Sixteen months after the Constitution came into force, the government that had written it began rewriting it. The Constitution (First Amendment) Act, 1951 was moved by Jawaharlal Nehru on 10 May 1951, reported on by a select committee within a fortnight, passed in the first days of June, and assented to by President Rajendra Prasad on 18 June 1951. It altered the free speech guarantee, the equality guarantee and the property guarantee, and it created a schedule whose entire purpose was to place certain statutes beyond the reach of fundamental rights challenge. The legislature that did all this had never faced an electorate.

That combination of speed, scope and mandate is why the First Amendment is still argued about with more heat than any other constitutional change in India apart from the Emergency amendments. The argument, though, is usually conducted without the record. Most accounts present the amendment as a single decision with a single motive: either Nehru protecting land reform from reactionary courts, or Nehru curbing a press that had begun to criticise him. Neither reading survives contact with what actually happened, because the amendment was not one decision at all. It was three separate responses to three separate judicial defeats, drafted in different parts of the government for different reasons, bundled into one bill because one bill was easier to pass than three.

This article traces each clause of the amendment back to the judgment that provoked it, follows the bill through the select committee and the floor, and identifies what changed between the text Nehru introduced and the text that received assent. That gap is the analytical heart of the episode, because the most consequential single word in the amended free speech provision was not in the bill as introduced in the form it finally took, and its arrival changed the character of the restriction from a political instrument into a justiciable standard. The procedure by which all of this happened is set out separately in the guide to how the Constitution of India is amended, and this article assumes it rather than repeating it.

The constitutional settlement that lasted fifteen months

The Constitution came into force on 26 January 1950 with a chapter on fundamental rights that was, by the standards of the time, unusually generous and unusually enforceable. Part III did not merely declare rights; Article 13 rendered inconsistent laws void, and Article 32 gave the Supreme Court an original jurisdiction that a citizen could invoke directly. The framers had deliberately built a rights chapter with teeth, and they had done so knowing that the state they were creating intended an ambitious programme of economic redistribution.

The tension in that design was visible to anyone who looked. A government committed to abolishing intermediary tenures across the northern plains had written itself a constitution that guaranteed property, required compensation for acquisition, and gave every affected landholder a direct route to the highest court. A government facing communal violence, a communist insurgency in Telangana, and a hostile press on both flanks had written itself a constitution whose free speech clause permitted restriction only on a short and specific list of grounds. A government whose southern provinces had operated communal reservation in education for decades had written itself an equality clause that forbade discrimination on grounds of caste without expressly saving affirmative action in education.

None of these tensions was hidden. What nobody had tested was how a court would resolve them. Between May 1950 and April 1951 the Supreme Court and several High Courts resolved them, and in every significant instance they resolved them against the government. The First Amendment is the government’s answer, and the answer arrived within weeks of the last of those decisions.

It is worth being precise about what “against the government” means here, because the phrase is often used loosely. The courts did not strike down land reform in principle, did not hold that the state could never restrict speech, and did not hold that the state could never assist backward classes. They held, in each case, that the particular instrument the state had used exceeded what the constitutional text permitted. The government could have redrafted the instruments. It chose instead to redraft the text.

Why did the government amend the Constitution instead of redrafting the laws?

Because redrafting would have taken years and would not have cured the underlying problem. Land reform statutes in six or seven provinces were already in litigation, each on slightly different grounds. A constitutional amendment fixed every pending challenge at once and gave the states certainty. Speed, not principle, drove the choice of instrument.

That answer is the honest one, and it is uncomfortable in both directions. It concedes to the critics that the amendment was an instrument of convenience rather than a considered constitutional revision. It concedes to the defenders that the practical problem was real and that the alternative route was genuinely slow. Both concessions are necessary if the episode is to be understood rather than merely condemned or excused.

The first defeat: the press cases of 1950

The two decisions that produced the free speech clause of the amendment were delivered on the same day and concerned publications at opposite ends of the political spectrum, which is the fact that makes them so useful and which almost every popular account omits.

In Romesh Thappar v State of Madras, the government of Madras had banned the entry and circulation of Crossroads, a left-wing weekly published from Bombay, using a provision of the Madras Maintenance of Public Order Act that permitted restriction for securing public safety and the maintenance of public order. Thappar came directly to the Supreme Court under Article 32 rather than approaching the High Court first, and the Court held that he was entitled to do so, which is itself a foundational holding about the character of Article 32. On the merits, the Court struck the order down. Article 19(2) as originally enacted permitted restriction of free speech on grounds relating to the security of the State, and the Court held that public order was a wider and vaguer concept than the security of the State. A provision authorising restriction for the broader purpose could not be saved merely because it might also cover the narrower one.

In Brij Bhushan v State of Delhi, the Chief Commissioner of Delhi had required Organiser, a weekly associated with the Rashtriya Swayamsevak Sangh, to submit material relating to communal matters and to Pakistan for scrutiny before publication, under the East Punjab Public Safety Act. The Court struck that down too, treating pre-censorship as a restriction on the freedom of the press that the existing Article 19(2) did not authorise.

The pairing matters enormously to the political reading of the amendment. The state governments that lost these cases were acting against a communist paper and a Hindu nationalist paper respectively, and the constitutional problem was identical in both. The judicial holding was not that the state may not restrain either publication. It was that the constitutional text as written did not give the state the grounds it was relying on. The government’s response was to supply those grounds.

There is a further point in Romesh Thappar that gets lost. The Court’s reasoning turned on the doctrine that a law which is too wide is bad in its entirety when the wider and narrower purposes are not severable. That is a doctrine about drafting, not about ideology. It told legislatures that they would have to write restriction provisions that tracked the constitutional grounds precisely. A government confident of its position might have accepted that instruction and redrafted its public order statutes. Instead the amendment widened the constitutional grounds so that the existing statutes would fit inside them.

How many separate changes did the amendment make to Article 19?

It replaced the original Article 19(2) with a new clause that added public order, friendly relations with foreign states, and incitement to an offence to the permitted grounds of restriction, and it inserted the requirement that any such restriction be reasonable. It also gave the new clause retrospective effect and amended Article 19(6) to protect state trading and nationalisation.

Each of those four moves does something different, and treating them as a single act of curtailment is the most common error in writing about this amendment. Widening the grounds expanded state power. Requiring reasonableness contracted it, and contracted it in a way that has been litigated continuously ever since. Retrospectivity saved statutes already on the books. The trading amendment had nothing to do with speech at all and was inserted to answer an entirely separate line of challenges to state monopolies in commodities such as motor transport.

The second defeat: Champakam Dorairajan and the communal General Order

In State of Madras v Champakam Dorairajan, decided in 1951, the Supreme Court struck down the arrangement by which the State of Madras allocated seats in engineering and medical colleges among named communal categories. The order under challenge had its roots in the pre-independence politics of the Madras Presidency, where community-wise allocation in public employment and education had been established practice for decades and was regarded by its beneficiaries as the only reason non-Brahmin communities had access to professional education at all.

The Court’s reasoning was structurally simple and, on the text as it then stood, difficult to answer. Article 29(2) forbade denial of admission to a state-maintained or state-aided educational institution on grounds only of religion, race, caste or language. Article 15 forbade discrimination on those grounds. Article 16(4) expressly permitted reservation in public employment for backward classes, and the absence of any parallel saving in the education provisions was, in the Court’s reading, deliberate. The directive principles, on which the state relied, could not override an enforceable fundamental right.

The reaction in Madras was immediate and political rather than legal. The judgment did not merely invalidate an administrative order; it threatened an entire settlement about access to professional education in a state where that settlement was the central fact of political life. The government of India’s response was Article 15(4), which the First Amendment inserted to permit the state to make special provision for the advancement of socially and educationally backward classes and for the Scheduled Castes and Scheduled Tribes.

The precision here matters, because the popular description is wrong. Article 15(4) did not create reservation, did not mandate it, and did not fix any quantum. It is an enabling provision: it removes the constitutional objection that Article 15(1) and Article 29(2) would otherwise raise, and leaves the decision whether and how far to act with the state. Every subsequent controversy about ceilings, creamy layers, and the relationship between backwardness and caste flows from this enabling clause rather than from any command in it. The long lineage that runs from this insertion through the Mandal litigation is treated in the survey of social justice legislation in India, and this article stops at the point of insertion.

The other detail worth recording is what the amendment did not do. It did not touch Article 29(2), and it did not extend the new enabling power to minority institutions or to admissions generally. The saving was drafted narrowly around the problem the state actually had, which is a pattern that recurs through the whole amendment.

The third defeat: zamindari abolition in the High Courts

The property litigation is the largest of the three strands and the one that generated the most durable constitutional machinery, yet it is also the one where the popular account is furthest from the record, because it compresses a scattered mass of High Court litigation into a single imagined Supreme Court judgment that never existed.

