafters of 2005 had written into Section 24: the exemption does not cover information pertaining to allegations of corruption, which the listed organisation must still supply with the Central Information Commission’s approval, so the notification changed the channel of the CBI’s accountability without closing it, and applicants continued to extract the agency’s administrative and corruption-related records through the proviso’s gate. The episode is the exemption architecture working exactly as the anti-corruption reading describes: the boundary of secrecy was contested at the point of corruption, by name, and the contest was fought over the proviso rather than over the right, because the proviso was where the theft investigations lived.

Disclosure then reached the two institutions that most resisted it, and the sequence belongs to the record. On 3 June 2013 a full bench of the Central Information Commission, Chief Information Commissioner Satyananda Mishra with Information Commissioners M. L. Sharma and Annapurna Dixit, held in the complaints brought by the Association for Democratic Reforms and the applicant Subhash Chandra Agrawal that six national parties, the Indian National Congress, the Bharatiya Janata Party, the Communist Party of India (Marxist), the Communist Party of India, the Nationalist Congress Party and the Bahujan Samaj Party, were public authorities under Section 2(h), substantially financed indirectly by the central government through concessional land allotments in the capital, government accommodation at concessional rents, income tax exemption under Section 13A of the Income Tax Act and free airtime on state media during elections. The bench directed the six parties to designate their own information officers within six weeks and to answer the pending applications within four. None of the six complied. The six parties, adversaries on nearly every other question before the country, presented a common front against the order, arguing in statements and in Parliament that their internal functioning was already regulated by the Representation of the People Act and the Income Tax Act and that the Commission had misread the financing clause. The government of the day introduced a bill in the Lok Sabha in August 2013 to amend the Act and remove political parties from its scope; after sustained campaign opposition the bill was referred to a standing committee rather than pressed to a vote, and it lapsed with the Lok Sabha in 2014, and the Commission itself recorded in March 2015 that it lacked the machinery to enforce its order against parties that declined to appear. The episode cuts both ways in the adjudication below: the architecture reached party finance, the citadel of Indian political opacity, and the architecture’s own enforcement ceiling was exposed in the same act.

The judiciary’s turn closed the sequence. On 13 November 2019 a five-judge Constitution Bench of the Supreme Court, presided over by Chief Justice Ranjan Gogoi and including Justices N. V. Ramana, D. Y. Chandrachud, Deepak Gupta and Sanjiv Khanna, held in Central Public Information Officer, Supreme Court of India v Subhash Chandra Agarwal that the office of the Chief Justice of India is a public authority under the Act, upholding the Delhi High Court’s Full Bench judgment of 12 January 2010. The bench held that the exemption for information held in a fiduciary capacity did not cover the asset declarations judges had made to the Chief Justice under the Court’s own 1997 resolution, and it remitted the collegium correspondence questions to the Court’s information officer for decision under the public interest test and the third-party notice procedure. The judgment is the apex of the information-rights lineage that runs from Raj Narain in 1975 through S. P. Gupta v Union of India in 1981, where the Supreme Court located the citizen’s right to know in Article 19(1)(a) and read down the class of documents entitled to secrecy, to the 2019 bench: the courts built the right’s doctrine, and the 2005 Act built its machinery, and in 2019 the machinery was applied to the doctrine’s own house. The same year completed the Act’s territorial reach. The Jammu and Kashmir Reorganisation Act 2019, effective 31 October 2019, extended the central Right to Information Act to the new Union Territories of Jammu and Kashmir and Ladakh and repealed the Jammu and Kashmir Right to Information Act 2009, ending the exception that had stood since 2005, when the central Act was the rare Indian statute that did not extend to the state whose own access law it had left in place. The state commission at Srinagar gave way to the central commission’s jurisdiction, and the applicants of the valley entered the same appellate queue as the applicants of every other state, a unification of the transparency map that proceeded from the constitutional changes of August 2019 and that the information commissions’ own reporting registered without commentary on its origins.

A disclosure regime that reaches the Chief Justice’s office and the national parties’ funding has a fair claim to be the deepest transparency settlement in the series this article belongs to, deeper in reach than the most-consequential-constitutional-moment framework suggests for judgments alone, because the RTI settlement operates daily, at the level of the tehsil office and the works register, without waiting for a Constitution Bench; and it was assembled, clause by clause, in full view of the domestic-crisis-through-trust-vote framework that governs how Indian governments count the cost of confrontations they cannot afford to lose.

