The 86th Amendment is the rare constitutional change whose text is the least interesting thing about it. Three short clauses, inserted, substituted and added in December 2002, did less work than the decade of litigation that forced them into being and the seven years of executive silence that followed them. The interpretive question at the centre of this article is one the constitutional text left entirely open: where does a fundamental right come from when the Constitution, as adopted, does not contain it? For education, the answer given by the Supreme Court was that it had been inside the right to life all along. The answer given by Parliament, a decade later, was narrower, more conditional, and, for more than seven years, purely theoretical, because the amendment that carried the right into Part Three of the Constitution also carried a commencement clause that left its arrival to a government notification.

That sequence, creation by judgment, restriction by judgment, codification by amendment, and commencement by notification, is the spine of what follows. Each stage changed not only what the right to education was but who held it against whom. A reader who follows the sequence carefully ends with something more valuable than the holdings themselves: a working map of how fundamental rights actually arrive in the Indian legal order, which is rarely by the front door of the constitutional text and often by a side entrance that a court opens and a legislature then rebuilds to its own dimensions. The private education sector, whose admission fees started the whole story in a Karnataka medical college, appears at every stage, and its position at the end of the sequence is very different from its position at the start.

Before the Constitution: compulsion attempted, and refused, for forty years

The choice the framers made in 1949, to promise education rather than to guarantee it, was not made in ignorance of the alternative. Compulsory elementary education had been demanded in India for two generations before the Constituent Assembly sat. In 1911, Gopal Krishna Gokhale introduced an elementary education bill in the Imperial Legislative Council that would have made primary schooling compulsory in areas where a prescribed proportion of children were already attending, with the cost shared between the state and local bodies. The bill was debated and defeated, the official majority voting it down on grounds that combined finance, administrative readiness and the alleged unreadiness of the population itself. The arguments deployed against Gokhale’s bill in 1911 and 1912 are recognisable in every later decade of this story: the entitlement was conceded in principle, the machinery was pronounced absent, and the cost was treated as a reason for postponement rather than as a problem for budgeting.

The provincial record after the transfer of education to Indian ministers under the dyarchy reforms was better than the imperial record and still fell far short of universality. Several provinces passed compulsory primary education legislation, municipalities were empowered to introduce compulsion within their limits, and enrolment climbed steadily, but compulsion on paper coexisted with attendance that nobody enforced and provision that nobody completed. The pattern that Indian education law would repeat for a century was set in these decades: a statute announcing compulsion, an apparatus unequal to it, and a gap between the two that official reports acknowledged and official budgets declined to close. When independence came, the leaders of the new state inherited both the aspiration and the habit of deferring it, and the Constitution they wrote bears the marks of both inheritances at once.

A promise filed under directions: education in the Constitution as adopted

The Constitution that commenced on 26 January 1950 contained no fundamental right to education. What it contained was a direction. Article 45, in Part Four, the directive principles of state policy, provided that the state shall endeavour to provide, within a period of ten years from the commencement of the Constitution, for free and compulsory education for all children until they complete the age of fourteen years. Two features of that sentence deserve to be held apart, because the whole later history turns on them. The first is its placement. Part Four is expressly non-justiciable: Article 37 declares that the directive principles shall not be enforceable by any court, though they are fundamental in the governance of the country and it is the duty of the state to apply them in making laws. The second is its deadline. Article 45 was the only directive principle written with a clock attached. The framers gave the republic ten years to put every child under fourteen into free and compulsory schooling, and the ten years expired in 1960 with the undertaking unredeemed.

Around Article 45 sat two neighbouring provisions that later became load-bearing. Article 41, also a directive principle, directed the state, within the limits of its economic capacity and development, to make effective provision for securing, among other things, the right to education. Article 46 directed the state to promote with special care the educational and economic interests of the weaker sections of the people. Neither provision conferred an enforceable claim on any child. A parent whose child had no school in 1965, or in 1985, held a constitutional promise addressed to governments, not a constitutional right addressed to courts, and the distinction was not a technicality. It determined which institution could be asked to act, and what happened when the institution declined.

The non-delivery is a matter of record rather than rhetoric. Successive governments expanded schooling substantially in absolute terms, building schools and enrolling children in numbers the framers would have found astonishing, and still the universal elementary education that Article 45 demanded within a decade remained incomplete four decades later. The reasons were fiscal, administrative and social, and they are the subject of the implementation literature rather than of this article. What matters for the legal story is the posture the failure created. By the early 1990s, Article 45 stood in the text as a promise the political branches had made to themselves and broken on a schedule they had set themselves. That posture, a directive principle with an expired deadline, is what the Supreme Court picked up when a dispute about a medical college fee reached it in 1992. The Court did not treat the expired deadline as an embarrassment to be passed over. It treated it as an argument.

The placement of the promise in Part Four was itself a debated compromise, and recovering the terms of that compromise matters because both sides of the 1990s litigation claimed the framers’ authority. Members of the Constituent Assembly pressed for education, along with other social and economic goods, to be placed among the justiciable fundamental rights. The counter-argument, which prevailed, was frank about its ground: a state born amid partition, refugee resettlement and an empty treasury could not honour writs commanding it to build a school system instantly, and a right that the state would predictably violate from its first day would damage the authority of every other right in Part Three. The directive principles were the Assembly’s device for holding both ends of that dilemma. The goods were named as fundamental in the governance of the country, the state was directed to pursue them, and the courts were excluded. Article 45’s ten-year deadline was the residue of the losing side’s case: if education was not to be a right, the Assembly at least recorded, in the text, the speed at which it expected the promise to be kept. The deadline is what distinguishes Article 45 from the general run of directives, and it is what made the provision, decades later, usable as a lever by a court inclined to pull it.

The states, for their part, spent the decades of Article 45’s dormancy rehearsing the manoeuvre the Union would later perform on the Constitution itself. Most major states enacted compulsory primary education statutes in the years after independence, laws that empowered local authorities to require attendance, prescribed penalties for defaulting parents, and then lay substantially unenforced for want of schools, teachers and administrative will. The statutes were not fraudulent; they were aspirational instruments in a system that had learned to separate the announcement of compulsion from its delivery. A legal historian reading the state statute books of the 1960s and 1970s encounters the same structure that Section 1(2) of the 86th Amendment would later embody at the summit of the legal order: an obligation enacted in full solemnity, whose operative life awaited an executive decision that fiscal reality kept postponing. The appointed day, and the appointed decade, are the Indian state’s characteristic compromise between the rights it proclaims and the resources it commands, and education is merely the subject on which the compromise is most fully documented.

Could a citizen enforce Article 45 in a court?

No. Article 37 states that the directive principles are not enforceable by any court, and Article 45 sat among them, so a parent could not sue on the promise. The provision bound the state politically and guided legislation, but conferred no claim a judge could decree, which is why the right later had to be found inside Part Three.

The instrument: how Article 21 learned to carry unlisted rights

The route into Part Three ran through Article 21, the guarantee that no person shall be deprived of life or personal liberty except according to procedure established by law. For education to become a fundamental right without a constitutional amendment, a court had to hold that education was already inside “life”, and that holding was possible only because of the interpretive career Article 21 had led in the decades after 1970. The habit of reading the fundamental rights generously, and of reading them together rather than as sealed compartments, was not invented for the education cases. In the bank nationalisation case, decided in 1970, an eleven-judge bench had already rejected the older view that each right in Part Three occupies its own watertight field and that a law tested against one article need not answer to another; the account of how the bank nationalisation judgment reordered the reading of fundamental rights belongs to that article, and what matters here is the method it restored. Rights were to be read as a connected code, and state action was to be measured against the whole of it.

The second step in the method came in 1978, when the Court held that the procedure which takes away life or liberty must itself be just, fair and reasonable, and that Article 21 draws content from the dignity of the individual rather than from a bare animal existence. Once “life” meant a life of dignity, the question in any new case became empirical as much as textual: what does a dignified human existence require that the state may not withhold? In the years that followed, the Court answered that question for livelihood, for health, for a clean environment and for the conditions of prisoners, each time deriving an unlisted entitlement from the listed one. Education was an obvious candidate for the same treatment, because the argument writes itself: a person who cannot read cannot exercise the franchise intelligently, cannot use the freedom of speech and expression in its printed forms, cannot approach a court with understanding, and cannot compete for the occupations the Constitution elsewhere protects. The derivation had been available for years before any litigant put it at the centre of a case. It took a fee dispute in Karnataka to do that.

The derivations that prepared the ground were not abstract. In Francis Coralie Mullin in 1981, the Court held that the right to life includes the right to live with human dignity and all that goes with it, the bare necessities of life and the facilities for reading, writing and expressing oneself among them, a formulation in which education is present in everything but name. In Olga Tellis in 1985, the pavement dwellers’ case, the Court held that the right to livelihood is part of the right to life, since a person deprived of livelihood is deprived of life itself, and it tested a municipal eviction programme against that standard. By the time the education claim arrived, the pattern of reasoning had been rehearsed across the whole terrain of social existence: identify the condition without which a dignified life is impossible, locate that condition within Article 21’s protection, and measure the state’s conduct against the entitlement so derived. Education was not a harder case than livelihood. If anything it was an easier one, because the directive principles spoke about it in terms, with a deadline, in language the Court could quote. The surprise in the education story is not that the derivation succeeded in 1992. It is that it waited until 1992 to be attempted at the level of a court of record, a delay explained by the sociology of litigation rather than by doctrine: the children the right concerned do not file writ petitions, and the claim reached the Court only when a fee dispute carried it there in the luggage of the professional colleges.

There is a standing objection to this method, and an honest account should state it before the education cases are even reached. Derivation of unlisted rights gives judges a power that looks legislative: the content of the Constitution expands without the special majorities Article 368 demands, and the expansion is difficult to reverse, since a larger bench or a constitutional amendment is needed to undo it. Defenders of the method reply that the alternative is worse, that a right to life read as mere animal existence would betray the document’s own preamble and its promise of dignity, and that the directive principles supply a constitutionally legitimate source for the content poured into Article 21, so the court is harmonising the text rather than inventing one. The education cases are among the strongest exhibits for both sides of that argument. The right they produced was real, was grounded in an expired promise the framers themselves had written, and was nonetheless a right no voter had ever been asked to approve in that form. Parliament’s eventual response, adopting the right while cutting it down, can be read as the democratic correction the critics of the method had demanded. It can equally be read as proof that the method works: the court moved first, and the legislature, forced to engage, took responsibility for the final shape.