The intermediary tenures that the new state proposed to abolish went by different names in different provinces: zamindari in Bihar and the United Provinces, jagirdari and inamdari elsewhere, malguzari in the Central Provinces. The statutes abolishing them were provincial statutes, drafted separately, with different compensation formulae, different classification of estates, and different transitional arrangements. When the holders challenged them, they challenged them in their own High Courts, and the outcomes were not uniform.

The Bihar litigation is the one that mattered most. The Patna High Court held the Bihar Land Reforms Act unconstitutional on the ground that its compensation scheme classified proprietors into groups and applied sharply different rates to them, which the court read as a violation of the equality guarantee in Article 14 rather than of the property guarantee itself. That is a crucial distinction and it is routinely misreported. The vulnerability the government discovered was not principally that the courts would insist on full market compensation; it was that the graduated, deliberately redistributive structure of the compensation formulae was exposed to an equality challenge, and that a court could strike down an entire abolition statute on that basis without ever reaching the property question.

Elsewhere, notably in the United Provinces and Madhya Pradesh, the abolition statutes survived challenge. The government therefore faced not a wall of judicial resistance but something more awkward: an unpredictable patchwork in which the constitutional fate of an agrarian programme depended on which High Court heard the case, with appeals to the Supreme Court pending and no prospect of a settled position for years. Land could not be redistributed while title was in doubt, revenue arrangements could not be reorganised, and the states were being told by their own law officers that the position was uncertain.

The response was two new articles and a new schedule. Article 31A provided that a law providing for the acquisition by the state of any estate, or for the extinguishment or modification of rights in an estate, would not be deemed void on the ground that it was inconsistent with the fundamental rights, subject to a proviso requiring presidential assent for state legislation of that kind. Article 31B went further and was the more radical instrument: it validated by name the statutes listed in a new Ninth Schedule, declared that they would not be deemed void or ever to have become void on fundamental rights grounds, and did so notwithstanding any judgment, decree or order of any court to the contrary.

What does Article 31B do that Article 31A does not?

Article 31A protects a category of laws defined by subject matter, and state laws qualify only with presidential assent. Article 31B protects named statutes listed in a schedule, unconditionally and retrospectively, and it operates notwithstanding any contrary judgment of any court. One is conditional; the other is absolute.

The retrospective phrasing is the point. Article 31B was drafted to reverse decisions that had already been given, not merely to prevent future ones, and the words used were about as strong as parliamentary drafting gets. A statute that a High Court had held void would be treated as never having been void at all. That is a validation clause operating on a constitutional plane, and nothing quite like it existed in the original Constitution.

The thirteen original entries were agrarian statutes from Bihar, Bombay, Madhya Pradesh, Madras, Uttar Pradesh and Hyderabad, dealing with the abolition of intermediary tenures and with land tenures more generally. That composition is the strongest argument available to the amendment’s defenders, and it deserves to be stated at its full strength: the schedule as enacted was a narrow, subject-specific device aimed at one programme, not a general-purpose immunity chamber. What it became is a separate question, and the answer to that question lies with the Parliaments that came later rather than with the one that created it. The schedule’s expansion into a repository for statutes having nothing to do with land is traced in the treatment of the 42nd Amendment and the Emergency-era additions.

The judgment-to-clause map of the First Amendment

The single most useful thing that can be said about this amendment is that every operative clause in it has a traceable parent in a specific judicial decision. Set out that way, the bill stops looking like a programme and starts looking like what it was, a schedule of repairs.

Provoking decision	What the court held	The clause the government inserted	Present status of that clause
Romesh Thappar v State of Madras (1950)	A restriction for public order could not be sustained under a clause permitting restriction only for the security of the State, and an over-wide provision fell in its entirety	New Article 19(2) adding public order, friendly relations with foreign states and incitement to an offence, with the reasonableness requirement	In force, and the reasonableness limb is the most heavily litigated part of Article 19
Brij Bhushan v State of Delhi (1950)	Pre-censorship of a publication was not authorised by the free speech clause as it then stood	The same replaced Article 19(2), given retrospective effect to save existing public safety statutes	In force
State of Madras v Champakam Dorairajan (1951)	Community-wise allocation of seats in state-aided professional colleges violated the equality and admission guarantees, and directive principles could not override them	New Article 15(4) enabling special provision for socially and educationally backward classes and for the Scheduled Castes and Scheduled Tribes	In force, and the source of the entire reservation-in-education jurisprudence
Bihar and allied zamindari abolition litigation in the High Courts (1950 to 1951)	Graduated compensation formulae in abolition statutes were vulnerable to an equality challenge, leaving agrarian reform in doubt across several provinces	New Article 31A saving estate acquisition laws, and new Article 31B validating named statutes	Article 31A in force; Article 31B in force but the immunity it confers has been substantially narrowed by later judicial doctrine
Challenges to state monopolies and trading ventures	State commercial monopolies were exposed to challenge under the freedom of trade guarantee	Amendment of Article 19(6) expressly saving state trading and nationalisation	In force

Reading the amendment through this table produces a different judgment about it than reading it as a manifesto. Nothing in the bill originated in a conception of what the Constitution ought to look like. Every clause originated in a courtroom loss. That is the finding that organises everything else in this article.

The reactive amendment pattern

The claim this article advances, and the one worth carrying away from it, is that the First Amendment established a mode of constitutional change that dominated Indian practice for the next three decades. Call it the reactive amendment pattern: when a court holds that a policy exceeds constitutional limits, the response is to move the limit rather than to reconsider the policy.

Three features define the pattern. The amendment is triggered by a judgment rather than by a considered proposal. It is drafted narrowly around the specific holding, which produces constitutional text that is oddly technical and reads like litigation drafting because that is what it is. And it is passed quickly, because the government’s interest is in restoring the status quo ante rather than in deliberation.

Once the pattern is named it becomes visible everywhere in the constitutional record that follows. The response to the bank nationalisation and privy purses decisions took the same form. The response to Golak Nath took the same form, in a bill that answered a specific holding with a specific clause. The response to the Allahabad High Court judgment setting aside an election took the same form and reached its most extreme expression there. The response to the reservation decisions of the nineties and after has repeatedly taken the same form. The pattern is not a Congress pattern or a Nehru pattern; it is an Indian constitutional pattern, and it starts here.

Naming the pattern also disciplines the argument about this amendment. If the reactive habit is the real inheritance, then the question worth asking is not whether Nehru was authoritarian but whether a constitution that can be amended by a two-thirds majority in a single House, in six weeks, in response to an adverse judgment, is going to behave like a constraint on government at all. That question is what the courts eventually answered with the basic structure limitation, and the sequence that runs from the litigation over this amendment to that answer is set out in the account of the basic structure doctrine and what Kesavananda actually decided.

Who drafted the bill and what it said on introduction

The drafting was done inside the Law Ministry with close involvement from the Prime Minister’s office, and the political direction came from Nehru personally. The Statement of Objects and Reasons appended to the bill is unusually candid by the standards of Indian legislative drafting and repays reading, because it does not pretend that the bill was a general improvement. It says in substance that challenges to agrarian and land reform statutes were pending in courts and were holding up large schemes of reform, and that the citizen’s right to freedom of speech had been given a scope by judicial decisions that the framers had not intended.

That candour is important for the historical argument. The government did not claim it was correcting a drafting oversight. It said plainly that the courts had read the Constitution in a way the government could not work with, and that the text would therefore be changed. Whatever one thinks of that position, it was stated openly on the floor and in the papers laid before the House, and the country’s newspapers reported it as exactly what it was.

The bill as introduced on 10 May 1951 already contained the essential architecture that survived into the Act: the new Article 15(4), a replacement of Article 19(2) widening the grounds of restriction, an amendment to Article 19(6) on state trading, the insertion of Articles 31A and 31B, and the addition of the Ninth Schedule. It also carried a set of unglamorous provisions having nothing to do with rights at all, which are discussed later in this article and which are almost universally ignored.

What the bill did not yet contain, in the settled form it eventually took, was the reasonableness requirement operating across the whole of the enlarged Article 19(2), and the precise formulation of the free speech grounds went through revision at the committee stage. The movement of that requirement from a contested proposal to the settled text is the most consequential thing that happened to the bill between introduction and assent.

The Provisional Parliament problem

No account of this amendment is complete without confronting the composition of the body that passed it, and no account is honest if it treats the answer as obvious.

The Constituent Assembly had been elected indirectly in 1946 by the provincial legislatures, which had themselves been elected on a franchise restricted by property, education and taxation qualifications covering a minority of the adult population. On 26 January 1950 that Assembly did not dissolve. Under the transitional provisions of the Constitution it continued as the Provisional Parliament until a Parliament elected under the new Constitution could be constituted, which happened only after the general election held between late 1951 and early 1952. The First Amendment was passed in that interval, by that body, in a single-chamber legislature with no Council of States yet in existence to check it.