Where does a rejected applicant carry the case after the first appeal?

The rejected applicant carries the case to the information commission for the jurisdiction: the Central Information Commission for central authorities and the State Information Commission for state authorities. The second appeal must be filed within ninety days of the first appellate decision, the commission hears it with the powers of a civil court, and its decision binds the public authority.

The Confluence of 2011: From the File to the Fast

The brief’s fifth feature assigns the post-2011 anti-corruption mobilisation to this article’s trajectory, and the assignment is historically exact, because the Anna Hazare movement of 2011 grew out of the transparency movement’s soil, personnel and evidentiary habits. Arvind Kejriwal, a former revenue service officer whose Delhi organisation Parivartan had used the Delhi Right to Information Act of 2001 against ration and works records in the resettlement colonies of east Delhi, received the Ramon Magsaysay Award in 2006 in recognition of that information work, and he carried the method into the drafting circle of India Against Corruption. When Hazare began his fast at Jantar Mantar on 5 April 2011 demanding a Jan Lokpal, an ombudsman with jurisdiction over the political executive, the movement’s exhibits were the files the disclosure regime had already produced: the spectrum allotments, the Commonwealth Games contracts, the Adarsh allotments, each documented in records obtained under the Act or in audit reports the Act’s climate had made politically usable. The fasts of April and August 2011 forced a parliamentary resolution on the Lokpal’s principles, and the Lokpal and Lokayuktas Act was enacted in December 2013, though the first Lokpal of India, Justice Pinaki Chandra Ghose, took office only in March 2019. The confluence matters to the adjudication for a reason neither the celebration nor the critique of 2011 states plainly. The Hazare mobilisation is usually read as the disclosure regime’s successor, the street succeeding the statute. The record reads better as the regime’s dividend: a decade of applicants had normalised the premise that public money leaves a public trail, and the 2011 crowd’s demand for an ombudsman was the demand of people who had read the trail and wanted a prosecutor at its end. The transparency movement’s own elders were divided about the fast and its draft, Roy among those who questioned the Jan Lokpal bill’s design in public while defending the mobilisation’s right to press it, and that division, between the movement that had built an audit and the movement that wanted a jailer, is the honest form of the 2011 chapter.

The Counter-Case at Its Strongest: Obstruction, Misuse and Risk

The evenhandedness this series requires means the case against the disclosure regime is stated by its holders, at full strength, before the adjudication. The administrative case was made earliest and most candidly from inside the government that had enacted the Act. In August 2006 the Union cabinet approved a proposal to amend the Act to exempt file notings, the recorded reasoning of officials on government files, from disclosure, together with material on cabinet deliberations and examinations; the proposal met a national campaign of opposition from the NCPRI and the press, the Prime Minister’s Office announced that the amendment would not be introduced in that session, and the proposal was shelved, revived in altered forms in later years and never enacted. Its proponents’ argument, stated in the government’s own briefings of the period, was that officials would cease to record frank advice if the advice could be read by the applicant it disappointed, and that the Act as drafted exposed the deliberative process itself rather than its outcomes. Six years later the maker of the Act stated the second half of the case. Addressing the annual convention of the Central Information Commission in October 2012, Prime Minister Manmohan Singh said that the right to information should not be allowed to become an instrument for obstructing the decision-making process, expressed concern at frivolous and vexatious applications, and raised the privacy of individuals and the discouragement of candid internal discussion as live questions for the regime. The statement matters because of its source: the head of the government that had carried the Act, in the Act’s own house, seven years into its operation.