One more piece of doctrine had to be in place before the derivation could be run for education, and it concerns the relationship between Part Three and Part Four. In the Constitution’s first decades, the directive principles were routinely treated as subordinate to the fundamental rights whenever the two appeared to conflict, on the strength of an early holding that Part Four must conform to and run subsidiary to Part Three. The Emergency-era attempt to invert that hierarchy by amendment, giving the directives primacy over Articles 14 and 19, was struck down in Minerva Mills in 1980, and the position the Court settled on was neither hierarchy but harmony: the two Parts were to be read together, each informing the content of the other, and the balance between them was itself part of the Constitution’s basic structure. That settlement is the jurisprudential licence for Unni Krishnan. A court working within it could say, without contradicting the framers’ refusal to make Article 45 enforceable, that the enforceable right to life in Article 21 should be given the content that Article 45, Article 41 and Article 46 collectively describe. The directive remains unenforceable in its own name. It becomes enforceable as the dictionary definition of a right that is.

The education cases also arrived at a moment when the social composition of the litigation had changed what the Court was being asked to do. The fee-paying professional college sector, concentrated in the southern and western states, had expanded rapidly as governments licensed private medical, engineering and dental colleges faster than public capacity grew. These institutions were neither the charitable trusts of the older imagination nor ordinary businesses; they were the principal route into the professions for a middle class that the public system could not seat, and their economics ran on fees, donations and payments that shaded into capitation. A court asked to pronounce on the right to education in 1992 was therefore not writing on a clean slate about village schools. It was writing, unavoidably, about the price of a medical degree, and every proposition it announced for the child in a primary classroom would be quoted back to it by the managements of professional colleges. The tension between the two readerships of the same judgment runs through the entire decade that followed, and explains why the benches kept growing: two judges in 1992, five in 1993, eleven in 2002, five again in 2003, seven in 2005, as the Court repeatedly had to reconcile what it had said to schools with what it had said to colleges.

Mohini Jain: the right declared at full width

The case that opened the door was not brought by a child denied elementary schooling, and the mismatch between the litigation and the right it produced is the first thing to understand about it. Karnataka had a statute whose very name announced its purpose: the Karnataka Educational Institutions (Prohibition of Capitation Fee) Act, 1984. Acting under that statute, the state government issued a notification on 5 June 1989 fixing the tuition fees private medical colleges could charge. The schedule was tiered by origin. A student admitted against a government seat paid Rs 2,000 a year. A student from Karnataka outside the government seats could be charged up to Rs 25,000 a year. A student from outside Karnataka could be charged up to Rs 60,000 a year. Miss Mohini Jain, a resident of Meerut in Uttar Pradesh, was offered a seat at a private medical college in Tumkur on those terms: Rs 60,000 for the first year and a bank guarantee for the fees of the remaining years. She also alleged that a capitation fee of four and a half lakh rupees had been demanded of her, an allegation the management denied. Unable to pay, she petitioned the Supreme Court under Article 32 against the notification.

A two-judge bench of Kuldip Singh and R. M. Sahai JJ. decided the case on 30 July 1992, and the judgment of Kuldip Singh J. ranged far beyond the notification. The Court held that the right to education flows from the right to life in Article 21, because the dignity of the individual, which the Constitution secures, cannot be assured unless it is accompanied by the right to education. It held that the state is under an obligation to make an endeavour to provide educational facilities at all levels to its citizens, and that every citizen has a right to education from the state, qualified only in the sense that the state’s obligation is shaped by its economic capacity. On the fee question, the holding was unqualified: charging a capitation fee for admission to an educational institution amounts to a denial of the citizen’s right to education, is patently unreasonable, and offends Article 14. The tiered notification, insofar as it authorised the higher charges, could not stand.

Two features of the decision explain both its influence and its short life as a complete statement of the law. The first is the level at which the right was pitched. The case concerned admission to a medical college, a form of professional higher education, and the judgment spoke of education at every level without distinguishing the child’s elementary schooling from the adult’s professional degree. Read on its own terms, the decision suggested that the fundamental right extended across the whole educational ladder, with the state’s capacity relevant to the manner of fulfilment rather than to the existence of the right. The second feature is what the judgment did to private institutions. By striking at the fee a private unaided college could charge its own students, and by grounding the strike in the student’s fundamental right, the decision implied that private colleges stood inside the rights framework as bearers of obligations, not merely as businesses regulated by licence. For institutions that received no state money and depended on fees to survive, the implication was existential, and they said so in the litigation that followed within the year.

The judgment’s method is worth isolating, because it is the template the Constitution Bench later formalised. Kuldip Singh J. did not rest the right on Article 21 alone. The reasoning moved through the Preamble’s promise of dignity and fraternity, through the directive principles in Articles 38, 39, 41 and 45, which oblige the state to secure a social order promoting welfare, to protect childhood against exploitation, and to endeavour to provide education, and arrived at Article 21 as the enforceable vessel into which that accumulated content could be poured. Education, in a much-repeated strand of the reasoning, was treated as the concomitant of the other fundamental rights: the freedoms of speech, movement, occupation and political participation are substantially hollow for a citizen who cannot read, and a Constitution that guarantees those freedoms in Part Three cannot be indifferent to the capacity that makes them usable. The state, the Court held, is obliged to endeavour to provide educational facilities, and the citizen’s corresponding right is to receive education from the state, subject always to the practical register in which the judgment itself was written, which was the register of obligation of endeavour rather than of instant result. Later readers, hostile and sympathetic alike, often missed that register and read the case as a promise of free education for all on demand. The text supported breadth about the right’s existence and caution about its remedy, and the confusion of the two is one source of the disappointment the case is sometimes said to have caused.

Why did a medical college fee dispute produce a fundamental right?

Because the Court refused to treat the fee as a private bargain. Once the bench located education inside the right to life, a charge that priced an eligible student out of a seat became a constitutional injury rather than a commercial disappointment, and the notification authorising it had to answer to Articles 14 and 21.

The reaction to Mohini Jain within the legal world was divided in a way that predicted the reconsideration. Sympathisers with the outcome found the reasoning difficult to defend at its full width. A judgment that grounded the right in dignity and declared it at every level of education, in a dispute about the price of a private medical seat, had supplied a proposition for the law reviews and a problem for everyone else: state governments read it as a potential obligation to fund education without limit, private institutions read it as the abolition of their fee autonomy by a bench of two, and both readings were textually available. The professional colleges’ counter-argument, when it came, was not that education is unimportant but that a right declared without a theory of its limits would be honoured by evasion, and that the proper course for a claim of this magnitude was a considered statement by a Constitution Bench. The Court accepted the invitation with unusual speed. The interval between the two decisions was just over six months, and the second was argued and delivered as what it was, a partial retraction dressed as a clarification, which is why the relationship between the two cases is so persistently misreported: Unni Krishnan is remembered as the case that created the right to education by readers who know Mohini Jain only as its footnote, when in truth the creation and the restriction were separate acts by separate benches with separate theories.

Unni Krishnan: the right kept, and bounded by the directive principles

The reconsideration came quickly and from a bench built for the purpose. J. P. Unni Krishnan and others, students and institutions affected by the fee regime, brought the correctness of Mohini Jain before a five-judge Constitution Bench, and on 4 February 1993 the Court, in a majority judgment written by B. P. Jeevan Reddy J., delivered the decision that organised this area of law for a generation. The bench affirmed Mohini Jain on the two propositions that mattered most and cut it down on the one that mattered most after them. Affirmed: the right to education does flow from Article 21, and the charging of a capitation fee is impermissible. Cut down: the proposition that the right runs, as a fundamental right to free education, at every level of the system.

The instrument of the cutting down was the directive principles themselves. The Constitution Bench reasoned that a fundamental right cannot be read in isolation from the rest of the document, and that the content of the right to education must be determined in the light of Articles 41, 45 and 46. Reading those provisions together, the Court held that every child has a right to free education until the age of fourteen, and that this limb of the right is a fundamental right enforceable against the state. Beyond fourteen, the picture changes. The claim to education continues, but it is subject to the limits of the economic capacity and development of the state, the express qualification Article 41 attaches to the right to education in Part Four. It follows, the Court held, that there is no fundamental right to professional education as such, and that Mohini Jain, insofar as it suggested the contrary, did not state the law correctly. The judgment also drew support from the scheme of the International Covenant on Economic, Social and Cultural Rights, which obliges states to realise the right to education progressively, to the maximum of available resources, a frame that matched the Court’s own two-band structure of an unconditional elementary core and a capacity-dependent remainder.

The passage of time did argumentative work in the judgment that deserves attention, because it is the move that converted a directive into a right. Article 45 had promised free and compulsory education to the age of fourteen within ten years. By 1993, more than four decades had run. The Court treated the long-expired deadline not as proof that the promise had been abandoned but as proof that the political branches had exhausted the patience the Constitution allowed them, so that the directive had matured, in substance, into an entitlement the courts could enforce. Critics of the judgment have never been comfortable with this move, and their objection is worth stating in its strongest form: a provision the framers deliberately made non-enforceable was rendered enforceable by the simple passage of time, on a theory the text does not state, by judges the electorate did not choose. The answer the judgment implicitly gives is that the Court did not enforce Article 45; it interpreted Article 21, using Article 45 only to fix the content and the age limit of the right it found there. Whether that distinction is a real one or a polite fiction is the deepest question the case leaves behind, and the 2002 amendment can be read as Parliament’s way of retiring it: once the right was written into the text, the argument about how it got there stopped mattering for the elementary band.

What the amendment did not adopt is as instructive as what it did. Unni Krishnan recognised the free-education right for every child until fourteen, speaking of childhood as a whole without a lower age boundary in the holding, since the directive it drew on ran from birth to fourteen. The amendment, nine years later, began the right at six and returned the years below six to a directive principle. The judgment’s right was also, in form, a right held against the state in respect of the whole education system it presided over; the amendment’s right was expressly a right to education provided “in such manner as the State may, by law, determine”, a qualification with no equivalent in the judgment. Anyone who treats the 86th Amendment as a simple transcription of Unni Krishnan into the text will misread both documents.

One holding of Unni Krishnan is quoted less often than the education right but did as much work in the decade that followed, and it concerns the institutions rather than the students. The bench held that while a citizen has a right to establish an educational institution, there is no fundamental right to recognition, to affiliation or to state aid, and that the state and the affiliating universities may impose conditions for recognition and affiliation directed to the maintenance of standards and the prevention of the abuses the case was about. This holding is what made the scheme enforceable without a statute: the scheme did not need to be imposed on colleges as a command, because it could be imposed as a condition of the recognition every professional college required in order to function at all. It is also the holding that survived the scheme’s destruction intact. After 2002, when the seat-sharing fell, the principle that recognition is a privilege carrying conditions remained, and the regulatory state that grew up around private professional education, entrance tests conducted or approved by public authority, fee committees under statute, affiliation inspections, is built on the ground this holding cleared. The judgment’s most famous proposition was cut down by amendment and time. Its quietest proposition became the foundation of the regulatory order.