The case against, stated at its strongest, is this. A constituent body derives its authority to write a constitution from the extraordinary character of the founding moment. Once the constitution is promulgated, that authority is spent. A body continuing in office by a transitional provision, elected on a narrow franchise by an electorate that no longer exists, chosen for a task it has completed, is the least appropriate institution imaginable to narrow the fundamental rights it has itself just enacted. The amendment was constitutionally valid and democratically illegitimate, and the fact that the same men were involved makes it worse rather than better, because it demonstrates how quickly the framers were willing to unmake their own work when it inconvenienced them.

The case for, stated at its strongest, is this. The Constitution expressly provided for the Provisional Parliament and expressly conferred on it the powers of Parliament, including the amending power, and there is no interpretive basis for reading in a limitation that the framers did not write. The alternative was to leave agrarian reform paralysed and public order statutes unenforceable for the eighteen months to two years that an election and a new Parliament would take, during which the courts would continue to invalidate statutes and the political costs would fall on a state with an insurgency in one region and communal violence in another. Waiting was not a neutral option with no consequences; it was a choice with its own costs, borne by different people.

Both cases are serious and neither is dishonest, and this article does not purport to settle a question that is finally about political values rather than about the record. What the record does settle is narrower and worth stating precisely. The government was aware of the objection, because it was made repeatedly and publicly by the leader of the opposition, by newspaper editors, by members of its own party, and privately by the President. It went ahead anyway. Whether that was courage or arrogance is a judgment each reader is entitled to make, but it was not inadvertence.

How many articles of the Constitution did the First Amendment touch?

Roughly a dozen, plus a new schedule. Three concerned rights, in Articles 15, 19 and 31. The rest were machinery corrections to the sessions and prorogation provisions for Parliament and the state legislatures, the specification of Scheduled Castes and Scheduled Tribes, and the transitional provisions on existing laws.

The select committee stage and the six dissents

The bill was referred to a select committee, which is a fact of some importance because it is frequently asserted that the amendment was rammed through without scrutiny. It was not rammed through without scrutiny; it was scrutinised at extraordinary speed. The committee reported within roughly a fortnight, and the House took the bill up again in the last days of May and the first days of June.

Six members of the select committee recorded minutes of dissent. That number deserves attention. A select committee on a government bill in a House with an overwhelming Congress majority producing six dissenting minutes is not a rubber stamp, and the dissents were not confined to the opposition benches. The political texture of the period was such that disagreement about the amendment ran through the governing party as well as across the aisle, and the departure of J B Kripalani from the Congress in the middle of the bill’s passage is a marker of how live that internal disagreement was.

The committee stage did real work on the text. The most significant change concerned the free speech clause, where the drafting was tightened and the reasonableness requirement was carried through the whole of the replaced Article 19(2). The importance of that change is developed below, but its origin is worth noting here: it emerged from a process in which the government was under sustained public criticism for the width of what it had proposed, and the concession, however grudging, came at the committee and floor stage rather than in the original draft.

This is a point that cuts against both of the usual narratives. The critics’ account, in which a supine legislature did what it was told, cannot explain six minutes of dissent and a textual concession on the central clause. The defenders’ account, in which the amendment received careful deliberation, cannot explain a total elapsed time from introduction to assent of under six weeks for a measure altering three fundamental rights.

The debate: the government’s case

Nehru’s defence of the bill on the floor was not primarily a constitutional argument, and pretending otherwise misreads him. It was an argument about the relationship between a written constitution and a programme of social transformation, and it was made with a bluntness that later governments learned to avoid.

The substance of it ran roughly as follows. The Constitution was a year old, and it was already producing results its authors had not intended and could not have wanted. Land reform, on which the entire agrarian settlement of the new republic depended, was being held up by litigation. The freedom of expression, which nobody wished to abolish, was being read so widely that the state could not act against speech that incited murder or inflamed communal feeling in a country that had just been through partition. A constitution was not scripture. It was a working instrument, and where it obstructed the purposes for which it had been made, it should be corrected. He was impatient with the suggestion that the framers’ work had acquired a sanctity that put it beyond revision by the same men who had done the work.

He also made a narrower and stronger argument that his critics have generally declined to engage with. The property provisions had been the subject of the most difficult compromise in the entire Constituent Assembly, and the compromise had been designed to permit exactly the kind of agrarian legislation now being struck down. On this reading the courts had not been protecting a settlement; they had been disturbing one, and the amendment restored the arrangement that had actually been agreed. That argument is contestable but it is not frivolous, and it is the best argument available for the property provisions of the amendment.

Where the government’s case is weakest is on the free speech clause. It is one thing to say that the state must be able to act against incitement to murder. It is another to insert friendly relations with foreign states as a ground for restricting the speech of citizens, which is a ground with no obvious connection to public safety and an obvious connection to the government’s diplomatic convenience. The government’s answer, that the ground was needed to control propaganda of a kind that could produce international incidents, was made and can be assessed on its merits. It has aged less well than the arguments about land.

The debate: the case against

Shyama Prasad Mookerjee led the opposition to the bill and framed it in institutional rather than partisan terms, which is why his speeches remain worth reading and why the attempt to dismiss his opposition as simple partisanship does not hold up against the record.

His central objection was about mandate. A body that had not been elected under the Constitution it was amending, sitting under a transitional provision, containing a large majority subject to the whip of a single party, was not the appropriate forum to narrow fundamental rights. He argued that the amendment should await the general election that was already scheduled, and that a government confident of its position had nothing to fear from putting the question to a Parliament with a mandate.

His second objection was about the free speech clause specifically. He argued that the existing restriction, framed around the security of the State, was the right test, and that adding public order handed the executive a general power to suppress criticism, since any sufficiently determined administration can characterise inconvenient publication as a threat to public order. That prediction has been tested repeatedly in the seven decades since, and the honest verdict is mixed: the ground has certainly been used expansively by executives of every political persuasion, and it has also been repeatedly reined in by courts applying the reasonableness standard that was inserted alongside it.

His third objection concerned the Ninth Schedule. He argued that a device for placing statutes beyond fundamental rights review was dangerous irrespective of the merits of the statutes first placed in it, because the device would outlive the occasion, and that a schedule created for zamindari abolition would in time carry whatever a future government wished to protect. Of the three objections, this is the one the record has most emphatically vindicated.

Outside the House, the opposition was broader than the parliamentary arithmetic suggests. The All India Newspaper Editors Conference pressed for wider consultation before the free speech clause was altered. The Speaker of the House, G V Mavalankar, was among those who urged that a measure of this kind should not be taken at such speed. This was not a fringe objection raised by defeated interests; it was a mainstream institutional concern raised through mainstream institutional channels, and it was overridden.

The President’s misgivings

President Rajendra Prasad’s private opposition to the bill is documented in the historical literature on the period and is a genuinely instructive episode about the constitutional office he held.

Prasad had doubts about the amendment and communicated them, raising both the question of the Provisional Parliament’s competence to narrow fundamental rights and his own position as the person who would have to assent to it. He was advised that his role was to act on the advice of his ministers and that he must give his assent. He gave it on 18 June 1951.

The episode is a useful early demonstration of something the Constitution left implicit and the courts and practice later made explicit: the head of state in the Indian system has an ample right to be consulted, to warn, and to require reconsideration, and no right at all to substitute his judgment for that of the government on a bill duly passed. Prasad’s misgivings were serious, expressed through the proper channel, and constitutionally irrelevant to the outcome. That is not a failure of the system. It is the system working as designed, and it is worth noticing that the design was tested this early and held.

It also disposes of a claim that circulates in popular argument, which is that the amendment was passed without any senior figure objecting. Senior figures objected. The leader of the opposition objected on the floor, six members of the select committee objected in writing, the Speaker objected on grounds of process, the newspaper editors objected publicly, and the President objected privately. The amendment passed anyway, because the government had the votes, and having the votes is what the amending procedure requires.

The gap between the bill and the Act

Every legislative history article in this series is built around one question: what changed between the text introduced and the text enacted, and who won that change. For the First Amendment the answer is compact and consequential.

The architecture survived intact. Article 15(4) went in substantially as proposed. Articles 31A and 31B went in, and the Ninth Schedule went in with its agrarian list. The state trading amendment went in. The government did not lose the bill or any of its major limbs, and anyone expecting the select committee to have gutted it will be disappointed.

What changed was the free speech clause, and what changed in it was the discipline imposed on the new grounds of restriction. The original Article 19(2) had permitted restriction on grounds relating to matters such as libel, slander, defamation, contempt of court, and matters offending against decency or morality or undermining the security of the State. It did not carry an express requirement that the restriction be reasonable. The replacement clause enacted by the First Amendment widened the grounds substantially and, at the same time, subjected every one of them to a reasonableness requirement.