The practitioner’s counter-case, stated by serving and retired officials in the essays of Prashant Bhushan’s counterparts across the debate and in the commission literature, adds three operational counts. Applications are used, in documented instances, to settle private scores, to extract information for commercial competitors under the cover of citizenship, and to burden small offices with volume requests disconnected from any public purpose, while the same offices lack the record-management systems the Act’s Section 4 presupposes, since the proactive disclosure duty was honoured late or not at all by a large share of public authorities. The commissions’ own transparency audits, commissioned by the Central Commission in the Act’s second decade, found that a substantial fraction of central public authorities had not completed the third-party audit of their Section 4 disclosures that the Commission’s directions required. The third count is safety. Applicants and activists who used the Act against land, mining, liquor and construction interests were assaulted and in named cases killed: Shehla Masood, shot in Bhopal on 16 August 2011 after using the Act against local interests; Amit Jethwa, shot near the Gujarat High Court in July 2010 after litigation and applications on illegal mining around the Gir forest. The roll of harmed applicants maintained by the Commonwealth Human Rights Initiative runs to scores of assaults and a number of murders debated in Parliament itself. Whatever the Act’s design intended, its use imposed risks on the user that its text allocates nowhere, and the Whistle Blowers Protection Act passed in 2014 was not brought into force in an amended and operative form during the period this article covers. The executive’s narrower attempts at containment belong in the same section of the record. Draft Right to Information Rules circulated by the Department of Personnel and Training in 2010 proposed to confine each application to a single subject and to a word limit, and to restrict the appeal process in ways the campaign read as a filter on volume applicants; the proposals were published for comment, met the NCPRI’s organised opposition and unfavourable press assessment, and were not carried into the rules as finally notified in 2012. The pattern across 2006, 2010 and 2019 is consistent enough to name: governments of both coalitions, when they moved against the regime, moved against its operating conditions, the noting, the word count, the tenure, and left the proclaimed right standing, because the right cost nothing to proclaim and the conditions were where the disclosure happened. A rereading that omitted those counts would be advocacy, not adjudication; they stand in the record, and the claim defended here must survive them rather than ignore them.

The Price of Asking: Violence Against the Applicant

One cost of the regime falls outside the counter-case’s administrative terms and must be recorded separately, because its victims used the Act exactly as the design intended. Satish Shetty, a land rights campaigner in Talegaon near Pune who had used information applications and litigation against land acquisition and highway compensation irregularities in Maharashtra, was murdered on 13 January 2010 while on a morning walk, in a case that passed to the Central Bureau of Investigation and became a parliamentary reference point for applicant safety. Amit Jethwa, who had filed public interest litigation and information applications against illegal mining in the Gir forest region of Gujarat, was shot dead outside the Gujarat High Court complex in Ahmedabad in July 2010. Shehla Masood, a Bhopal campaigner who had pursued applications against local administrative and political interests in Madhya Pradesh, was shot outside her home on 16 August 2011. The Commonwealth Human Rights Initiative’s documentation of attacks on information users, maintained across the Act’s second decade, records scores of assaults, harassments and killings of applicants and campaigners across states, the pattern concentrating where applications touched land, mining, liquor and construction, the four economies in which a paper record is most often the only proof of a theft. The Whistle Blowers Protection Act, enacted in 2014 after the Anna Hazare movement’s pressure, was not brought into operative force with its protective machinery during the period covered here, and the commissions possess no power to protect an applicant, only to penalise an officer. The anti-corruption reading must carry this entry on its own ledger: the Act distributed the risks of accusation to the accuser, and the state’s settlement of 2005 built no shelter into the distribution. That the applications continued to arrive in their millions, in the Commission’s annual counts, is a fact about the applicants, not about the adequacy of the design.

The Amendment of 2019: Tenure by Prescription

The counter-attack, when it came, was aimed at the machine rather than the right, and its form is the strongest evidence for this article’s claim. The Right to Information (Amendment) Bill was introduced in the Lok Sabha on 19 July 2019, passed that House on 22 July 2019 against opposition protests, passed the Rajya Sabha on 25 July 2019 after the government mustered support from regional parties whose votes the opposition had counted on, and received presidential assent on 1 August 2019. The Amendment Act changed three things, all inside Sections 13 and 16. It replaced the fixed term of five years for the Chief Information Commissioner, Information Commissioners and their state counterparts with a term as may be prescribed by the central government. It replaced the statutory salary parity with the Chief Election Commissioner and Election Commissioners with salaries and service conditions as prescribed by the central government, extending the centre’s prescribing power over the state commissions’ members as well. And it deleted the provision that had deducted from a commissioner’s salary any pension the commissioner drew for prior government service. The rules made under the amendment, notified in October 2019, prescribed a term of three years for the central and state commissioners and fixed the Chief Information Commissioner’s salary at two lakh fifty thousand rupees a month and an Information Commissioner’s at two lakh twenty-five thousand.