The scheme: a court drafting an admissions code

Unni Krishnan did not stop at doctrine. The Constitution Bench went on to frame a scheme governing admission and fees in private professional colleges, and directed that the scheme be imposed by every authority granting recognition or affiliation. No retelling of this period is accurate that treats the scheme as an afterthought; for the colleges that lived under it through the 1990s, the scheme was the judgment.

Its architecture was a division of every intake into two halves. Fifty per cent of the seats in a private professional college, the free seats, were to be filled on merit through a common entrance process at fees comparable to those charged in government colleges. The remaining fifty per cent, the payment seats, were also to be filled on merit, from candidates prepared to pay a higher fee, but that fee was capped at a maximum fixed by the state and could not be exceeded. No capitation fee could be charged under any name. Private unaided institutions were entitled to charge more than government institutions for the same course, recognising that they received no grant, but the ceiling converted that entitlement into a regulated price. Recognition and affiliation were conditional on compliance, which gave the scheme teeth without a statute behind it.

The scheme’s logic was cross-subsidy under judicial management. Payment seats would finance free seats, merit would govern entry to both, and the state’s fee ceiling would hold the whole structure against profiteering. Its difficulties were equally structural. A court order is a blunt instrument for running an admissions market: it fixed proportions that bore no relation to any individual institution’s costs, it treated merit lists and fee bands as the whole of educational policy, and it placed courts and committees in continuing supervision of decisions they were not designed to make. Institutions argued, with increasing force as the decade wore on, that the scheme treated the running of a college as a public function while denying it public funding, and that the right to practise an occupation under Article 19(1)(g), which includes establishing and administering an educational institution, could not survive a regime in which the state dictated half the intake and capped the price of the rest. That argument, deferred in 1993, was the argument that won in 2002.

The scheme’s central category, merit, concealed a difficulty that its administration exposed year after year. Merit in the scheme meant rank in a common entrance examination, and the free seats and payment seats were simply two price bands drawn across a single ranked list, so that the fiftieth candidate might study free and the fifty-first, separated by a mark, might pay the ceiling fee. The arrangement treated the examination rank as a morally sufficient basis for distributing both the education and its cost, and it is vulnerable to the objection that has since become standard in the literature on the period: an entrance rank measures preparation as much as aptitude, preparation is purchasable, and a scheme that allocates subsidy by rank therefore directs the subsidy disproportionately to candidates whose families had already purchased the best preparation. The bench had not designed the scheme as an instrument of social justice in admissions; reservation, the Constitution’s own instrument for that purpose, sat outside it, accommodated by the states within the scheme’s frame in ways that varied and were litigated. The scheme is best understood as an anti-profiteering device with a cross-subsidy attached, and judging it as a failed equality programme, as its critics sometimes do, condemns it for not attempting what it never claimed. What it claimed was that price would not decide admission. On that claim its record is mixed but real.

It is worth pausing on what kind of judicial act the scheme was, because the vocabulary used about it since often misses this. The bench did not strike down a statute and stop. It supplied, from the bench, the operative code for a sector, complete with seat matrices and fee bands, and made compliance a condition of legal existence for the institutions concerned. Defenders of the scheme answer that the vacuum was real, that legislatures had left capitation and profiteering substantially unregulated, and that a court which had just declared a fundamental right could not consistently leave its enjoyment to a market in which seats were sold to the highest bidder. Critics answer that the cure institutionalised the disease in a milder form, since the payment seat is itself a price on admission set by category rather than by auction, and that a court performing the functions of a fee commission for a decade demonstrates the limits of adjudication as government. Both positions were fully argued in the years that followed, and the eleven-judge bench that finally dismantled the scheme did not pretend the questions were easy.

Living under the scheme: the decade of the free seat and the payment seat

For roughly nine years, the Unni Krishnan scheme was the operating constitution of private professional education in India, and its lived operation explains the force of the reaction against it. Admission seasons ran through the scheme’s categories. State governments conducted common entrance tests and counselled candidates into free seats and payment seats in order of merit, while colleges calculated their finances around a fee ceiling they had not set and a free-seat intake for which no government paid them. The arithmetic at the heart of the scheme, that payment seats would subsidise free seats within each institution, worked unevenly at best: it presumed that the paying half of an intake could carry the free half at fees the state was willing to approve, in institutions whose costs varied enormously by course, city and speciality. Where the arithmetic failed, the pressure escaped sideways, into payments demanded under other names, donations routed through trusts, and the continuing capitation economy the scheme existed to abolish, which the courts then policed institution by institution in a stream of contempt and enforcement litigation.

The scheme also froze a particular theory of fairness into the admission process, and the theory had costs its authors had not priced. By defining entitlement through two fee bands within a merit list, it treated the price of a seat as the principal injustice in professional education and the reservation categories of the Constitution as a secondary matter to be accommodated around the scheme’s edges. State governments chafed at the loss of control over a sector educating their own citizens; institutions chafed at the loss of control over their own survival; and students who could pay something but not the payment-seat fee occupied a middle band the scheme did not recognise at all. Each of these grievances reached the Supreme Court in some form across the 1990s, usually as an application for clarification, and the clarifications never quite clarified, because the difficulty was not in the scheme’s wording but in its premise that a court could administer a market indefinitely. When the eleven-judge bench was finally constituted, it was constituted less to answer a new question than to relieve the Court of a government it had been running for nine years.

It is only fair to record what the scheme achieved in its decade, because the reaction against it tends to erase the record. It destroyed the open legitimacy of the capitation fee, which before 1992 had been defended in public as a normal incident of private education and after 1993 could be defended nowhere. It established, as a matter of enforceable practice rather than rhetoric, that admission to a professional college could be ordered by merit rather than purchased outright, and the common entrance test as the gateway to professional education, which Indian admission systems take for granted, was entrenched in this period under the scheme’s discipline. And it held open a space, at regulated prices, for meritorious students without wealth in institutions that would otherwise have sold every seat. The eleven judges who dismantled the scheme did not deny any of this. Their holding was about who may govern, not about whether the governance had been worthless, and the distinction matters for reading the judgment honestly.

TMA Pai: the scheme falls, the right survives

The scheme’s fate was settled on 31 October 2002, when an eleven-judge bench delivered judgment in TMA Pai Foundation v State of Karnataka, the largest bench assembled on education and one of the largest in the Court’s history. The sheer number of opinions tells part of the story: the eleven judges spoke through several opinions, with the majority position framed by B. N. Kirpal CJ., and the precise ratio of the case was argued about for years afterwards. On the scheme, however, the outcome was clear. The bench held that the Unni Krishnan scheme, insofar as it imposed a sharing of seats and a fee regime on private unaided institutions, was an unreasonable restriction on the right of those institutions to carry on their occupation under Article 19(1)(g), and it could not stand. The right to establish and administer an educational institution was recognised as an occupation protected by the Constitution, for minorities under the express guarantee of Article 30(1) and for all citizens under Article 19(1)(g). A private unaided institution was entitled to devise its own admission process, provided it was fair, transparent and merit-based, and to charge a reasonable fee, provided the fee involved no profiteering and no capitation. Regulation in the interest of excellence and fairness remained permissible; appropriation of the institution’s intake did not.

Two clarifications protect readers against the standard misdescription of this case. The first is that TMA Pai did not overrule the right to education. The eleven judges were not asked to reconsider whether a child has a fundamental right to elementary education, and nothing in the decision disturbs the Unni Krishnan holding on that point, which concerned the obligations of the state. What fell was a mechanism: the judicially drafted seat-sharing and fee scheme as applied to institutions the state did not fund. Pages that describe 2002 as the year the right to education was overruled have confused the mechanism with the right. The second clarification concerns minority institutions. Article 30(1) gives religious and linguistic minorities the right to establish and administer educational institutions of their choice, and TMA Pai read that guarantee as a substantive protection of institutional autonomy, not a mere equality clause. The distinction between aided and unaided, and between minority and non-minority, became the grid on which all later education litigation was plotted.

The sequel arrived within a year. In Islamic Academy of Education v State of Karnataka in 2003, a five-judge bench, confronted with state governments and institutions reading TMA Pai in opposite directions, set up two committees in each state as an interim arrangement: one, headed by a retired High Court judge, to oversee admissions and ensure they were fair and merit-based, and another of the same character to examine and approve the fee structure proposed by each institution, testing it against profiteering and capitation. The committees were a characteristically Indian judicial device, regulation by nominated body in the interval before legislation, and they kept a floor under the autonomy TMA Pai had restored without reviving the seat-sharing that TMA Pai had destroyed. Their existence is the clearest evidence that the eleven judges had settled a principle without settling a system, and that everyone in the field knew it.

The bench in TMA Pai framed eleven questions for itself, and the spread of those questions across Articles 19, 21, 29 and 30 is the measure of how much of the constitutional law of education the case rewrote at once. Its answers built the grid that still organises the subject. Institutions were sorted along two axes that the judgment treated as constitutionally significant in themselves: aided against unaided, since an institution that accepts state money accepts the conditions attached to it, and minority against non-minority, since Article 30(1) confers on minorities a right of establishment and administration that has no exact counterpart for the majority. Within that grid, the judgment gave its most quoted holdings. The right to establish and administer an institution includes the right to admit students, to appoint staff, to set a fee structure and to use the institution’s property for its purposes, and each incident is an aspect of the occupation the Constitution protects. A minority institution, aided or unaided, holds its Article 30(1) right against the state, and while an aided minority institution may be regulated and may be required, as a condition of aid, to admit a measure of students from outside its community, the earlier St. Stephen’s College decision had fixed the working balance by permitting an aided minority college to prefer its own community up to half its intake. Regulation, the bench repeated in several formulations, must be for the maintenance of standards and the prevention of maladministration; it must not become the annihilation of the right it regulates.

Within the grid, the judgment’s treatment of aided institutions completes the picture and is too often dropped from summaries. An institution that accepts state aid accepts it on the state’s conditions, and TMA Pai accepted that the state may attach to its grant requirements bearing on admissions, including a measure of seats filled through the state’s own process and the application of its reservation policy as a condition of the aid relationship. The autonomy of an aided institution is therefore real but purchased at a discount the institution itself negotiated when it took the money. It was the unaided institution, minority and non-minority alike, for which the judgment’s protection ran at full strength, and the line between the two classes of institution explains why the later battles were fought almost entirely over the unaided: the aided sector’s terms had been settled by the logic of the grant. The judgment also attracted a standing criticism that any honest account must record. By constitutionalising the occupation of running an institution and confining the state to regulation against maladministration and profiteering, the critics argued, the bench had placed the expansion of Indian higher education, which the unaided sector undeniably delivered in volume, under a charter that treats education as a trade the state may police but not direct. The defenders’ answer was equally structural: the seats existed because private capital built them, capital builds where its rights are secure, and a state that wished to direct admissions had the remedy of building its own institutions or paying for places in others. The 93rd Amendment, three years later, chose a third route between those positions, and its existence is the best evidence that neither the criticism nor the defence had wholly prevailed in the political forum.