That trade is the whole story of the clause, and both halves of it are real.

The government gained public order, which is a much wider concept than the security of the State and which has been used ever since to justify restrictions the original clause would not have permitted. It gained friendly relations with foreign states, a ground of restriction serving the executive’s diplomatic interest rather than any interest of the public. It gained incitement to an offence. And through the retrospective operation of the clause, it rescued statutes that had already been held invalid.

The country gained a justiciable standard. Under the original clause, a restriction falling within a listed ground was constitutional; the question of proportion did not formally arise. Under the replaced clause, falling within a listed ground is necessary but not sufficient, and every restriction must additionally be shown to be reasonable. That word is the hook on which the entire Indian free speech jurisprudence of the following decades hangs. Every case in which a court has struck down an overbroad restriction, insisted on a proximate connection between the speech and the harm, refused to accept a remote or fanciful apprehension of disorder, or held a restriction disproportionate to its object, has been argued through that word.

The honest assessment is therefore not that the amendment curtailed free speech or that it protected free speech. It did both, in a single clause, and the balance between the two has been struck and re-struck by courts for seven decades. A reader who takes away only the widening has an incomplete picture. So does a reader who takes away only the reasonableness requirement, which is the defence most often offered by the amendment’s supporters and which conveniently omits that the requirement was the price of a substantial expansion of state power rather than a gift.

Did the First Amendment make free speech restrictions easier or harder?

Both, and that is not evasion. It made the categories of permissible restriction substantially wider, particularly through public order. It simultaneously made every restriction testable for reasonableness, which the original clause did not require. Governments gained more grounds; courts gained a stronger standard of review.

What the amendment did that nobody discusses

The First Amendment is remembered for rights, and roughly half of its sections have nothing to do with rights at all. This is not a trivial observation, because it changes the character of the bill.

The Act made changes to the provisions governing the summoning of the Houses of Parliament and the sessions and prorogation arrangements, along with parallel changes for state legislatures, and it altered the addresses and messages provisions. It touched the provisions dealing with the specification of Scheduled Castes and Scheduled Tribes. It made changes to the transitional provisions concerning existing laws and their adaptation. These were technical corrections of the kind that any newly operative constitution generates in its first year, when the machinery is used for the first time and the drafting problems become apparent.

Their presence in the same bill as the free speech clause tells us two things. First, the government was not treating the amending power as reserved for grave constitutional occasions; it was treating it as an ordinary legislative instrument for fixing whatever needed fixing, which is precisely the attitude that made the reactive pattern possible. Second, the bundling made opposition harder, because a member who objected to the free speech clause was voting against a package that also contained uncontroversial machinery corrections.

Anyone assessing the amendment as a constitutional event should hold both halves in view. It was simultaneously a serious narrowing of two fundamental rights, an enabling of affirmative action in education, a rescue operation for agrarian reform, and a housekeeping bill.

Shankari Prasad and the question that would not go away

The amendment was challenged almost immediately. In Shankari Prasad Singh Deo v Union of India, decided in 1951, the Supreme Court upheld it.

The argument against the amendment had an elegant structure. Article 13 provides that the state shall not make any law that takes away or abridges the fundamental rights, and that any law made in contravention is void to the extent of the contravention. A constitutional amendment is made by Parliament, is enacted in the form of an Act, and takes away or abridges fundamental rights. Therefore, ran the argument, an amendment abridging fundamental rights is void under Article 13.

The Court rejected it, holding that the word law in Article 13 means ordinary legislation made in exercise of legislative power, and does not include a constitutional amendment made in exercise of constituent power. On that reading, Article 13 constrains what Parliament may do as a legislature and says nothing about what Parliament may do as the amending authority. The amendment therefore stood, and the Ninth Schedule technique stood with it.

Three things about that holding deserve to be flagged, because they set up everything that follows in this cluster.

First, it is a holding about the meaning of a single word, and the distinction it draws between legislative power and constituent power is a distinction the constitutional text does not spell out. The Court supplied it. Whatever one thinks of the result, the reasoning was more constructive than declaratory, which becomes significant when later benches are accused of inventing doctrine.

Second, the holding was not obviously stable, and the Court itself did not treat it as closed for long. The same question returned in 1965 and returned again in 1967, and by 1967 a majority had reversed the position. The instability was not because the judges changed their minds about statutory interpretation; it was because the amending power kept being used in ways that made an unlimited power look increasingly alarming.

Third, and most important for the argument of this article, Shankari Prasad settled the narrow question and left the wide one untouched. It held that Article 13 does not limit the amending power. It did not hold that nothing limits the amending power. Nobody in 1951 had a reason to ask the wider question, because the amendments being made, however contested, were plainly amendments rather than replacements. The wider question arrived only when the amending power was used at a scale that made the difference matter, and the answer it eventually received is set out in this series’ treatment of what Kesavananda Bharati actually decided.

What the passage record predicted

A legislative history is worth writing only if the record tells you something about how the statute would behave afterwards. The record here predicted four things, and it was right about all four.

It predicted that the Ninth Schedule would grow. A device that offers immunity from fundamental rights review, created by a government that had just discovered how inconvenient such review can be, was never going to remain confined to thirteen agrarian statutes. Mookerjee said so at the time. The schedule expanded across the following decades to hold statutes concerning trading, mining, urban land, industrial regulation and much else, and its most aggressive expansion came during the Emergency. What eventually contained it was not parliamentary restraint but judicial doctrine.

It predicted that public order would become the workhorse ground for restricting speech. A ground that is wide, undefined, and satisfied by an administrative apprehension of disorder will be used more often than a ground requiring proof of a threat to the security of the State. That is exactly what happened, and the litigation that followed is essentially the story of courts building the proximity and proportionality requirements that the constitutional text does not contain in terms.

It predicted that Article 15(4) would generate more litigation than the provision it rescued. An enabling clause with no ceiling, no definition of backwardness, and no method for identifying beneficiaries hands every one of those questions to the courts. The Champakam problem was solved in six weeks; the questions that Article 15(4) opened have never been closed.

And it predicted the reactive habit itself. A government that discovers within sixteen months that an adverse judgment can be answered by a constitutional amendment passed in six weeks has learned something it will not unlearn, and it did not. The clearest confirmation is the frequency and character of the amendments that followed through the fifties and sixties, most of which followed the same trigger-and-repair structure. The wider legislative style of the period, of which this amendment is the founding instance, is examined in this series’ account of Nehru-era legislation and the planned state.

Testing the four claims that dominate the argument

The First Amendment attracts more assertion per paragraph than any other constitutional change in India, and four claims recur constantly enough to be worth examining directly against the record.

Did the First Amendment abolish free speech?

No. It widened the grounds on which speech may be restricted, which is a real narrowing of the right and a fair basis for criticism. It also inserted the reasonableness requirement, which made every restriction judicially testable and is the foundation of the free speech litigation that followed. Both statements are true.

The claim that Nehru abolished, destroyed or gutted free expression is not supportable, and it is not made more supportable by the seriousness of what the amendment did do. Speech remained a fundamental right, enforceable under Article 32, subject to restrictions that had to be both within an enumerated ground and reasonable. What changed is that the enumerated grounds became wider and included at least one, friendly relations with foreign states, that has no persuasive public interest justification.

The mirror-image claim, that the amendment was harmless because it merely clarified what the framers intended, is equally unsupportable. The government did not describe it as a clarification. It described it as a correction of judicial readings it could not work with, and the two press decisions it was correcting had struck down censorship of publications, which is about as central to press freedom as a case can get.

The defensible position lies between, and it is not a fudge: the amendment traded a substantial widening of state power over speech for a substantial strengthening of judicial control over how that power is exercised, and whether that trade was worth making depends on how much weight one gives to what courts have subsequently done with the reasonableness standard.

Was the amendment really only about land reform?

No, and the defenders who say so are relying on the strongest part of the amendment to carry the weakest. The land reform justification is genuinely strong. The agrarian programme was the central redistributive commitment of the new state, it commanded support well beyond the governing party, the litigation was real and was producing inconsistent outcomes across provinces, and the compensation formulae were exposed to an equality challenge that could have destroyed the programme through a doctrine having nothing to do with property at all.

But the free speech clause was not about land, and it did not become about land because it travelled in the same bill. It was a separate response to two separate judgments about press censorship, and the government said so in its own statement of reasons. Treating the whole amendment as an agrarian measure is the mirror of treating the whole amendment as a censorship measure. Both work by picking one strand and calling it the bill.

Was the Ninth Schedule always a device for evading rights?

Not as enacted, and this distinction matters. As enacted the schedule was a narrow validation device carrying thirteen agrarian statutes, addressing a specific and identifiable problem. The claim that it was designed from the outset as a general-purpose immunity chamber is not supported by the original list or by the debate.