The parliamentary contest stated both cases at full strength, and the record of July 2019 gives each side its best form. The government’s case, moved by the Minister of State for Personnel, Jitendra Singh, was institutional tidiness: the Election Commission is a constitutional body created by Article 324, the information commissions are statutory bodies created by an Act, and equating their service conditions confused two categories of institution; the amendment rationalised an anomaly, touched neither the Act’s disclosure provisions nor the commissions’ powers, and left appointments to the same statutory committee as before. The opposition’s case, stated on the floor by members including Shashi Tharoor of the Congress and Derek O’Brien of the Trinamool Congress and outside Parliament by former commissioners in a public letter, was that independence held by prescription is independence held at pleasure: a commissioner deciding appeals against the central government, on a three-year term the central government prescribes and a salary the central government fixes, with the state commissions’ terms also prescribed from Delhi in a federal system, would price the government’s displeasure into every contested order, and the deletion of the pension offset advertised the posts to retired officials dependent on the government’s goodwill. Former Chief Information Commissioner Wajahat Habibullah and former Information Commissioner Shailesh Gandhi were among the signatories and speakers who carried that argument into print, and the debate’s federal count, that the centre had taken power to prescribe service conditions for state functionaries, was pressed by state governments of the opposition. The adjudication of this contest belongs to the section after next; the structural fact belongs here. Parliament in 2019 did not narrow Section 3, did not widen Section 8, did not touch the thirty-day clock or the penalty. It changed who guards the guardians’ tenure, which is precisely the point a government chooses when its objection is not to disclosure in principle but to an enforcement machine it does not control. The contrast with the coalition-era settlement, in which disclosure was one of the governing programme’s advertised reforms, is legible against the integrated three-tensions framework that maps how the post-2014 state re-settled the terms of institutional independence across domains.

Vacancies and Backlog: The Appointment Record

The machine’s other vulnerability required no legislation at all, and its record is measured by the regime’s own auditors. The Satark Nagrik Sangathan’s Report Card on the Performance of Information Commissions in India, compiled annually from data the commissions themselves supply, recorded in its 2019 edition that appeals and complaints pending before the commissions had crossed two lakh, and its 2024-25 edition recorded 4,13,972 appeals and complaints pending across the twenty-nine information commissions as on 30 June 2025, with Maharashtra’s state commission carrying the largest backlog at over ninety-five thousand. The same report card series documented commissions working below sanctioned strength for long stretches, state commissions in Jharkhand and Himachal Pradesh defunct for periods during 2024 and 2025 for want of any appointed commissioner, and estimated disposal horizons, at prevailing clearance rates, running to years and in one state commission’s case to decades for a fresh second appeal. The Central Commission itself went headless for stretches, including the months between Chief Information Commissioner Sushma Singh’s retirement at the end of December 2013 and Rajiv Mathur’s appointment in May 2014, and again from September 2025 in the report card’s account. These figures are past this article’s locked date and are included under the brief’s fifth feature, which extends the trajectory past 2019; they are the appointment record of the post-amendment settlement, and they are attributed here to the report card that compiled them rather than asserted as this article’s own count.

The statute anticipated its own audit and then watched the audit be neglected. Section 25 requires the Central and State Information Commissions to prepare an annual report on the implementation of the Act, covering the requests received by each public authority, the appeals decided, the penalties imposed and the reforms recommended, and requires the governments to lay the report before Parliament and the state legislatures. The report card literature draws its numbers from those reports and from the commissions’ replies to information applications about themselves, and it records a secondary failure inside the primary one: annual reports tabled late, in some jurisdictions years late, so that the legislature charged with watching the machine received the machine’s account after the account had ceased to describe it. The transparency statute’s own transparency loop, in other words, ran on the same appointment-and-delay economy as the commissions it reported on, a recursion the report cards document year after year without any government of either coalition moving to break it.

The judiciary had already named the mechanism. In Anjali Bhardwaj v Union of India, decided on 15 February 2019, the Supreme Court directed the central and state governments to fill commission vacancies on time, to begin the selection process one to two months before a vacancy arose, to publish the criteria and the applications received on official websites, and to complete appointments for the Central Commission vacancies then pending, observing that the commissions are the institutions envisaged by the Act to ensure its effective implementation and that leaving them under strength defeats the statute. The petitioners in that case were transparency campaigners including the NCPRI’s Anjali Bhardwaj, and the judgment stands as the courts’ contribution to the trajectory: the doctrine expanded the right in 2019, in the Chief Justice’s office case, while the administration of the right contracted through the appointment diary. Read with the amendment of the same year, the 2019 record is double-sided in a way neither the celebration nor the obituary of the Act captures: the right grew at its apex and thinned at its base in the same twelve months, and which of those movements defines the regime is a question about where the Act’s force actually lives, in its doctrine or in its machinery.