The hardest interpretive problem TMA Pai left behind was its own text. Eleven judges produced a majority opinion and a set of concurring and partly dissenting opinions, and in the months after the judgment, state governments and institutions mined different paragraphs for opposite propositions, each side finding genuine support. Passages could be read to permit a state common entrance test for unaided institutions, or to forbid it; to allow a government seat quota in aided colleges, or to confine the state to the conditions of its grant. The confusion was not bad faith on either side. It was the predictable consequence of answering eleven questions at a level of generality that the immediate disputes had not required, and it is the direct cause of the Islamic Academy reference the following year and of the Inamdar reference two years after that. The TMA Pai episode is therefore also a lesson in the costs of very large benches: authority increases, but so does the surface area of the text from which rival readings can be quarried, and the Court spent the rest of the decade deciding what its largest education bench had meant.

The committee regime of Islamic Academy had a legislative afterlife that completes this part of the story. What the Court had improvised as an interim device, state legislatures progressively converted into statutory form, enacting admission and fee regulation laws for professional institutions that placed fee fixation and admission oversight on a statutory footing, with committees of the Islamic Academy pattern absorbed into Acts of the state legislatures. The conversion mattered constitutionally. A judicial committee supervising fees year after year is an anomaly that the doctrine tolerates as a bridge; a statutory committee is ordinary economic regulation of a kind legislatures enact routinely, and challenges to it must meet the ordinary tests of reasonableness rather than the extraordinary objection that the courts are governing. By the time Inamdar was argued, therefore, the practical question had narrowed considerably. Nobody proposed returning to an unregulated market in professional seats, and the state had acquired, through statute, much of the supervisory apparatus the scheme had pioneered. What statute could not give the state, after TMA Pai, was the one thing reservation required: a share of the seats themselves. That is the precise and limited ground on which Inamdar was fought.

Inamdar: the reservation wall of 2005

The remaining question was whether the state could do by policy what the Court had forbidden itself to do by scheme: appropriate a share of seats in unaided institutions for its own reservation policy. On 12 August 2005, a seven-judge bench answered in P. A. Inamdar v State of Maharashtra. The bench held that the state could not impose its reservation policy on private unaided professional institutions, whether minority or non-minority, and could not appropriate any quota of their seats. To do so would not be a reasonable restriction on the occupation right but a nationalisation of seats in institutions the state had neither established nor funded. The Islamic Academy committees survived the judgment, recast as a regulatory check on fees and admissions until such time as suitable legislation occupied the field, but the core of the decision was the wall it built: after Inamdar, the admission policy of an unaided institution was constitutionally insulated from the state’s social justice programme, however compelling that programme might be.

The constitutional consequence followed by a logic nobody in 2005 disputed. If the obstacle was the text of Articles 15 and 19, the remedy was to change the text. Parliament had used that remedy before, inserting Article 15(4) in 1951 to restore special provisions after an early judgment had struck down a communal order in admissions. Within months of Inamdar, the remedy was used again, and its product, the 93rd Amendment, is examined below after the amendment that concerns this article most, because the two amendments are cousins: both wrote judicially created pressure into constitutional text, and both were then tested in court on the question whether the writing had gone too far.

Inamdar’s reasoning deserves a slower reading than its conclusion usually receives, because the bench was at pains to separate three propositions that public argument habitually fuses. The first is that reservation is a constitutional value: nothing in the judgment questions the legitimacy of reservation in state institutions, and the bench treated the advancement of backward classes as an obligation the Constitution itself imposes. The second is that an unaided institution is not the state: the duty to pursue the Constitution’s social programme falls on the state, and the state cannot discharge it by commandeering the admissions of a body it does not fund, any more than it could discharge a housing obligation by requisitioning rooms in private houses without compensation or authority of law. The third is that the boundary is the text, not judicial preference: Articles 15(4), 15(5) as it later became, 16(4) and 30 mark where the Constitution itself has authorised inroads into institutional and individual claims, and outside those authorisations the occupation right governs. The judgment’s treatment of the Islamic Academy committees followed the same logic. Supervision of fees and admissions to prevent profiteering and capitation was regulation of the manner of carrying on the occupation, which Article 19(6) permits; a quota was an appropriation of the substance of the right, which it does not. Committees could stay until legislation replaced them; quotas could not be imposed at all.

The political response to Inamdar is sometimes described as Parliament overruling the Court, and the description, while constitutionally loose, captures the sequence’s rhythm across this whole article. The Court identifies what the existing text forbids; the political branches decide whether the forbidding is acceptable; where it is not, the text is changed by the special process the Constitution provides; and the Court then tests the change against the one limit it guards itself, the basic structure. Each of the great education turning points fits this rhythm. Mohini Jain read a right into the text and Unni Krishnan adjusted the reading. TMA Pai found the scheme beyond the text and the scheme ended. Inamdar found quotas beyond the text for unaided institutions, and the 93rd Amendment rewrote the text to admit them, within months, with a speed that showed how completely the political consensus for reservation in admissions commanded both Houses. The amendment was then litigated for eight years on precisely the basic structure question the rhythm predicts. Understanding the sequence as a dialogue conducted in holdings and amendments, rather than as a contest with a winner, is the surest way to keep the individual cases in proportion.

The 86th Amendment: what Parliament wrote, and what it withheld

The amendment that carried the right to education into the text had its own decade of gestation, and its drafting history is written into its final shape. A first attempt, the Constitution (Eighty-third Amendment) Bill of 1997, proposed to insert the right and to delete Article 45 outright, and it was examined by the department-related parliamentary standing committee on human resource development, which recommended that a provision for children below six be preserved rather than deleted, and by the Law Commission, whose 165th Report addressed free and compulsory education for children. The 1997 bill was never enacted. The measure that succeeded was the Constitution (Ninety-third Amendment) Bill of 2001, whose numbering records the order of its introduction rather than the order of its enactment, a perennial source of confusion between this amendment and the 93rd Amendment of 2005, which is a different instrument on a different subject. The 2001 bill was passed by the Lok Sabha in November 2001 and by the Rajya Sabha on 14 May 2002, and received the assent of the President on 12 December 2002, becoming the Constitution (Eighty-sixth Amendment) Act, 2002.

It is sometimes said that the amendment was ratified by the states before assent, and the point is worth correcting precisely because this series treats the amendment process as a subject in its own right. Ratification by the legislatures of not less than half the states is required only where an amendment touches the provisions listed in the proviso to Article 368(2): the election of the President, the extent of executive power, the judiciature chapters, the distribution of legislative powers, the Seventh Schedule lists, the representation of states in Parliament, or Article 368 itself. The 86th Amendment touched none of these. It altered Part Three, Part Four and Part Four-A, and so travelled on the special majority of Parliament alone, a route whose mechanics are set out in the account of the constitution amendment process in India. The states’ absence from the enactment story makes their presence in the commencement story, seven years later, entirely a matter of executive choice.

The enacted text did three things, each of which can be quoted exactly because each is short. It inserted Article 21A: the state shall provide free and compulsory education to all children of the age of six to fourteen years in such manner as the state may, by law, determine. It substituted Article 45 with a provision on early childhood care and education, directing the state to endeavour to provide early childhood care and education for all children until they complete the age of six years. And it added clause (k) to Article 51A, making it the duty of a parent or guardian to provide opportunities for education to his child or ward between the age of six and fourteen years.

Set against Unni Krishnan, the text is a study in adoption and reduction performed in the same instrument. Adopted: the core holding that free and compulsory education to the age of fourteen is a fundamental right of the child against the state, and the banishment of the whole subject from the merely aspirational register of Part Four. Reduced, in three separate moves. The age band was cut at the bottom: the judgment, following the original Article 45, addressed children until fourteen without a floor, while the amendment guarantees the right from six, and consigns the first six years to the substituted directive, where protection is weaker by design. The right was made conditional in its very wording: “in such manner as the State may, by law, determine” is a phrase that subordinates the content of the entitlement to future legislation, and it has no counterpart in the judgment, where the right’s content was fixed by the Court itself. And a duty was redistributed: clause (k) placed a constitutional duty on parents and guardians alongside the state’s obligation, a move critics read as diluting the state’s sole responsibility. The educationist Nalini Juneja, analysing the bill as it moved through Parliament, objected to each of these features: the exclusion of children below six, the conditional phrasing, the shift of obligation toward parents, and financial estimates attached to the measure that fell well short of the figures the Tapas Majumdar Committee had produced in 1999 for universal elementary education. The government’s answer, running through the amendment’s statement of objects and reasons, was that the right had to be framed so that a consequential statute could define its content, and that an unqualified textual guarantee without machinery would have been a promise of the Article 45 kind all over again. Both readings were available in 2002. The commencement story that followed gave the critics their strongest evidence.

Did the states have to ratify the 86th Amendment?

No. Ratification by half the state legislatures is required only for amendments touching the provisions listed in the proviso to Article 368(2), such as the Seventh Schedule lists or the judiciary chapters. The 86th Amendment altered Parts Three, Four and Four-A, so the special majority of Parliament alone enacted it.

The road from judgment to amendment, 1993 to 2001

Between the Constitution Bench’s judgment and the amendment’s introduction lies an eight-year interval that the standard accounts skip, and it contains the explanation for the amendment’s peculiar shape. Unni Krishnan had made the elementary right judicially enforceable, but a right whose content is fixed by a judgment is an awkward possession for a political system: legislatures could neither plan expenditure against it nor adjust its edges, and the executive lived under a standing reproach it could convert into policy only by legislating. Through the middle 1990s the demand for codification gathered institutional sponsors. A committee of state education ministers recommended that free and compulsory education be made a fundamental right by constitutional amendment. The Law Commission’s 165th Report took up free and compulsory education for children and supported the constitutional route. An expert group chaired by Tapas Majumdar was charged with estimating what universal elementary education would cost, and its figures, running far beyond the allocations governments had been making, became the permanent backdrop of the debate: everyone who favoured the amendment had to answer for the money, and everyone who opposed it had to answer for the right.

The 1997 bill was the product of this accumulation, and its failure is as instructive as the 2001 bill’s success. Introduced as the Constitution (Eighty-third Amendment) Bill, it proposed the insertion of the right and the deletion of Article 45, and it ran into the combined resistance that the under-six question has attracted ever since: child development advocates objected that deleting the directive abandoned the youngest children at the moment of constitutionalising the older ones, the parliamentary standing committee on human resource development recommended preserving a provision for early childhood, and the bill was allowed to drop. When the proposal returned in 2001, the design had absorbed the lesson. Article 45 would not be deleted but substituted, redirected to the under-sixes, converting a defeat into a reallocation. The manner-by-law qualification and the parental duty travelled in the same direction, concessions to the fiscal and administrative anxieties that had kept Article 45 a directive for fifty years. The amendment that finally passed is unintelligible without the failed bill behind it, because nearly every feature of the enacted text that critics attack as a dilution is a scar left by a specific objection raised between 1997 and 2001.