The claim that survives is different and more damaging. The schedule was designed as a device that could be used that way, with no limiting principle written into it, no subject-matter restriction, and no sunset. A validation mechanism that operates by listing statutes has no natural stopping point, and Parliament faced no textual obstacle to listing anything it wished. The design flaw was structural rather than motivational, and the eventual constitutional response, which subjected post-1973 additions to the schedule to review against the core fundamental rights, was a response to the structure rather than to any bad faith in 1951.

Did the Provisional Parliament have the authority to amend at all?

Legally, yes, and this is not seriously arguable. The Constitution expressly continued the Constituent Assembly as the Provisional Parliament and conferred the powers of Parliament on it, including the amending power. No court has held otherwise and there is no textual foothold for the contrary view.

Politically and morally, the question is entirely open, and treating the legal answer as though it disposed of the political one is a common move that does not work. Legal competence and democratic legitimacy are different properties. A body can have the first and lack the second. Recognising that is what allows a reader to hold two positions at once without contradiction: that the First Amendment was validly enacted, and that a government which chose to narrow fundamental rights six months before an election it was confident of winning made a choice that deserves to be criticised on its own terms.

What is left of the First Amendment

Almost all of it, which is the fact that should end any suggestion that the amendment was a repudiated aberration.

Article 15(4) is in force and is the constitutional basis for reservation in educational institutions. The replaced Article 19(2) is in force and its reasonableness limb is invoked in essentially every free speech case decided in India. Article 19(6) as amended continues to protect state trading. Article 31A is in force. Article 31B and the Ninth Schedule are in force, though the immunity they confer is no longer absolute for statutes added after the date the courts fixed as the dividing line.

The only significant part of the constitutional architecture the amendment touched that is gone is Article 31 itself, which was repealed later when the right to property was moved out of Part III altogether, and that repeal was the work of a different Parliament in a different decade for entirely different reasons.

An honest verdict on the amendment therefore has to reckon with durability. This was not a measure that a later generation reversed once the political weather changed. It is load-bearing constitutional architecture, relied on daily by governments and litigants of every political persuasion, including by those most inclined to denounce the circumstances of its passage. Readers who want to hold the whole cluster in view, from the amending procedure through the doctrinal limits that were eventually imposed on it, will find it useful to keep their own linked notes on each amendment and the litigation attached to it, and VaultBook is built precisely for that kind of private, offline-first research file that grows across a long course of reading.

The verdict

The First Amendment was a competent, narrow, fast piece of repair work that solved the government’s immediate problems and created two long-term ones.

It solved them well. The agrarian programme proceeded. The public order statutes became enforceable. The Madras admissions settlement was restored on a constitutional footing that could be defended. Judged as an exercise in legislative problem-solving, it did what it set out to do, within six weeks, without losing a single major clause.

The first long-term problem it created is the Ninth Schedule, a device with no limiting principle that was always going to be used beyond its occasion and duly was. The second is the reactive habit itself, the assumption that an adverse constitutional judgment is a problem to be legislated around rather than a limit to be respected. That habit produced the amendments of the sixties and seventies, and it produced the confrontation that eventually forced the courts to construct a limit on the amending power that the text does not contain.

There is a real irony in that sequence, and it is the strongest thing that can be said against the amendment. The basic structure doctrine exists because the amending power was used the way it was used starting in 1951. A government that had proceeded more slowly, litigated its way through the property cases, and redrafted its statutes rather than the Constitution might well have ended up with a constitution that was more amendable, not less, because the courts would never have needed to build a defence against constitutional self-destruction. The First Amendment got what it wanted and, over two decades, cost future Parliaments something larger.

That is the reckoning this article offers, and it is offered as analysis rather than as a settled account of a contested episode. Aspirants preparing for the polity papers should note that this amendment is examined more often than any other single amendment except the 42nd and the 44th, usually through the Champakam link, the Ninth Schedule origin, or the Provisional Parliament fact, and working through how the question has actually been framed is more useful than memorising the clause list, which is why it is worth testing yourself against previous-year questions on the ReportMedic explorer and building a daily practice habit on ReportMedic rather than rereading notes.

The six weeks, reconstructed

Setting the sequence out in order makes the pace visible in a way that a summary does not.

The Constitution came into force on 26 January 1950. The two press decisions were delivered in the course of 1950, both striking down state action against publications. The zamindari challenges were argued and decided in the High Courts across 1950 and into 1951, with the Bihar outcome the most damaging to the government. The Madras admissions decision came in the first half of 1951. By the spring of that year the government had lost, in different forums, on speech, on equality in education, and on the constitutional durability of its agrarian programme.

Nehru moved the amendment bill on 10 May 1951. The House debated it and referred it to a select committee. The committee worked through the second half of May and reported, with six members recording minutes of dissent. The House took the reported bill up at the very end of May and into the first days of June, when Nehru made his principal defence of the measure. The bill passed. The President assented on 18 June 1951, and the amendment took effect from that date.

Thirty-nine days from introduction to assent, for a measure that altered the free speech guarantee, the equality guarantee and the property guarantee, and created a new schedule of immunised statutes.

That number is the most eloquent fact in the whole episode, and it is eloquent in both directions. It demonstrates that the amending procedure, as designed, imposes no meaningful delay on a government with the numbers. It also demonstrates that the framers’ choice of a two-thirds special majority rather than a referendum or a constitutional convention was a choice with consequences, and that those consequences became apparent almost immediately rather than after decades of drift. Whether thirty-nine days is scandalous or simply efficient depends on whether one thinks constitutional change should be difficult in itself or merely difficult to command a supermajority for, and reasonable constitutional theorists have taken both positions.

What the number does settle is the standard defence that the amendment was thoroughly deliberated. It was debated, it was scrutinised by a committee, it attracted recorded dissent, and it was amended in committee. All of that happened inside thirty-nine days, alongside the ordinary business of the House, in a legislature that also had a general election to prepare for.

Article 31A: the definition problem and the assent proviso

Article 31A is the less famous of the two property articles the amendment inserted and, in the working life of the courts, it did more sustained work than Article 31B.

The article protected a law providing for the acquisition by the state of an estate or of rights in an estate, or for the extinguishment or modification of such rights, from being struck down on the ground that it was inconsistent with the fundamental rights. That looks like a broad shield until one notices that everything depends on the meaning of the word estate, and the definition became a battleground almost at once. Tenures across India were organised under a bewildering variety of local systems with local names, and whether a particular category of holding fell inside the protected definition determined whether the statute abolishing it was immune from challenge. Litigation over the boundaries of that definition continued for years and prompted further amendments to extend and clarify it, which is a small illustration of a general truth about protective drafting: an immunity clause that operates by category invites litigation about the category.

The proviso is the more interesting half. State legislation of this kind attracts the protection of the article only where the law, having been reserved for the consideration of the President, has received presidential assent. That requirement matters for three reasons that are rarely drawn out.

It gave the Centre a supervisory grip on the agrarian programme. Land was a state subject and the abolition statutes were state statutes, but the constitutional immunity that made them safe was available only where the Centre, through the President acting on ministerial advice, had signed off. A state government that wanted the protection had to negotiate for it.

It also gave a reader a reliable way to test whether a particular state statute enjoyed the protection at all. A practitioner encountering a state land statute cannot assume immunity; the reservation and assent history has to be checked, and the absence of assent is fatal to the claim of protection under the article.

And it demonstrates that the amendment was not the crude instrument it is often described as. The drafters built a conditional immunity with a federal check inside it, which is a more careful piece of constitutional engineering than the Ninth Schedule beside it, where protection follows automatically from listing and no condition applies at all. The contrast between the two articles inserted by the same bill on the same day is the clearest evidence that the government understood the difference between a limited device and an unlimited one, and chose to enact both.

Retrospectivity: the drafting move that did the real work

Three separate limbs of the amendment operate backwards in time, and the retrospective operation is where most of the practical effect sits.

The replacement of Article 19(2) was given retrospective effect so that state public safety and public order statutes already on the books, some of which had been held bad, would be treated as valid. Article 31A was inserted with words providing that it should be deemed always to have been inserted. Article 31B validated the listed statutes not merely going forward but declared that they should not be deemed void or ever to have become void, and did so notwithstanding any contrary judgment, decree or order of any court.

The cumulative effect is a constitutional amendment that reaches back and alters the legal consequences of decisions already delivered. That is a serious thing to do, and it is worth naming plainly rather than leaving buried in the drafting.

It is also, on the government’s side of the argument, the only thing that would have worked. Prospective validation would have left every statute already struck down still struck down, every proprietor already successful still successful, and the agrarian settlement varying province by province according to the accident of litigation timing. If the objective was uniformity, retrospectivity was not an aggravation of the measure; it was the measure.