Three Readings, One Statute: The Adjudication

The information-rights framing, held by the global freedom of information literature and by the Act’s own preamble, reads the statute as India’s entry into a family of over a hundred national access laws, its significance lying in the right’s universality: any citizen, any record, no reason required. The policy-history framing, held by the legislative historians, reads it as the product of a drafting sequence, Press Council model to Shourie draft to the dead letter of 2002 to the National Advisory Council’s revisions, its significance lying in the improvement of the text across drafts. Both readings capture something real, and both fail the same three exhibits. Neither explains why the demand for the law arose from a wage dispute rather than from a newsroom or a law school: the information-rights family has members drafted by ministries, and none of the others began with labourers reading muster rolls aloud. Neither explains the statute’s particular load-bearing choices, the personal penalty, the corruption exception inside the security exemption, the sample of road material inside the definition of the right, which are not standard items in the global template and are direct transcriptions of jan sunwai practice. And neither explains the politics of 2019, in which a government content to leave the right’s text untouched legislated instead against the tenure of its referees, a choice that makes sense only if the Act’s operative force was always understood, by its friends and its opponents alike, to lie in its use as an instrument against the misuse of public money and public office.

The holders of the rival readings can be named, as this series requires. The information-rights framing’s standard scholarly form is the comparative survey tradition represented by Toby Mendel’s Freedom of Information: A Comparative Legal Survey, published for UNESCO, which scores the Indian Act within a global family of access laws; the global rating exercises conducted by the Centre for Law and Democracy with Access Info Europe placed the Indian statute at or near the top of their index of legal frameworks in the years after 2011, a verdict on the text that the framing treats as the measure of the regime. Alasdair Roberts, the scholar of government secrecy who examined the Act’s first years in Public Administration Review in 2010, asked whether the law was great and revolutionary and answered with the administrator’s caution, finding the implementation machinery already straining under the caseload the right had generated; his essay is the policy-history framing’s strongest scholarly form, attentive to drafts, forms and disposal rates. Both accounts are honoured in the adjudication that follows, and both, it is argued there, mistake the part for the whole: Mendel’s index scores the promise, Roberts’s caution scores the queue, and the politics of the Act lives in the relation between the two, in a promise written by a movement for use against theft and a queue lengthened by every institution the promise threatened.

A reading that cannot state its own defeat conditions is a mood rather than an argument, so the conditions belong on the record. The anti-corruption reading would fail if the Act’s enemies had behaved as the information-rights framing predicts, treating the statute as a neutral access utility to be tuned for efficiency: they did not, and across three decades of attempts, in 2006, in 2010, in the parties’ common front of 2013 and in the amendment of 2019, the pressure fell on the deliberative file, the volume applicant, the funding record and the referee’s tenure, never on the reading room’s opening hours. The reading would fail if the usage record resembled the policy-history framing’s implied user, a researcher assembling a legislative past: the commissions’ own reporting shows the docket dominated by entitlements, land, employment and service grievances, the jan sunwai’s agenda at national scale. And the reading would fail if the machinery’s distinctive clauses, the personal penalty, the severability rule, the corruption exception inside the security exemption, could be found in the access statutes of countries whose laws were drafted by ministries without a movement at the door; the comparative surveys the information-rights framing relies on record those clauses as Indian particularities. The claim therefore stands on evidence that could have been otherwise and was not.

The anti-corruption-political reading defended here absorbs the other two rather than denying them. The complication stated in the brief is genuine: much of the Act’s daily operation is informational in character, Section 4’s publication duties serve planning and research as much as accusation, and the commissions’ dockets are thick with service grievances that name no scandal. The reading does not deny that content; it explains its origin and its distribution. The Act serves those informational ends through machinery designed for an accusatorial purpose, in the way a plough designed for one field still turns others, and the proof is in the record of what the machinery’s enemies attack: not the right, not the exemptions, but the commissioners’ tenure and the commissions’ strength. The five features therefore stand as the citable anatomy: a movement that defined corruption as a false entry in a public record; a first state statute that showed a law without teeth decays; a decade of state laboratories that documented the administrative counter-moves; a central Act that answered each counter-move with a clause, the clock against delay, the penalty against the officer, Section 4 against the locked file room, the commissions against the departmental appeal; and a trajectory in which the settlement was extended by courts and commissions to parties and to the Chief Justice’s office, then narrowed at the point of the machine’s independence by the amendment and the appointment diary of 2019 and after. That is how the 2005 RTI operates as anti-corruption politics, and it is why the settlement belongs in the economic-policy-as-politics account of the period, alongside the detailed treatment in the detailed-political-architectural-substantive-content framework of how the 1991 reforms were assembled: in both cases the durable politics is in the machinery the moment built, and in both cases later governments reveal their true settlement by what they do to the machinery when nobody is watching the preamble.