The amendment in Parliament: an easy passage and a contested text

The parliamentary history of the 86th Amendment presents a paradox that the rest of this article helps to resolve: a bill that cut down a court-declared right passed both Houses without a division that history remembers. The Lok Sabha approved the Ninety-third Amendment Bill in November 2001 with support across the political spectrum, and the Rajya Sabha followed in May 2002. Parties that disagreed about nearly everything in education policy agreed on the bill, and the near-unanimity is not mysterious once the bill’s structure is understood. It allowed every participant to claim the achievement and defer the argument. Supporters of the right could celebrate its arrival in Part Three. Supporters of restraint could point to the six-year floor, the manner-by-law qualification and the appointed-day clause as evidence that the obligation had been domesticated. The fiscal argument, which had defeated compulsion since Gokhale’s bill ninety years earlier, was not confronted but postponed, because nothing in the amendment fixed a deadline for commencement or an appropriation for delivery. The bill was, in this precise sense, a compromise with time itself: the Constitution would carry the right immediately, and the republic would pay for it later.

The contest that did occur took place in the space around the bill rather than on the floor. The rally of teachers, students and activists that gathered in Delhi on the day of the Lok Sabha’s debate, demanding radical changes to the bill, represented a formidable current of the education movement, and members across parties voiced parts of its case inside the House: that the exclusion of children below six abandoned the age group in which deprivation does its deepest damage; that the parental duty in clause (k) invited the state to blame the poorest families for its own defaults; and that the financial memorandum’s estimates understated the cost of the universal elementary education that the Tapas Majumdar Committee had costed three years earlier. These objections did not defeat the bill, but they fixed the terms in which it has been judged ever since, and every one of them reappeared in the commencement debate of the following decade. The standing committee process of 1997 had already shown one objection succeeding: the 1997 bill’s proposal to delete Article 45 was abandoned, and the under-sixes survived as a directive. In 2001 and 2002 the remaining objections were noted and outvoted by consensus, and the text that emerged is best read as the record of which objections the consensus could absorb and which it could not.

One further feature of the passage deserves record because later accounts routinely garble it. The amendment required no ratification, as the preceding section explained, but it did require the special majority of Article 368, a majority of the total membership of each House and two-thirds of the members present and voting, and it obtained that majority in both Houses without the bargaining that federal amendments usually extract. The absence of the states from the process is constitutionally orthodox and politically significant in the same breath. Elementary education sits in the Concurrent List, delivered overwhelmingly by state governments and state school systems, and the financial burden of the right would fall substantially on state budgets. A right created by the Centre, delivered by the states, and commenced at the Centre’s discretion concentrated every incentive toward delay in exactly the place where the delay occurred. The seven-year silence that followed was not inscribed in the amendment’s politics by accident. It was inscribed there by a design that separated the credit for the right from the cost of it, and then left the connection between the two to a notification.

A fundamental right with an appointed day: the seven-year silence

Section 1(2) of the 86th Amendment Act provided that the Act shall come into force on such date as the Central Government may, by notification in the Official Gazette, appoint. The clause is a familiar drafting device in Indian legislation, and this series has met it before at the highest level of the legal order: the 44th Amendment contained a provision, amending the property articles, that the executive never notified at all, a failure examined in the account of the 44th Amendment Act 1978. What happened to the 86th Amendment is the counterpart case, and in one way the more startling of the two. The unnotified provision of 1978 was a deletion. The uncommenced provision of 2002 was the creation of a fundamental right. Between 12 December 2002 and 1 April 2010, the Constitution of India, as printed and sold, contained an Article 21A that no child could invoke, because the amendment inserting it had not been brought into force. A fundamental right existed as enacted law and did not exist as operative law, and the only bridge between the two states was an executive notification that took more than seven years to issue.

The ordinary explanation for the delay deserves to be put at full strength, because it is genuinely strong. Article 21A was drafted to be completed by a statute: the state shall provide free and compulsory education “in such manner as the State may, by law, determine”. A right in that form has no determinate content until the law determining the manner exists. Commencing the article in 2003 without the statute would have handed the courts an enforceable entitlement whose content, school norms, teacher standards, neighbourhood criteria, financial sharing between the Centre and the states, they would have had to invent case by case, repeating the Unni Krishnan scheme experience at a far larger scale. The drafting of the consequential legislation was itself a substantial federal exercise: the first versions of a central education bill circulated from 2005, the bill in its enacted form was introduced in the Rajya Sabha in December 2008, and it passed through Parliament in 2009. On this view, commencing the amendment and the statute together on a single appointed day was not neglect but design, and the seven years measure the time a complex social statute took to build.

The objection is not to the design but to the duration, and it too deserves full strength. Seven years is longer than drafting, even federal drafting, requires, and the record of those years shows long intervals in which the bill was not moving at all, versions circulated and were withdrawn, and no interim commencement was even proposed for the parts of the amendment, such as the substituted Article 45, that needed no machinery. During every one of those years, children passed through the six-to-fourteen band and out of it; for a child who turned six in 2003, the right arrived, if it arrived in time at all, at the very edge of its coverage. A right whose beneficiaries age out of it while it waits for notification is not merely delayed in the way a delayed bridge is delayed. Its promise is consumed by the delay. The deeper institutional point is the one this series has tracked across dozens of statutes: the appointed-day clause transfers a fragment of the legislative power to the executive, because Parliament’s decision to enact can be held in suspense by the government’s decision not to commence, and no court has been willing to compel a commencement notification. In the education case the suspense ended well, with a statute and a commencement together. The mechanism that permitted the delay is identical to the mechanism that, in the 44th Amendment’s case, permitted a permanent refusal, and it operates on the Constitution itself with no greater friction than it operates on an ordinary Act.

One fact about the interval deserves separate emphasis, because it converts the delay from an abstraction into a paradox. Throughout the years in which Article 21A slept, the Union was operating the largest elementary education programme in its history. Sarva Shiksha Abhiyan, the flagship programme for universal elementary education launched at the start of the decade, was building schools, recruiting teachers and driving enrolment upward across the country under a framework of targets, missions and shared funding. The substance of the elementary education project, in other words, was not idle for seven years; money moved, classrooms rose and children enrolled in very large numbers. What slept was only the form: the conversion of the project from a programme, which a government may expand, shrink or rename at discretion, into a right, which a citizen holds as an entitlement and a court may protect. The distinction between the two is the whole subject of this article, and the interval displays it in its purest form. A state that delivers schooling as policy while withholding it as right retains the discretion that rights exist to remove, and the seven years are best understood not as years in which nothing happened but as years in which everything happened on the state’s terms rather than on the child’s.

The interval was not empty of law, and describing what filled it sharpens the oddity of the amendment’s dormancy. Because Article 21A was not in force, the enforceable right to education between 2002 and 2010 remained the judge-made right of Unni Krishnan under Article 21, and litigants went on invoking it in the High Courts on matters from school infrastructure to the conditions of teachers, so that Indian education law ran for seven years on a judicial text that Parliament had already voted to replace. State compulsory education statutes continued their older, quieter existence underneath. Within the Union government, the consequential bill went through the full cycle of Indian social legislation: drafts prepared with the advice of the Central Advisory Board of Education, circulation to the states whose systems would have to operate it, state objections concentrated on the sharing of cost between the Centre and the states, redrafting, and eventual introduction in the Rajya Sabha in December 2008, six years after the amendment it existed to serve. Each stage of that cycle was defensible in itself. Their sum was a delay longer than the entire parliamentary life of many governments, endured by a right whose text had promised nothing about waiting.

Nor was the silence legally challengeable, and the reason completes the constitutional picture. In A. K. Roy v Union of India, decided in 1982 in the shadow of the 44th Amendment’s own uncommenced provision, the Supreme Court had declined to direct the Central Government to bring an enacted constitutional provision into force, holding that the commencement power is an executive function into which a mandamus does not run, however much the Court regretted the provision’s dormancy. The doctrine survived the education amendment untouched: no court, between 2002 and 2010, could have ordered the notification that would switch Article 21A on, and none was seriously expected to. The appointed-day clause thus combines two features that each look innocuous alone. It is a standard drafting device, found in hundreds of statutes for sound administrative reasons. And at the constitutional level it amounts to a standing executive veto over the timing of a fundamental right, exercisable by inaction, unreviewable in practice, and invisible in the printed Constitution that citizens read. The education case is the veto’s most consequential exercise, precisely because it ended: the notification came, the statute came with it, and the veto was never converted into the permanent refusal that the 44th Amendment’s history records.

The implementing statute, when it came, was the Right of Children to Free and Compulsory Education Act, 2009, which received assent on 26 August 2009 and commenced, with the amendment, on 1 April 2010. That statute’s provisions, its norms for schools and teachers, its prohibition of screening and capitation, and its reservation of seats in private schools, are the operative content of the right in practice, and they belong to the statute’s own article: the Right to Education Act 2009 guide carries that account, and this article deliberately does not. What belongs here is the constitutional significance of the pairing. From 1 April 2010, the right to education ceased to be a judicial construction resting on Article 21 and became a textual right, Article 21A, with a statutory code beneath it. The courts’ thirty-year project of deriving the right was complete, and, in the manner of such completions, the text that finished the project was not the text the courts would have written.

The 93rd Amendment and the settlement on private institutions

The Inamdar wall stood for less than five months before Parliament began dismantling it by the only route available. The Constitution (Ninety-third Amendment) Act, 2005 was passed by Parliament in December 2005, received assent on 20 January 2006, and came into force the same day. It inserted clause (5) into Article 15, providing that nothing in Article 15 or in the occupation guarantee of Article 19(1)(g) shall prevent the state from making any special provision, by law, for the advancement of any socially and educationally backward classes of citizens or of the Scheduled Castes or the Scheduled Tribes, insofar as the provision relates to their admission to educational institutions, including private educational institutions, whether aided or unaided by the state, other than minority educational institutions referred to in Article 30(1). Three features of that drafting carry the whole later history. The power is enabling, not self-executing: it operates “by law”, so a legislature must still enact the reservation. The reach expressly covers unaided institutions, which is the Inamdar holding reversed in terms. And the minority exclusion is written into the clause itself, so the protection of Article 30(1) was preserved by the same instrument that overbore Article 19(1)(g).