For the reader trying to assess the amendment, the retrospective limbs are the sharpest test of where they stand. A person who thinks constitutional amendment is an ordinary instrument of government will find nothing troubling here. A person who thinks the point of a rights chapter is that a citizen who has won in court has actually won will find the retrospective validation of statutes notwithstanding contrary judgments to be the single most objectionable thing the amendment did, more objectionable than the widening of the speech grounds, because it concerns not the scope of a right but whether a judicial determination of a right means anything at all.

What happened in the provinces after June 1951

The amendment worked, and the evidence for that is administrative rather than judicial.

The abolition statutes listed in the Ninth Schedule ceased to be litigable on fundamental rights grounds, and the machinery of abolition moved forward. Intermediary tenures were extinguished across the northern and central states over the following years, revenue administration was reorganised to deal directly with cultivators, and the compensation was paid on the graduated formulae the courts had questioned. The programme had defects that are properly the subject of the land reform literature, including the widespread transfer of land into the personal cultivation category ahead of the operative dates and the near-total failure of the tenancy reform and ceiling components in most states, but the abolition of intermediaries itself was carried through, and it was carried through because the constitutional obstacle had been removed.

On the speech side the effect was equally practical and less visible. State public safety statutes were restored to operation, and the ground of public order became the standard basis on which orders restricting publication, procession and assembly were made. The litigation that followed shifted from whether the state had a permissible ground to whether the restriction it had imposed was reasonable, which is exactly the shift the amended clause was drafted to produce, and it is the shape of Indian free speech litigation to this day.

On the education side, Madras restored a scheme of allocation on the footing of the new enabling clause, and the practice spread. Every subsequent question about how far affirmative action in education may go, who counts as backward, whether there is a ceiling, and how the identification of beneficiaries is to be reviewed, has been argued in the space Article 15(4) opened.

The general point is that the three strands of the amendment did not merely settle three lawsuits. Each of them set the terms of a field of Indian public law that is still being argued about, which is the strongest available answer to anyone who treats the First Amendment as a historical curiosity.

What the First Amendment still does in a courtroom

The practical payload is easy to state and easy to underestimate.

An advocate arguing any free speech matter is arguing inside the clause this amendment wrote. The two questions are whether the restriction falls within one of the enumerated grounds and whether it is reasonable, and both limbs come from the replaced Article 19(2). The proximity requirement between the speech and the apprehended disorder, the insistence that the connection not be remote or conjectural, and the refusal to accept a bare administrative assertion of an apprehension of disorder are all judicial constructions built on the reasonableness limb, and none of them would have a textual home in the original clause.

An advocate dealing with a state land statute has to establish whether it is protected by Article 31A, which means checking whether the subject matter falls within the estate definition as it now stands and whether the statute was reserved and received presidential assent, or whether it is protected by listing in the Ninth Schedule, which raises the separate question of when it was added and therefore what standard of review applies to it. These are not academic distinctions; they determine which grounds of challenge remain open.

An advocate on an admissions or reservation matter is working with an enabling clause, which means the state’s power is discretionary and the challenge is almost never to the existence of the power but to its exercise, the identification of beneficiaries, the quantum, and the evidence base. Framing such a challenge as an attack on the constitutional competence to reserve at all is a wasted argument, and it is made surprisingly often.

Because these questions turn on statute-specific facts such as assent history, insertion dates and the precise terms of the order under challenge, the general position described here is not a substitute for advice on an individual matter, and limitation and forum rules in constitutional litigation are unforgiving of approximation.

Five errors in the standard account

The standard treatments of this amendment repeat five errors consistently, and identifying them is a fair test of whether a source has done any work.

The first is dating the amendment to an elected Parliament. It was passed by the Provisional Parliament before the first general election, and any account that describes the Lok Sabha passing it has not checked.

The second is describing Article 15(4) as creating reservation in education. It enables special provision; it does not create, require, or quantify anything, and the difference is the reason the clause has generated so much litigation.

The third is treating the Ninth Schedule as having been designed open-ended. It was created with an agrarian list and grew later. The criticism that survives is about the absence of a limiting principle in the design, not about the original intent, and the stronger criticism is the structural one.

The fourth is presenting the amendment as a single act with a single motive. It answered three unrelated lines of litigation and also carried routine machinery corrections, and any account that gives it one purpose has chosen a strand and called it the whole.

The fifth is omitting the reasonableness requirement when describing what happened to Article 19. This omission is nearly universal in polemical writing on both sides, because the requirement is inconvenient for the argument that the amendment simply curtailed speech and equally inconvenient for the argument that it merely clarified the framers’ intent.

An account that avoids all five is rare enough that a reader can use the list as a quality filter on anything else they read about this amendment.

The consultation question, which is separate from the merits

One strand of the contemporary objection had nothing to do with whether the amendment was right and everything to do with how it was made, and it is the strand most worth recovering because it has almost disappeared from the modern argument.

The Speaker, the newspaper editors, and several members pressed for the bill to be circulated for public opinion before it was taken up, which was and remains a recognised parliamentary route for a measure of unusual significance. The argument was not that the government would lose such a consultation. It was that a constitutional amendment narrowing rights in a republic sixteen months old ought to be seen to have been made in the open, with the affected interests heard, so that the resulting text would carry an authority that a rapid party-whipped passage could not give it.

The government declined, and the reason it gave was the reason it gave for everything about this bill: delay was the harm being avoided.

There is a real cost to that choice that is separate from the cost of the amendment itself. A constitutional change made after visible consultation is harder to attack later, because the critic has to argue against a process as well as a text. A constitutional change made in thirty-nine days by a legislature nobody had elected is permanently vulnerable to the charge that it was imposed, regardless of how defensible its content turns out to be. The First Amendment has been carrying that vulnerability for its entire life, and it will carry it indefinitely, and the government of the day chose that outcome in exchange for a few months.

This is also the origin of a recurring feature of Indian legislative practice that later Parliaments repeated with less excuse: significant constitutional and statutory changes taken up and passed at a pace that precludes meaningful scrutiny, defended on the ground that the problem being solved is urgent. The pattern has been followed by governments of every political composition, and the precedent that made it respectable was set here. The wider question of how bills are scrutinised, when committees are used and when they are bypassed, belongs to this series’ account of how a bill becomes law in India and is not re-argued here.

What the amendment reveals about the Constituent Assembly’s own view of Part III

There is a reading of this episode that is more unsettling than the partisan one, and it deserves a hearing because it complicates everybody’s position.

The people who wrote the fundamental rights chapter and the people who narrowed it sixteen months later were substantially the same people, sitting in the same room, under a different name. That fact is usually deployed as an accusation, and it works well enough as one. But it can also be read as evidence about what the framers thought they had done.

On this reading, the Assembly had never regarded Part III as a set of absolutes. It had drafted a rights chapter alongside a directive principles chapter that committed the state to redistribution, and it had done so knowing the two would collide. It had written a property provision that was the product of the most difficult compromise in the entire drafting process, precisely because the members knew that agrarian legislation was coming. It had written a free speech clause with an enumerated list of restrictions rather than a general guarantee, which is the drafting choice of people who expect restriction to be normal.

When the courts read those provisions more protectively than the drafters had expected, the drafters, now sitting as a legislature, corrected the reading. Whether that is a scandal or simply the framers explaining themselves depends on a prior question about what a constitution is: whether the text, once promulgated, acquires a meaning independent of its authors, or whether the authors remain the best evidence of what it means.

The Indian answer to that question was settled against the framers, and it was settled by the same institution that had provoked the amendment. The courts proceeded on the basis that the Constitution means what it says as interpreted by the judiciary, not what its authors say it means, and within two decades they had constructed a limit on the amending power that no framer had contemplated. The First Amendment is therefore both the first assertion of the framers’ continuing authority over their text and, in the long run, the episode that most contributed to the loss of that authority.

The counterfactual worth thinking about

Historians earn their keep on counterfactuals, and this one is unusually tractable because the alternatives were all live at the time.

Suppose the government had waited for the general election. It would have won it comfortably, and it would have returned with a Lok Sabha elected on universal adult franchise and a newly constituted Council of States. An amendment passed by that Parliament would have been identical in substance and immune to the mandate objection that has dogged the actual amendment for seventy years. The cost would have been roughly a year of continued uncertainty in the agrarian programme, some further High Court decisions, and the political awkwardness of fighting an election with land reform in constitutional limbo.

Suppose instead the government had accepted the judgments and redrafted. The compensation formulae could have been restructured to survive an equality challenge, at the price of paying more to larger holders or of accepting a slower programme. The public safety statutes could have been redrafted to track the security of the State ground precisely, at the price of losing the ability to act against a wider class of speech. The Madras allocation could have been rebuilt around criteria other than community, which is broadly what the constitutional law of reservation eventually required anyway.

Neither counterfactual is a fantasy. Both were argued at the time by serious people. What the government chose instead was the fastest route, and the choice was not driven by ideology in any interesting sense; it was driven by an assessment that the programme could not survive delay.