Frequently Asked Questions

Q: When was the RTI Act 2005 enacted?

The Right to Information Act was passed by the Lok Sabha on 11 May 2005 and by the Rajya Sabha on 12 May 2005, received the assent of President A. P. J. Abdul Kalam on 15 June 2005, and came fully into force on 12 October 2005, the 120th day fixed by its own commencement clause. It replaced the Freedom of Information Act 2002, which had never been notified into operation. The Act extended to the whole of India except the then state of Jammu and Kashmir, which kept its own statute until the central Act was extended after 2019.

Q: Who was Aruna Roy in the RTI movement?

Aruna Roy was a co-founder of the Mazdoor Kisan Shakti Sangathan, established at Devdungri in Rajsamand district, Rajasthan, on 1 May 1990, and the most recognised public face of the campaign that produced the 2005 Act. A former officer of the Indian Administrative Service, which she left in 1974, she had worked at the Social Work and Research Centre at Tilonia before settling at Devdungri in 1987. She led the jan sunwai public hearings of 1994 and 1995, the forty-day Beawar dharna of 1996, and the National Campaign for People’s Right to Information, and she later sat on the National Advisory Council that shaped the 2004 bill. Her account, The RTI Story: Power to the People, appeared in 2018.

Q: What is MKSS in the RTI history?

The MKSS, the Mazdoor Kisan Shakti Sangathan or Organisation for the Empowerment of Workers and Peasants, is the Rajasthan movement organisation founded by Aruna Roy, Nikhil Dey, Shankar Singh and village colleagues at Devdungri on 1 May 1990. It began in minimum wage struggles on drought-relief works, discovered that wages were stolen through false muster rolls, and converted the discovery into the jan sunwai, the public hearing at which official records were read aloud and verified by the workers named in them. The MKSS carried the demand from village hearings to the Beawar dharna of 1996 and into the national campaign, and its practices supplied the 2005 Act’s defining machinery of deadlines, penalties and public audit.

Q: What was the 1997 Tamil Nadu Right to Information ordinance?

Tamil Nadu’s access law of 1997, debated through an ordinance phase before enactment, became the Tamil Nadu Right to Information Act, numbered Act 24 of 1997, which received the Governor’s assent on 4 May 1997 under the Dravida Munnetra Kazhagam government of M. Karunanidhi. It was the first right to information statute enacted by any Indian state. Campaigners treated it as a warning rather than a model: its exemptions were drawn widely, its appeal lay to the government itself rather than to an independent body, and it carried no penalty for an officer who failed to answer, so the right it granted decayed quickly in administrative hands.

Q: How did the RTI operate as anti-corruption politics?

The Act operated as anti-corruption politics because its machinery was designed against the documented techniques of petty and grand theft. The thirty-day clock and the deemed refusal rule attacked delay, the standard burial ground of inconvenient requests. The personal penalty in Section 20 made a named officer, rather than an office, pay for default. Section 4 forced the routine record into the open before anyone asked, shrinking the file room in which false entries hid. Social audits under the employment guarantee programme read muster rolls aloud on the jan sunwai pattern, and applicants used disclosed records against land, spectrum and housing allotments, so the statute functioned as a standing accusation mechanism rather than a reading room.

Q: What is the post-2005 RTI institutional architecture?

The architecture has three storeys. At the base, every public authority designates Public Information Officers, and Assistant Public Information Officers in sub-district offices, to receive applications and supply information within thirty days, or forty-eight hours where life or liberty is at stake. Above them sits a first appellate authority, a senior officer inside the same public authority, who hears appeals within thirty days. At the top stand the Central Information Commission and the State Information Commissions, independent bodies of a Chief Information Commissioner and up to ten commissioners, which hear second appeals and complaints with the powers of a civil court, issue binding decisions, and impose the Act’s penalties.

Q: What is Section 8 of the RTI Act?