The first exercise of the new power shows what “by law” meant in practice. Within the year of the amendment’s commencement, Parliament enacted the Central Educational Institutions (Reservation in Admission) Act, 2006, prescribing reservation, including twenty-seven per cent for other backward classes, in the central institutions the Union itself maintains, and state legislatures acquired the corresponding competence for institutions within their spheres. The two-step structure is constitutionally significant. Article 15(5) fixed no percentage, named no institution and reserved no seat; it removed a disability. Every operative choice, the percentage, the categories, the creamy layer’s treatment, the institutions covered, remained with legislatures answerable for them, which is why the subsequent litigation attacked the choices and the amendment in the alternative, and why the amendment’s survival did not conclude the challenges to the laws made under it. The 93rd Amendment is therefore best characterised not as a reservation measure but as a jurisdiction measure: it decided which institution, Parliament and the state legislatures rather than the courts administering TMA Pai, would hold the power to decide whether reservation reached the unaided sector.

The amendment was challenged at once, and its judicial history resolves into two holdings three years and then six years apart. In Ashoka Kumar Thakur v Union of India in 2008, a five-judge Constitution Bench upheld Article 15(5) against a basic structure challenge insofar as it applied to state-maintained and aided institutions, in a case that also sustained the twenty-seven per cent reservation for other backward classes in central educational institutions; the bench expressly left open the validity of the clause as applied to private unaided institutions, on which one member, Bhandari J., had taken the view that imposing reservation on unaided institutions abrogated the occupation right. The question left open was answered in Pramati Educational and Cultural Trust v Union of India on 6 May 2014, when another five-judge Constitution Bench held Article 15(5) valid in its application to private unaided institutions as well, disapproved the contrary view taken in Ashoka Kumar Thakur, and held that the minority exclusion in the clause was itself constitutionally sound, since minority institutions form a separate class protected by Article 30. The same bench completed the settlement on the statute side: the Right to Education Act, valid in itself, does not apply to minority institutions, aided or unaided, because applying its obligations to them would abrogate the Article 30(1) guarantee. An earlier three-judge decision in 2012, in the Society for Unaided Private Schools of Rajasthan case, had upheld the Act’s twenty-five per cent admission quota for unaided non-minority schools while holding unaided minority schools outside it; Pramati generalised the minority exclusion and closed the doctrinal file. How the twenty-five per cent quota works on the ground is the subject of the account of the RTE twenty-five per cent quota’s implementation; the constitutional point for this article is that the quota rests on two distinct foundations, the statute’s own command for the schools it covers, and Article 15(5) as the enabling answer to the occupation-right objection that Inamdar had made unanswerable by ordinary law.

The 2008 stage of the settlement carried a condition that shaped a decade of admission practice and is worth isolating. In upholding the twenty-seven per cent reservation for other backward classes in central institutions, the Constitution Bench in Ashoka Kumar Thakur insisted on the exclusion of the creamy layer, the socially advanced members of a backward class, from the benefit of the quota, treating that exclusion as integral to the reservation’s constitutional validity rather than as a policy option. The condition mattered beyond its immediate setting because it confirmed that the courts would continue to police the internal design of reservation even after the enabling text had been supplied by amendment. Parliament could open the door that Inamdar had closed; it could not, by the same instrument, exempt the design of what walked through the door from equality review. The two-stage validation of Article 15(5), for aided institutions in 2008 and for unaided institutions in 2014, with the creamy layer condition laid down on the way, is the complete answer to anyone who supposes that a constitutional amendment ends judicial supervision of a subject. It relocates the supervision. It does not retire it.

Two features of the settlement’s slow judicial completion repay attention. The first is the role of the basic structure doctrine, which appears in this story not as a slogan but as a working test with a discriminating answer. Article 15(5) was attacked as an abrogation of the occupation right dressed as an amendment, and the attack failed in two stages because the benches treated the question as one of balance rather than of absolutes: an amendment that widens the equality code’s reach into private education, while carving out the minority right the Constitution elsewhere guarantees, rearranges the relations among rights without destroying any of them. Bhandari J.’s contrary reasoning in 2008, that reservation imposed on an institution that receives nothing from the state annihilates the institution’s freedom, was the strongest form of the attack, and its express disapproval in 2014 marks the point at which the settlement became final. The second feature is the asymmetry the settlement entrenches, which is deliberate and textual. A non-minority unaided school may be subjected, by law, to reservation and to the admission obligations of the education statute. A minority school, aided or unaided, may not, and the reason is not judicial solicitude but the words of Article 30(1) as the 93rd Amendment itself respected them when it wrote the minority exclusion into Article 15(5). Parliament chose the boundary; the courts policed it. Debates about the boundary’s wisdom belong to constitutional politics, and they continue there, but they should not be misreported as unsettled law.

Are minority institutions bound by the same rules as other private institutions?

No, and the difference is constitutional rather than statutory. Minority institutions, aided or unaided, stand outside both the Article 15(5) enabling power and the Right to Education Act, because the 2014 Constitution Bench held that subjecting them to those regimes would abrogate the Article 30(1) right to establish and administer institutions of their choice.

The architecture that emerges from the two amendments and the case law can be stated as a set of matched pairs. The child’s right runs against the state, and is delivered, in part, through obligations laid on private schools by statute. The private institution’s autonomy runs against the state, and is protected, for non-minority institutions, only up to the point where a constitutional amendment has authorised the intrusion, and for minority institutions, beyond that point altogether. Reservation in admissions, which the courts in 2005 placed beyond the reach of ordinary legislation aimed at unaided institutions, was placed within reach by constitutional amendment within the year, and the amendment was then sustained, in stages, by the institution that had provoked it. Anyone who describes the sequence as courts versus Parliament has the protagonists wrong. The sequence is better described as an iterated settlement, in which each institution in turn stated the most it would accept, and the final position is the one no actor in the sequence originally proposed.

Two words in the text: what “free” and “compulsory” bind, and whom

Two words of Article 21A carry more analytical weight than their brevity suggests, and the doctrine cannot be stated accurately without separating them. “Free” qualifies the education the state must provide, and its constitutional function is to remove price as a barrier between the child and the school the state provides or recognises for the purpose: a system that charges the child for the elementary education the article guarantees would contradict the text in the act of implementing it. “Compulsory” is the word that misleads foreign readers most reliably, because in several legal traditions compulsory education is a duty enforced against parents, with truancy offences behind it. In the Indian constitutional lineage the word has run the other way. From the original Article 45 onward, compulsion has described the obligation of the state: the state is compelled, by the constitutional text, to provide the schooling, to build the system that makes attendance possible, and to organise admission so that the right can be exercised. The parental duty that the 86th Amendment added in Article 51A(k) is a duty to provide opportunities for education, framed as a fundamental duty and unenforceable in itself, not a penal obligation to send a child to school on pain of prosecution. The implementing statute follows the same distribution: its commands fall on governments, local authorities and schools, and its treatment of parents is facilitative rather than punitive. A reader who imports the truancy model into Article 21A will misunderstand both the text and the politics, since the amendment’s settlement deliberately places the compulsion where the Indian debate has always placed it, on the provider and not on the poor.

A word on placement completes the reading of the text. Article 21A was inserted immediately after Article 21, inside the cluster of rights that the courts and the 44th Amendment had made the most protected ground in Part Three, and it is the only fundamental right in the Constitution held by a class defined by age: its beneficiaries are children from six to fourteen, who hold it without any act of assertion, and who lose it by the passage of time rather than by any act of their own. Rights of this shape are rare in the document, and the shape explains much of the surrounding design, the parental duty that shadows it, the directive that covers the years before it opens, and the anxiety, running through the whole history traced here, that a right held by those who cannot litigate for themselves will be exactly as strong as the institutions willing to speak for it.

The doctrine as it stands, and what remains open

The propositions a reader can carry away as settled are these. Free and compulsory education for children from six to fourteen is a fundamental right under Article 21A, enforceable against the state, and its detailed content is the content of the implementing statute, because the article itself assigns the manner of provision to law. The years below six are governed by a directive principle, Article 45 in its substituted form, and confer no equivalent entitlement. Education beyond fourteen, including higher and professional education, is not a fundamental right; the claims it generates are claims on the state’s capacity and policy under Article 41, not claims of constitutional entitlement. Private unaided institutions, minority and non-minority, hold a constitutionally protected autonomy in establishment, admissions and fees, bounded by the prohibitions on profiteering and capitation fees and by fair, transparent, merit-based processes. That autonomy yields, for non-minority institutions, to reservation and admission obligations imposed under Article 15(5) and the education statute; for minority institutions it does not yield, and the exclusion is itself part of the constitutional text and was reaffirmed in 2014. Capitation fees have been unconstitutional in every formulation of the doctrine since 1992, the single proposition that no bench and no amendment has ever disturbed.

The insertion of Article 21A also rearranged the argumentative geography around the right in ways that show up in litigation strategy. Before 2010, an education claim was argued upward from Article 21, with the directive principles as content and the two education judgments as authority, and the state answered in the same register. After commencement, the constitutional argument begins from the text of Article 21A and descends immediately into the statute, because the text directs it there: the manner of provision is what the law determines, so the contested questions are questions about the law’s validity, its coverage and its administration. Claims that the amendment was designed to exclude have fared accordingly. A child below six has no Article 21A claim to early childhood education, and the substituted Article 45, being a directive, supplies none that a court can decree; the remedy for the under-six gap is legislative, and its absence is a policy failure rather than a constitutional violation. A student above fourteen stands in the same position in the other direction, remitted to Article 41’s capacity language and to whatever statutes and schemes the state maintains. The edges of the band are therefore hard edges, fixed by the constitutional text, and litigation at the margins has been about the statute’s admission rules and the state’s classification choices under Article 14, not about stretching the band itself. This hardness is the deliberate consequence of writing the right as a bounded article rather than leaving it as a derivation, and it is the clearest single difference between the right the courts made and the right Parliament kept.

The practical consequences of the doctrine sort by reader, and stating them plainly is part of what an interpretation article owes. For a parent, the enforceable claim is twofold and the two halves must not be fused: against the state, Article 21A guarantees that free and compulsory elementary education be provided for a child from six to fourteen; against a school, the claim runs through the implementing statute, its admission categories, its norms and its complaint authorities, and it differs sharply according to whether the school is a government school, an aided school, an unaided non-minority school or a minority school. For an institution, the first legal question is classification, because category decides almost everything that follows: aid brings conditions, minority status brings the Article 30 shelter, and unaided non-minority status brings autonomy qualified by the statute and by Article 15(5) legislation. For a legislature, the settlement defines the available instruments with unusual completeness: reservation in admissions may be extended to unaided non-minority institutions by ordinary law resting on Article 15(5); it may not be extended to minority institutions without a further constitutional amendment that would itself face the Article 30 guarantee and the basic structure test; and the content of the elementary right may be reshaped through the statute that Article 21A designates, within the limits of the right’s own text. Each of these positions was purchased by a specific case or clause in the sequence, and none of them can be altered by administrative preference.