The instructive part is what the choice cost in institutional terms rather than in political ones. By demonstrating that an adverse constitutional judgment could be answered in six weeks, the government taught every subsequent government the same lesson, and it taught the courts a corresponding one: that the constitutional text was not going to protect itself, and that if there were to be limits on the amending power, the judiciary would have to construct them. Everything in the Indian constitutional confrontation of the following twenty-five years grows from those two lessons, learned in the summer of 1951.

How this amendment changed the amending power itself

The formal amending procedure was not touched by the First Amendment. What it changed was the practical understanding of what the procedure is for, and that change is more consequential than any textual alteration.

Before May 1951, the amending power in the Indian Constitution was an untested provision whose character was unknown. It might have developed as an instrument reserved for structural change, used rarely, deliberated at length, and treated as constitutionally exceptional. That is broadly how the amending powers of several other constitutional systems have developed, not because their texts require it but because their political cultures settled into that practice early.

After June 1951, the Indian amending power was something else: a normal legislative instrument, available to a government with the numbers, usable in response to a specific judgment, and capable of being deployed within weeks. That understanding was established by practice in the first use of the power, and practice in the first use of a power is disproportionately determinative of everything that follows.

Two consequences flowed from it directly. The number of amendments in India grew rapidly compared to systems where the amending power is treated as exceptional, which is a fact usually reported as evidence of the Constitution’s flexibility and is at least equally evidence of the political culture established in 1951. And the judiciary, watching a power of this character being exercised at this frequency, was pushed steadily towards the position that some limit had to exist independently of the text, because a power that is unlimited in substance and effortless in procedure is not a power a constitution can survive.

That is the deepest sense in which this article’s namable claim holds. The reactive amendment pattern was not just a habit of one government. It defined what the amending power in India is, and everything the courts subsequently built was built in response to it.

The trading amendment nobody mentions

Buried among the rights provisions is a change to Article 19(6) that has done quiet, continuous work ever since and is almost never mentioned in accounts of the amendment.

Article 19(1)(g) guarantees the right to practise any profession or to carry on any occupation, trade or business, and Article 19(6) permits reasonable restrictions on that right in the interests of the general public. The problem the government faced was that its economic programme involved the state entering trades directly and excluding private operators, whether in road transport, in the distribution of commodities, or in industries it intended to nationalise. A private operator excluded by a state monopoly could argue that the exclusion was not a restriction on his trade but its extinction, and that no restriction can be reasonable if it eliminates the right altogether.

The amendment answered that argument by expressly saving laws relating to the carrying on by the state, or by a corporation owned or controlled by the state, of any trade, business, industry or service, whether to the exclusion of citizens or otherwise. The effect is that a state monopoly does not have to justify itself as a reasonable restriction on the trade of those it excludes; the constitutional objection is removed at the threshold.

The significance of this is easy to miss and hard to overstate. It is the constitutional foundation on which the public sector was built. Every state transport undertaking with an exclusive route scheme, every state trading corporation, every nationalised industry that excluded private participants, operated on the footing this clause supplied. A constitutional amendment remembered for what it did to newspapers also underwrote four decades of state economic organisation, and it did so in a subclause that most summaries of the amendment do not even list.

It is also the clearest evidence against reading the First Amendment as a rights measure at all. This clause has nothing to do with any judgment about speech or equality. It is economic architecture, inserted because the state’s commercial ambitions were exposed to constitutional challenge, and it belongs in any honest account of what the bill was for.

Where the Ninth Schedule went

The schedule created for thirteen agrarian statutes did not stay agrarian, and the trajectory is the strongest vindication of the contemporary objection to it.

Successive amendments added statutes to it across the following decades. The subject matter drifted first into adjacent territory, which was defensible, covering further land reform, ceiling and tenancy statutes as the states legislated. It then drifted well beyond land, picking up statutes on trading, industrial regulation, urban property, reservations, and other matters having no connection to the agrarian settlement that had produced the device. The largest single expansion came during the Emergency, when the schedule was used as a general repository for statutes the government wished to insulate.

The constitutional response came in stages and came from the courts rather than from Parliament. The first stage was the recognition that the amending power itself has limits, which meant that placing a statute in the schedule could no longer guarantee immunity if the statute or the amendment placing it there damaged the constitutional core. The second stage was the fixing of a dividing line: statutes added to the schedule after the date of the decision that established the limitation on the amending power became testable against the fundamental rights, while earlier entries retained a stronger protection. The third stage was the working out of the test to be applied to post-dividing-line entries, which is now settled in a form that asks whether the statute violates the core rights guarantees read together.

Two lessons follow, and both are general rather than particular to this schedule.

A validation device with no subject-matter limit written into it will be used for whatever the government of the day wants validated, and the absence of a limiting principle is a design failure regardless of the good faith of the designers. The Ninth Schedule is the clearest illustration in Indian constitutional law that drafting for a present purpose without providing for future misuse is not neutrality; it is a decision, and its costs fall on a later generation.

And where a constitutional device outgrows its occasion, the correction in the Indian system has come from the judiciary and not from Parliament. No Parliament has ever repealed the Ninth Schedule, pruned it, or written a subject-matter limit into it. The containment was entirely judicial, which is why the argument about whether judicial limits on the amending power are legitimate cannot be conducted honestly without acknowledging what happened to this schedule in the absence of any other check.

The equality trap in the land cases

The property strand of the amendment is routinely described as a response to courts insisting on full market compensation for expropriated landlords. That description is wrong, and correcting it changes the assessment of what the government was actually facing.

The compensation provisions of the original Constitution were the product of the most laboured compromise in the entire drafting process, and they had been written by people who knew that abolition statutes were imminent. They did not require the state to pay market value. What they required was that a law providing for compulsory acquisition specify either the compensation or the principles on which it was to be determined, and the framers had gone to some trouble to keep the adequacy of that compensation away from judicial assessment.

The vulnerability the abolition statutes actually exhibited came from a different direction. The Bihar statute classified proprietors by the size of their annual income from the estate and applied sharply different multipliers to each class, so that a large proprietor received proportionately far less than a small one. That structure was deliberate and redistributive, and it was exactly what a graduated approach to expropriation looks like. The Patna High Court held that it offended the equality guarantee, because the classification bore no defensible relation to the object of the legislation once the object was characterised as compensating owners for property taken.

That is a devastating finding for a redistributive programme, and it is devastating precisely because it is not about property at all. If graduated compensation is an equality problem, then every abolition statute in the country with a graduated formula was exposed, and redrafting to survive the challenge would have meant flattening the formula and paying substantially more to the largest holders, which would have consumed the fiscal capacity of the states that had to pay it.

Understanding this changes the reading of Articles 31A and 31B. They were not drafted to defeat a compensation standard; they were drafted to close off the whole of Part III as a source of challenge, including the equality guarantee, because the government had learned that the fatal attack could come from any article in the chapter and not merely from the property provisions. That explains the width of the language chosen, which shields the listed statutes from inconsistency with any of the rights conferred by Part III rather than with the property provisions alone.

It also explains something about the schedule’s later career. A device built to defeat challenges from any direction is a device with no natural subject-matter boundary, and the reason it had none was that its designers had been surprised once and drafted to avoid being surprised again. Overbreadth in constitutional drafting usually has a history like this, and it is usually a rational response to a specific defeat rather than an act of malice.

The honest assessment of the property strand is therefore more favourable to the government than the standard account and more damaging in a different way. The problem was real, the diagnosis was correct, and the instrument chosen was wider than the problem required because the government wanted certainty rather than proportion. Certainty was purchased, and the price was a constitutional device that later Parliaments found convenient for purposes nobody in 1951 had in mind.

A note on the name, because it causes real confusion

The phrase First Amendment carries a meaning in one constitutional tradition that has nothing to do with the Indian instrument, and the collision produces a steady stream of confused searching and confused argument.

In the American constitutional vocabulary the First Amendment is a protection of speech, religion and assembly, part of a bill of rights added to a constitution that had been ratified without one. In the Indian constitutional vocabulary the First Amendment is a restriction of speech, an enabling of affirmative action, and a shield for land redistribution, made to a constitution that had been promulgated with an unusually strong rights chapter already in it. One expands rights against the state; the other adjusts them.

The difference is not a curiosity. It reflects two different constitutional problems. The American amendment answers a founding anxiety that a new central government would suppress its critics, and it was drafted by people who had just fought a war against a state they regarded as overbearing. The Indian amendment answers a founding anxiety that a rights chapter enforceable by courts would prevent an elected government from carrying out the redistribution it had been formed to pursue, and it was drafted by people who regarded state capacity as the scarce resource rather than the threat.