Section 8 lists the information exempt from disclosure, in ten clauses covering the sovereignty, security and strategic interests of India and relations with foreign states, court-forbidden material and contempt, parliamentary privilege, commercial confidence and intellectual property, fiduciary information, confidences from foreign governments, the safety of persons and confidential law-enforcement sources, material that would impede investigation or prosecution, cabinet papers until a decision is complete, and personal information without public interest. Three limits cut the list down: Section 8(2) permits disclosure where public interest outweighs the harm, Section 8(3) ends most exemptions for records over twenty years old, and Section 22 overrides the Official Secrets Act 1923.

Q: What did Section 4 of the RTI Act oblige public authorities to publish on their own initiative?

Section 4(1)(b) obliged every public authority to publish seventeen categories of information about itself within one hundred and twenty days of the Act’s commencement: its functions and duties, the powers and duties of its officers, its decision-making channels, the norms for its work, its rules, regulations and manuals, the categories of documents it holds, its public consultation arrangements, its boards and committees, a directory of officers with their monthly remuneration, its budget and expenditure reports, the details of subsidy programmes and their beneficiaries, the recipients of permits and concessions, and the names of its information officers. Section 4(2) required the publication to be updated and pushed outward through notice boards and other media so that citizens would need the application route as little as possible.

Q: Who was the first Chief Information Commissioner of India?

Wajahat Habibullah, a retired officer of the Indian Administrative Service from the Jammu and Kashmir cadre and a former secretary in the central government, became the first Chief Information Commissioner when the Central Information Commission was constituted in October 2005. He led the young commission through its formative years, building its hearing practice, its annual reporting to Parliament and its reading of the exemption clauses, and he served until 2010. After the 2019 amendment changed commissioners’ tenure and salary to terms prescribed by the central government, Habibullah was among the former commissioners who argued in public that the amendment weakened the independence the original settlement had protected.

Q: What penalties could an information commission impose on a defaulting Public Information Officer?

Under Section 20, where a commission found that a Public Information Officer had refused an application without reasonable cause, missed the statutory deadline, denied a request mala fide, knowingly supplied incorrect, incomplete or misleading information, destroyed the information sought, or obstructed its supply, the commission was required to impose a penalty of two hundred and fifty rupees for every day of default, up to a maximum of twenty-five thousand rupees. The penalty fell on the officer personally rather than on the department, the officer had to be given a hearing first, and the commission could in addition recommend disciplinary action under the officer’s service rules.

Q: What was the Freedom of Information Act 2002, and why did it never operate?

The Freedom of Information Act 2002 was the central government’s first access statute, passed by Parliament in December 2002 and given presidential assent in January 2003, descended from a weakened reworking of the H. D. Shourie committee draft of 1997. The campaign rejected it as a dead letter by design: its exemptions were wide, it provided no independent appellate commission and no penalty for default, and its commencement was left to a government notification. The notification was never issued, so the Act never operated for a single day, and the Right to Information Act 2005 repealed it on coming into force, replacing a promised right with an enforceable one.

Q: What role did the National Campaign for People’s Right to Information play in the road to the 2005 Act?

The National Campaign for People’s Right to Information, founded in Delhi in 1996 by the MKSS together with journalists, lawyers, retired civil servants and academics, was the bridge between the Rajasthan village hearings and the statute book. It supported grassroots struggles for records across states, drafted and refined model legislation through the Press Council NIRD line of drafts, assessed the state laws in operation, campaigned against the weak Freedom of Information Act 2002, and pressed the movement’s four tests, a short clock, an independent commission, a personal penalty and proactive disclosure, into the National Advisory Council’s revisions of 2004, from which the enacted bill took its machinery.

Q: How did the Beawar dharna of 1996 change the scale of the information movement?

The Beawar dharna, a forty-day sit-in at Chang Gate in Beawar town beginning on 5 April 1996, converted a state wage campaign into a national political question. Its largest contingent was rural women from the Bhim region demanding neither relief nor a wage rise but the legal right to inspect public accounts, a year after Chief Minister Bhairon Singh Shekhawat’s announced opening of development records had gone unimplemented. National editors, including Prabhash Joshi of Jansatta, carried the hearings’ slogan, we shall know, we shall live, into print across India. Within the year the movement had founded the National Campaign for People’s Right to Information and the demand had a national platform.

Q: What did the Central Information Commission decide about political parties in June 2013?