What remains open is structural rather than doctrinal, and can be described without pretending to a forecast. The age band of Article 21A is a standing invitation to amendment politics: proposals to extend the right downward to early childhood or upward through secondary schooling recur in public debate, and each would require the full Article 368 process, since the band is fixed by the constitutional text and not by the statute. The gap between the substituted Article 45’s aspiration for the under-sixes and any enforceable provision for them is the mirror image of the gap this article began with, a directive principle carrying a promise the courts have not been asked, in terms, to mature. And the relationship between the textual right and the judge-made right that preceded it retains one live edge: Article 21 continues to exist alongside Article 21A, and the question whether the older, wider derivation from the right to life could ever be used to outflank the limits Parliament wrote into Article 21A has been answered, in practice, by the courts’ acceptance of the amendment as the measure of the right, without a definitive theoretical burial of the derivation. The general position described in this article is legal information drawn from the constitutional text and the reported judgments; it is not advice on any individual admission, fee or reservation dispute, for which the facts and the forum rules will decide what the doctrine means in practice. Readers revising this case line for examinations can keep the sequence straight by working through it as a set of slides, and reviewing the five stages in a browser-based presentation viewer is a serviceable way to fix the order of the benches before the details are layered back on.

The conflation problem: three errors that organise the public confusion

No corner of Indian constitutional law is misdescribed more efficiently than this one, and the misdescription follows three standard errors so faithfully that correcting them is a service in itself. The first error attributes the 1993 holding to the 1992 case, usually in the form of the sentence that Mohini Jain made education a fundamental right up to the age of fourteen. That sentence fuses two decisions that held materially different things. Mohini Jain declared the right without the age limit and in the setting of professional education; the age limit of fourteen, the economic-capacity qualification beyond it, and the denial of a fundamental right to professional education are the work of Unni Krishnan, which cut Mohini Jain down in the act of affirming its core. A writer who attributes the bounded right to the earlier case has lost the entire plot of the sequence, which is a story of a right announced at one width and then deliberately narrowed by a larger bench within seven months.

The second error states that Article 21A has applied since 2002. It has not, and the error is the most consequential of the three because it conceals the article’s central finding. The amendment was enacted in December 2002 and commenced on 1 April 2010; between those dates the right it inserted was not in force, and any account of Indian education law in, say, 2006 that proceeds on the footing of an operative Article 21A is describing a legal order that did not exist. The third error describes TMA Pai as having overruled the right to education, or, in its milder form, as having held that the state cannot regulate private education. The eleven judges struck down a scheme, not a right; the fundamental right to elementary education against the state survived 2002 untouched, because it was not in issue; and the judgment expressly preserved regulation aimed at standards, fairness, merit and the suppression of profiteering and capitation. What TMA Pai withdrew was the courts’ own administrative machinery and the state’s power to appropriate unaided seats by scheme, two things narrower and more interesting than the abolition it is commonly credited with. Readers who hold the three corrections together, the right narrowed in 1993 rather than born bounded, the text dormant until 2010, and the 2002 bench dismantling a mechanism rather than a right, hold the accurate skeleton of the whole subject.

The right to education in five acts

The table below compresses the sequence this article has traced. Its payload is the final column read against the first: the constitutional right existed in enacted text from 2002 and in operative force only from 2010, so for more than seven years the row that mattered most was the one in which nothing was in force at all.

Stage	What the right was at that moment	Who was bound	The private sector’s position
Judicial creation, 1992	A fundamental right to education derived from Article 21, stated without an age limit and extending in principle to every level	The state, under an obligation shaped by its economic capacity	Capitation fees unconstitutional; a private college’s fee could be struck down as a denial of the right
Judicial restriction and the scheme, 1993	A fundamental right to free education until fourteen; beyond fourteen, subject to the economic capacity and development of the state	The state for the elementary band; courts and affiliating bodies as enforcers of the scheme	Half the seats free at regulated fees, half payment seats at capped fees; recognition conditional on the scheme
The scheme struck down, 2002	The elementary right untouched; the mechanism for delivering it through private colleges dismantled	The state, for the right itself; nobody, for the dismantled scheme	Autonomy restored under Articles 19(1)(g) and 30(1); fair admissions and reasonable, non-profiteering fees required
The constitutional amendment, 2002	Article 21A enacted: free and compulsory education from six to fourteen, in the manner law determines; under six returned to a directive principle	The state, on paper; no one in practice until commencement	The admission and fee position governed by the 2002 case law while the amendment slept
The reservation amendment, 2005	Article 15(5) enabled the state, by law, to advance backward classes, Scheduled Castes and Scheduled Tribes in admissions, including to unaided institutions, minority institutions excepted	Legislatures empowered; institutions exposed to laws made under the clause	Unaided non-minority institutions opened to reservation by law; minority institutions expressly excluded
Commencement, 2010	Article 21A in force from 1 April 2010, with the Right to Education Act supplying its operative content	The state, enforceably, for every child from six to fourteen	Obligations delivered substantially through the statute, including the admission quota for covered schools

Verdict: a right assembled by three institutions, none of which got its way

No single author wrote the Indian right to education. The Supreme Court created it in 1992 at a width its own Constitution Bench could not sustain in 1993. The Constitution Bench bounded it by an act of harmonisation that its critics regard as a reading down and its defenders regard as the only version of the right that could have survived contact with a poor state’s budget. The eleven-judge bench of 2002 protected a different set of rights holders, the institutions, and in doing so destroyed the only delivery mechanism the right had ever had inside the private sector. Parliament then wrote the right into the text at a narrower width than the courts had recognised, cut its lower age boundary, qualified it by the manner of future law, and left it uncommenced for seven years while the qualifying law was built. The settlement of 2014, which closed the file, upheld nearly everything: the amendment against the institutions, the statute against the non-minority schools, and the minority exclusion against the equality challenge.

Two judgments of value can be made on that record without overstating it. The first is that the iterated process produced a more durable right than any single stage of it would have done. Mohini Jain’s version was unenforceable at its stated width; Unni Krishnan’s scheme was ungovernable; a constitutional text imposed in 2002 without the political settlement of 2005 and the statute of 2009 would have been litigated for a decade on its content. The right that operates is narrower than the right that was proclaimed, and it operates, which the proclaimed one did not. The second judgment runs the other way, and honesty requires it. At each narrowing, the cost was borne by the same constituency: children below six, whose claim was returned to a directive; children outside the state system, whose delivery mechanism was struck down in the name of institutional autonomy; and the cohorts of 2003 to 2010, whose right was postponed past the age at which it could have served them. A constitutional order that arrives at a defensible right by successively disappointing the people the right is for has solved a problem of doctrine and created a problem of memory. The text of Article 21A carries no trace of any of this. That is why the sequence, and not the text alone, is the thing worth knowing.

The two standing debates the brief of this subject always generates can be closed with the positions stated at their strongest, since the evidence assembled here supports verdicts on neither being suppressed. On Unni Krishnan, the charge is that reading a right down by reference to the directive principles inverted the constitutional order, making the enforceable Part Three take its limits from the unenforceable Part Four, and that a bolder court would have held the line at Mohini Jain’s width. The answer is that the inversion charge mistakes the doctrine the judgment used: the directives fixed the content of a right whose existence and enforceability came from Article 21 alone, and the alternative the charge proposes, a fundamental right to professional education enforceable against a state that could not fund its primary schools, would have discredited the right it exalted. The verdict this article reaches is that the reading down was the price of the right’s survival, and that Parliament’s later adoption of the same boundary, fourteen, in the constitutional text is the strongest evidence the price was correctly judged. On TMA Pai, the charge is that institutional autonomy was purchased with access, and that the poor paid for the colleges’ freedom. The answer is that the autonomy protected was the autonomy of institutions the state had refused to fund, that the access the scheme provided was financed by a mechanism no legislature had enacted and no budget had approved, and that when the political system decided access should be extended into the unaided sector it amended the Constitution to say so, which is where such decisions belong. The verdict here is narrower: TMA Pai was right about the scheme and incomplete about the system, and the incompleteness, not the holding, is what the following decade of amendments and benches was spent repairing.

A final comparison fixes the article’s thesis in the series’ longer argument. The appointed-day device has appeared at both ends of the constitutional scale of values. At one end stands the 44th Amendment’s property provision, a deletion of rights that the executive declined ever to commence, so that a part of the enacted Constitution was stillborn. At the other stands Article 21A, a creation of rights that the executive commenced after seven years, so that a part of the enacted Constitution slept and then woke. The distance between the two outcomes is not legal; the clause is the same clause, and the doctrine of A. K. Roy covers both. The distance is political: a deletion nobody mobilised to demand remained deleted, while a right with a vast constituency of parents, teachers and movements behind it eventually acquired the pressure that a notification requires. The lesson the sequence teaches, and it is a lesson about the Indian Constitution rather than about education alone, is that the text is the beginning of a right’s career and not its end. Between enactment and life there stands an executive decision, and behind that decision there must stand either a statute ready to commence with it or a public unwilling to let it sleep. Education, uniquely, supplies the case where both were eventually assembled. That is the most that can be claimed for the process this article has traced, and it is not a small claim.

Frequently Asked Questions

Q: What did the Supreme Court decide in Unni Krishnan?

A five-judge Constitution Bench held in 1993 that the right to education flows from the right to life under Article 21, and that every child has a fundamental right to free education until the age of fourteen. Beyond fourteen, the claim to education continues but is subject to the limits of the economic capacity and development of the state, drawn from Article 41. The bench affirmed that capitation fees are impermissible, partly overruled Mohini Jain insofar as it suggested a fundamental right at every level, and framed a scheme of free seats and payment seats for private professional colleges.

Q: How did Mohini Jain and Unni Krishnan differ?

Mohini Jain, decided by a two-judge bench in 1992, declared the right to education at full width: it flowed from Article 21, extended in principle to education at every level including professional courses, and made a private college’s capitation fee unconstitutional as a denial of the right. Unni Krishnan, decided by a five-judge Constitution Bench in 1993, kept the source of the right and the capitation holding but bounded the right itself. Free education is a fundamental right only until fourteen; beyond that it depends on the state’s economic capacity, and there is no fundamental right to professional education as such. Unni Krishnan also added the free seats and payment seats scheme, which Mohini Jain had not attempted.

Q: Why was the Unni Krishnan scheme struck down?

The eleven-judge bench in TMA Pai Foundation held in 2002 that the scheme, which required private unaided professional colleges to surrender half their seats as free seats at regulated fees and to fill the rest as payment seats under a state fee ceiling, was an unreasonable restriction on the right to establish and administer an educational institution. That activity is an occupation protected by Article 19(1)(g) for all citizens and by Article 30(1) for minorities. A regime that dictated an unaided institution’s intake and capped its prices, the bench reasoned, went beyond regulation in the public interest and appropriated the institution itself, so the scheme could not survive, although the underlying right to elementary education was untouched.

Q: What did the 86th Amendment insert into the Constitution?

The Constitution (Eighty-sixth Amendment) Act, 2002 made three changes. It inserted Article 21A, providing that the state shall provide free and compulsory education to all children of the age of six to fourteen years in such manner as the state may, by law, determine. It substituted Article 45, replacing the original direction on free and compulsory education to fourteen with a direction to endeavour to provide early childhood care and education for children below six. And it added clause (k) to Article 51A, making it a fundamental duty of a parent or guardian to provide opportunities for education to a child or ward between six and fourteen.

Q: Why did Article 21A take years to come into force?

Because the amendment was drafted not to commence itself. Its first section provided that it would come into force on a date appointed by the Central Government by notification, and no notification issued until the one appointing 1 April 2010. The delay also had a stated rationale: Article 21A promises education in the manner that law determines, so the right had no operative content until the implementing statute existed, and drafting that statute, the Right to Education Act, took from the middle of the decade until 2009. Critics answer that seven years exceeded any drafting necessity, that versions of the bill circulated and stalled, and that children aged out of the six-to-fourteen band while the right waited.

Q: How did the 93rd Amendment enable reservation in private colleges?

The seven-judge bench in P. A. Inamdar had held in 2005 that the state could not impose its reservation policy on private unaided institutions, because Articles 15 and 19(1)(g) stood in the way. The Constitution (Ninety-third Amendment) Act, 2005 answered by inserting Article 15(5), which provides that nothing in Article 15 or in Article 19(1)(g) prevents the state from making special provision, by law, for the advancement of socially and educationally backward classes and of the Scheduled Castes and Scheduled Tribes in admission to educational institutions, including private institutions, aided or unaided, other than minority institutions. The clause removed the textual obstacle; a legislature still had to enact each reservation.

Q: Is there a fundamental right to higher education in India?

No, not as a freestanding entitlement. Unni Krishnan settled the point in 1993: the fundamental right to free education ends at fourteen, and education beyond that age, including higher and professional education, is subject to the limits of the economic capacity and development of the state under Article 41. The 86th Amendment confirmed the boundary by writing only the six-to-fourteen band into Article 21A. This does not leave higher education unprotected. Admission policies in state institutions must satisfy Article 14, reservations operate under Articles 15(4) and 15(5), and students hold the ordinary rights against arbitrary state action. What a citizen cannot do is demand a seat in a university or professional college as a fundamental right.

Q: What did TMA Pai decide about private institutions?

The eleven-judge bench held in 2002 that establishing and administering an educational institution is an occupation protected by Article 19(1)(g), and for religious and linguistic minorities by the express guarantee of Article 30(1). A private unaided institution may therefore devise its own admission process, provided it is fair, transparent and based on merit, and may charge a reasonable fee, provided the fee involves no profiteering and no capitation fee. The state may regulate in the interest of educational excellence and fairness, but it may not appropriate the institution’s seats or impose the kind of seat-sharing and fee regime the Unni Krishnan scheme had created. Those propositions organised every later dispute about private education.

Q: What did Mohini Jain decide about capitation fees?

The case arose from a Karnataka notification of 5 June 1989 that let private medical colleges charge Rs 2,000 a year to government-seat students, up to Rs 25,000 to other Karnataka students, and up to Rs 60,000 to students from outside the state. The two-judge bench held that a capitation fee charged for admission amounts to a denial of the citizen’s right to education, is patently unreasonable, and offends the equality guarantee of Article 14, and it struck down the notification insofar as it authorised the higher charges. The proposition that capitation fees are unconstitutional has never been disturbed: Unni Krishnan affirmed it in 1993, and TMA Pai, while restoring institutional autonomy in 2002, continued to prohibit profiteering and capitation in terms.

Q: What did the original Article 45 promise about free and compulsory education?

As adopted in 1950, Article 45 directed the state to endeavour to provide, within ten years of the Constitution’s commencement, free and compulsory education for all children until they complete the age of fourteen years. It was the only directive principle carrying its own deadline, and the deadline expired in 1960 without the undertaking being fulfilled, although schooling expanded greatly in absolute terms. Because Article 37 makes the directive principles unenforceable by any court, the promise bound governments politically but gave no child a claim a court could decree, a weakness the Supreme Court later worked around by reading the substance of the promise into Article 21, and Parliament eventually answered by substituting the article in 2002.

Q: What does “in such manner as the State may, by law, determine” mean in Article 21A?

The phrase is the qualification built into the heart of Article 21A, and it assigns the content of the right to legislation. Unlike the judge-made right under Article 21, whose content the courts fixed themselves, the textual right promises free and compulsory education only in the manner a law determines, so questions such as which schools must provide it, on what norms of infrastructure and teachers, within what neighbourhood, and with what sharing of cost between the Centre and the states are answered by the implementing statute rather than by the constitutional text alone. The Right to Education Act, 2009 is that determining law. The phrase is why commencing the amendment before the statute existed would have created an enforceable right with no defined content.

Q: What duty does Article 51A(k) place on parents and guardians?

Clause (k) of Article 51A, added by the 86th Amendment, makes it the duty of every citizen who is a parent or guardian to provide opportunities for education to his child or, as the case may be, ward between the age of six and fourteen years. It was the eleventh fundamental duty added to the original ten. Like the other fundamental duties, it is not enforceable by a court against an individual parent, and no penalty attaches to its breach by itself; duties of this Part operate as constitutional guidance and can be given teeth only by separate legislation. Its significance in the amendment’s design is distributive: alongside the state’s obligation under Article 21A, the text places a corresponding civic responsibility on the family.

Q: Why does Article 21A begin at six and end at fourteen?

The upper boundary came from the original Article 45 and from Unni Krishnan, both of which fixed fourteen, the age marking the end of elementary schooling and the threshold of lawful employment in the constitutional scheme. The lower boundary was Parliament’s choice in 2002, and it cut against the judgment: Unni Krishnan, following the original Article 45, spoke of children until fourteen without a floor at six. By starting the fundamental right at six, the amendment matched it to the age of entry into formal schooling, and moved the years below six into the substituted Article 45 as early childhood care and education, where the state’s obligation remains a non-enforceable directive. Critics of the amendment treat that division as its largest single retreat.

Q: What did Islamic Academy decide about admission and fee committees?

Islamic Academy of Education, decided by a five-judge bench in 2003, was the attempt to make TMA Pai workable after states and institutions read it in opposite ways. The bench directed each state to set up two committees as an interim arrangement: one, headed by a retired High Court judge, to supervise admissions in private professional institutions and ensure the process was fair, transparent and merit-based, and another of similar composition to examine each institution’s proposed fee structure and approve it if it involved no profiteering or capitation fee. The committees did not revive the seat-sharing scheme TMA Pai had destroyed. They were a regulatory bridge, expressly framed to operate until legislatures enacted their own machinery for fees and admissions.

Q: What did P.A. Inamdar hold about state quotas in unaided institutions?

The seven-judge bench held in 2005 that the state cannot impose its reservation policy on private unaided professional institutions, minority or non-minority, and cannot appropriate a quota of their seats for state nominees. To do so, the bench reasoned, is not a reasonable restriction under Article 19(6) on the occupation of running an institution but a takeover of admissions in an institution the state neither established nor funds, and for minority institutions it would also offend Article 30(1). The Islamic Academy fee and admission committees were allowed to continue as a regulatory measure until legislation occupied the field. The practical consequence was immediate: any reservation in unaided institutions thereafter required a constitutional amendment, which is what the 93rd Amendment supplied within months.

Q: Was Article 15(5) upheld for private unaided institutions?

Yes, but only at the second attempt and after six years of uncertainty. In Ashoka Kumar Thakur in 2008, a five-judge Constitution Bench upheld Article 15(5) for state-maintained and aided institutions against a basic structure challenge, and expressly left open its validity for private unaided institutions, where one judge had reasoned that imposing reservation on them abrogated the occupation right. The open question was closed by the five-judge bench in Pramati Educational and Cultural Trust in 2014, which held the clause valid for unaided non-minority institutions as well, disapproved the contrary reasoning of 2008, and confirmed that minority institutions stand outside the clause by the clause’s own terms and by force of Article 30(1).

Q: How does Article 21A relate to Article 21?

Article 21A is the codified successor of a right the courts had derived from Article 21. Between 1992 and 2010, the right to education existed as a construction placed on the right to life, which is why its content, including the age limit of fourteen, was fixed by judgments reading Part Three together with the directive principles. From 1 April 2010 the right has its own text, and in practice the courts treat Article 21A, read with its implementing statute, as the measure of the entitlement, including its six-year floor and its manner-by-law qualification. Article 21 has not been repealed or narrowed by the insertion; it continues to carry other derived rights. What changed is the source a litigant cites and the institution, Parliament rather than the bench, that last defined the right’s edges.

Q: Can a child enforce Article 21A against a private school?

Not directly, in the ordinary case. Article 21A, like the other rights in Part Three, is a right against the state: it obliges the state to provide free and compulsory education, and a writ for its enforcement runs against the state and its instrumentalities. Obligations reach private schools through the implementing statute rather than through the article itself. The Right to Education Act lays duties on unaided schools, including the admission of children from disadvantaged groups and weaker sections in the covered category, and supplies authorities and complaint routes for enforcing them. A dispute with a private school is therefore fought on the statute’s terms, while a failure of the state to provide schooling at all is the Article 21A claim.

Q: Why was the Eighty-sixth Amendment introduced as the Ninety-third Bill?

Because bills and Acts are numbered by different events. A constitutional amendment bill takes the next number in the sequence of amendment bills introduced in Parliament, while the resulting Act takes the number of its position in the sequence of amendments actually enacted. The education measure was introduced in 2001 as the Constitution (Ninety-third Amendment) Bill, since ninety-two amendment bills had preceded it in the introduction queue, counting bills that were later withdrawn, lapsed or enacted out of order. By the time it passed both Houses and received assent on 12 December 2002, it was only the eighty-sixth amendment to be enacted, so it entered the books as the 86th Amendment Act. The same gap explains why it must not be confused with the 93rd Amendment Act of 2005, a separate instrument inserting Article 15(5).

Q: What happened to the 1997 attempt to make education a fundamental right?

The Constitution (Eighty-third Amendment) Bill of 1997 was the first attempt, and it failed. It proposed to insert the right to education and to delete Article 45 altogether, leaving no directive for the years below the school age. The parliamentary standing committee on human resource development that examined it recommended instead that a provision for children below six be preserved, and the Law Commission’s 165th Report addressed free and compulsory education for children in the same period. The bill was never enacted, and the proposal returned four years later in altered form as the Ninety-third Amendment Bill of 2001, which kept a substituted Article 45 for early childhood care and education and became the 86th Amendment. The 1997 episode matters because the changes critics later attacked were revised, not invented, in 2001.