On 3 June 2013 a full bench of the Central Information Commission, Chief Information Commissioner Satyananda Mishra with M. L. Sharma and Annapurna Dixit, held in the complaints of the Association for Democratic Reforms and Subhash Chandra Agrawal that six national parties, the Congress, the Bharatiya Janata Party, the Communist Party of India (Marxist), the Communist Party of India, the Nationalist Congress Party and the Bahujan Samaj Party, were public authorities under Section 2(h), being substantially financed indirectly through concessional land, government accommodation, income tax exemption and free election airtime. The bench directed the parties to appoint information officers within six weeks. None complied, and the order remained unenforced.

Q: How did the Supreme Court treat the office of the Chief Justice of India under the RTI Act in 2019?

In Central Public Information Officer, Supreme Court of India v Subhash Chandra Agarwal, decided on 13 November 2019, a five-judge Constitution Bench presided over by Chief Justice Ranjan Gogoi held that the office of the Chief Justice of India is a public authority under the Act, upholding the Delhi High Court’s Full Bench judgment of January 2010. The bench ruled that judges’ asset declarations made to the Chief Justice under the Court’s 1997 resolution were not held in a fiduciary capacity exempt under Section 8, and it sent the questions on collegium correspondence back to the Court’s information officer for decision under the public interest test with notice to third parties.

Q: What did the Supreme Court direct on filling information commission vacancies in February 2019?

In Anjali Bhardwaj v Union of India, decided on 15 February 2019, the Supreme Court directed the central and state governments to fill vacancies in the information commissions on time, to begin each selection process one to two months before a vacancy was expected, to publish the appointment criteria and the particulars of applicants on official websites, and to complete the appointments then pending before the Central Information Commission. The Court reasoned that the commissions are the institutions through which the Act is made effective, and that leaving them below strength or headless defeats the statute the appointments exist to serve.

Q: Which organisations does the Second Schedule place outside the Act, and what exception survives for them?

Section 24 removes the intelligence and security organisations listed in the Second Schedule, among them the Intelligence Bureau, the Research and Analysis Wing, the Directorate of Revenue Intelligence, the Central Economic Intelligence Bureau, the Directorate of Enforcement, the Narcotics Control Bureau, the Border Security Force, the Central Reserve Police Force, the Indo-Tibetan Border Police, the Central Industrial Security Force, the National Security Guards and the Special Frontier Force, together with bodies notified by state governments. One exception survives the removal: information pertaining to allegations of corruption and human rights violations is not excluded, corruption information being supplied with the Central Information Commission’s approval and human rights information through a forty-five day procedure.

Q: What fee did the Act set for an application, and who paid nothing?

For central public authorities the application fee was prescribed at ten rupees, payable with the request, with copying charges for the pages supplied set by the rules. Applicants below the poverty line paid no fee at all and received the information free, a provision carried from the movement’s insistence that the price of a record must never exceed the means of the wage worker whose name stood in it. If the authority missed the statutory deadline, the information had to be supplied free of charge in any event, so the cost of official delay fell on the office rather than the applicant.

Q: How did the courts ground the right to information in Article 19(1)(a) before 2005?

The Supreme Court built the right a generation before Parliament legislated. In State of Uttar Pradesh v Raj Narain in 1975, Justice K. K. Mathew wrote that the people have a right to know every public act done in a public way by their public functionaries. In S. P. Gupta v Union of India in 1981 the Court held that the concept of an open government is the direct result of the right to know implicit in the freedom of speech and expression in Article 19(1)(a), and read the Official Secrets Act’s secrecy claims down accordingly. Later benches applied the doctrine to voters’ knowledge of candidates’ antecedents, in the Association for Democratic Reforms case of 2002 and the People’s Union for Civil Liberties case of 2003.

Q: What did the Satark Nagrik Sangathan report cards record about appeals pending before the commissions?

The Satark Nagrik Sangathan’s Report Card on the Performance of Information Commissions in India, compiled from the commissions’ own published data, recorded pendency climbing past two lakh appeals and complaints in its 2019 edition and reaching 4,13,972 pending matters across the twenty-nine information commissions as on 30 June 2025 in its 2024-25 edition, Maharashtra’s state commission holding the largest single backlog at over ninety-five thousand. The series also documented commissions working below sanctioned strength for long periods, some state commissions defunct for want of a single appointed member, and clearance horizons for a fresh second appeal measured in years and, at one commission, in decades.