Reading either through the other’s assumptions produces nonsense. An Indian reader who imports the American framing will conclude that the First Amendment must have been a rights-protecting measure and will be baffled by the record. An observer working the other way will conclude that Indian constitutionalism was compromised from the outset, without noticing that the Indian Constitution began with an enforceable rights chapter, direct access to the Supreme Court, and a judiciary willing to strike down state action in its first year, none of which the American constitution had at a comparable stage.

The numbering also misleads in a smaller way. The First Amendment to the Indian Constitution is simply the first in a sequence that has run to more than a hundred, and its number carries no special significance. It is remembered because of what it did and when it did it, not because it occupies the first position in a list.

Frequently Asked Questions

Q: Why was the First Amendment passed so soon after the Constitution came into force?

Because the government lost three separate lines of litigation within the Constitution’s first fifteen months and concluded that amendment was faster than redrafting. Two decisions in 1950 struck down state action against publications on the footing that the free speech clause did not permit restriction for public order. A 1951 decision struck down community-wise allocation of seats in professional colleges in Madras. Zamindari abolition statutes were being challenged across several High Courts with inconsistent outcomes, leaving the agrarian programme in doubt. Each problem could in principle have been solved by redrafting the offending statute, but that would have taken years and produced no certainty. A single constitutional amendment fixed all of them at once, retrospectively, and the government moved it thirty-nine days before it received assent.

Q: Which court judgments triggered the First Amendment of 1951?

Three strands of litigation produced the operative clauses. Romesh Thappar v State of Madras and Brij Bhushan v State of Delhi, both decided in 1950, struck down state restrictions on the circulation of a left-wing weekly and pre-censorship of a weekly associated with the Rashtriya Swayamsevak Sangh respectively, and together produced the replacement of Article 19(2). State of Madras v Champakam Dorairajan, decided in 1951, invalidated communal allocation of seats in state-aided professional colleges and produced Article 15(4). The zamindari abolition litigation in the High Courts, most damagingly the Bihar proceedings where the compensation formula was held vulnerable on equality grounds, produced Articles 31A and 31B and the Ninth Schedule. A separate line of challenges to state commercial monopolies produced the amendment to Article 19(6).

Q: Who passed the First Amendment when there was no elected Lok Sabha?

The Provisional Parliament, which was the Constituent Assembly continuing in office as a legislature under the transitional provisions of the Constitution until a Parliament elected under it could be constituted. That body had been elected indirectly in 1946 by provincial legislatures chosen on a restricted franchise, it sat as a single chamber with no Council of States yet in existence, and the Constitution expressly conferred on it the powers of Parliament including the amending power. Its legal competence to amend is not seriously arguable. Whether a body elected on a narrow franchise for the task of framing a constitution should have used the interval before a general election to narrow the fundamental rights it had itself enacted is a political question that was contested at the time and remains contested.

Q: What did the First Amendment change about free speech in Article 19?

It replaced Article 19(2) entirely. The new clause added public order, friendly relations with foreign states, and incitement to an offence to the grounds on which speech may be restricted, which is a substantial expansion of state power, particularly through public order, a concept the Supreme Court had expressly held to be wider than the security of the State. The same clause subjected every ground of restriction to the requirement that the restriction be reasonable, which the original clause did not contain and which made every restriction judicially testable. The clause was given retrospective effect so that public safety statutes already enacted, some already held invalid, would stand. Accounts that mention only the widening, or only the reasonableness requirement, describe half the change.

Q: Why was the Ninth Schedule created?

Because Article 31A alone would have left every land reform statute exposed to a fresh argument about whether it fell within the protected category of estate laws, and the government wanted certainty rather than another round of litigation about definitions. Article 31B, which the schedule serves, validates the listed statutes by name, provides that they shall not be deemed void or ever to have become void on fundamental rights grounds, and operates notwithstanding any contrary judgment, decree or order of any court. The original list carried thirteen entries, all concerned with the abolition of intermediary tenures and with land tenures generally in Bihar, Bombay, Madhya Pradesh, Madras, Uttar Pradesh and Hyderabad. The device was narrow as enacted and carried no limiting principle, which is why it did not stay narrow.

Q: What did the First Amendment add through Article 15(4)?

It added an enabling clause permitting the state to make special provision for the advancement of socially and educationally backward classes of citizens and for the Scheduled Castes and Scheduled Tribes. The clause was drafted in response to the Champakam Dorairajan decision, which had held that community-wise allocation of college seats offended the equality guarantee and the prohibition on denial of admission on grounds of caste, and that directive principles could not override an enforceable fundamental right. Article 15(4) does not create reservation, does not require any state to provide it, and fixes no quantum. It removes the constitutional objection and leaves the decision with the state, which is why the questions it opened about backwardness, ceilings and identification of beneficiaries have been litigated continuously ever since.

Q: Who opposed the First Amendment in the Provisional Parliament?

Shyama Prasad Mookerjee led the opposition on the floor, arguing that a body without a mandate under the Constitution it was amending should not narrow fundamental rights, that adding public order to the permissible grounds handed the executive a general power to suppress criticism, and that the Ninth Schedule was a dangerous device irrespective of the merits of the statutes first placed in it. Six members of the select committee recorded minutes of dissent, which is a substantial number in a House with an overwhelming governing majority. The Speaker, G V Mavalankar, urged that a measure of this kind should not be taken at such speed. The All India Newspaper Editors Conference pressed for wider consultation before the free speech clause was altered, and President Rajendra Prasad raised misgivings privately.

Q: How long did Parliament take to pass the First Amendment?

Thirty-nine days from introduction to assent. Jawaharlal Nehru moved the bill on 10 May 1951. It was debated and referred to a select committee, which worked through the second half of May and reported with six minutes of dissent recorded. The House considered the reported bill at the end of May and in the first days of June, when Nehru made his principal defence of it, and passed it. President Rajendra Prasad assented on 18 June 1951 and the amendment took effect from that date. Within that period the bill altered the free speech guarantee, the equality guarantee and the property guarantee, created a new schedule of immunised statutes, and made a series of unrelated machinery corrections to the Constitution.

Q: What is Article 31A and how is it different from Article 31B?

Article 31A protects a category of laws, while Article 31B protects a list of named statutes. Article 31A provides that a law for the acquisition by the state of an estate, or for the extinguishment or modification of rights in an estate, is not void for inconsistency with the fundamental rights, and where the law is a state law it attracts that protection only if it was reserved for the consideration of the President and received assent. Article 31B validates whatever statutes are listed in the Ninth Schedule, unconditionally, retrospectively, and notwithstanding contrary judgments. The practical differences are that Article 31A turns on whether the subject matter falls within the definition of an estate, which has itself been heavily litigated, and carries a federal check that the Ninth Schedule route does not.

Q: Did the First Amendment apply retrospectively to laws already struck down?

Yes, and the retrospective limbs did most of the practical work. The replacement of Article 19(2) was given retrospective operation so that public safety and public order statutes already enacted would be treated as valid. Article 31A was inserted with words providing that it should be deemed always to have been inserted. Article 31B provides that the listed statutes shall not be deemed void or ever to have become void, and expressly operates notwithstanding any judgment, decree or order of any court to the contrary. The consequence is that determinations already made by courts in favour of litigants were displaced by constitutional amendment. Supporters argue that prospective validation would have left the agrarian position varying province by province according to litigation timing, which was the problem being solved.

Q: Did the Supreme Court uphold the First Amendment in Shankari Prasad?

Yes. In Shankari Prasad Singh Deo v Union of India, decided in 1951, the challenge argued that Article 13 renders void any law that takes away or abridges the fundamental rights, that a constitutional amendment is made by Parliament in the form of an Act, and that an amendment abridging fundamental rights is therefore void. The Court rejected the argument, holding that the word law in Article 13 refers to ordinary legislation made in exercise of legislative power and does not include a constitutional amendment made in exercise of constituent power. The amendment and the Ninth Schedule technique both stood. That holding was doubted in 1965, reversed in 1967, and superseded by the very different framework the Supreme Court settled on in 1973.

Q: Which articles other than 15, 19 and 31 did the First Amendment amend?

Roughly a dozen provisions were touched in all, and several had nothing to do with rights. Beyond the new Article 15(4), the replaced Article 19(2), the amended Article 19(6) and the inserted Articles 31A and 31B, the Act made changes to provisions governing the sessions, prorogation and dissolution arrangements for Parliament and the addresses and messages that accompany them, made parallel changes for state legislatures, touched the provisions concerned with the specification of Scheduled Castes and Scheduled Tribes, and altered transitional provisions dealing with existing laws and their adaptation. These were the ordinary corrections that a newly operative constitution generates once its machinery is used for the first time, and bundling them with the rights clauses made the package harder to oppose cleanly.

Q: Did the select committee change the First Amendment Bill?

Yes, though it did not touch the architecture. Every major limb of the bill as introduced survived into the Act: Article 15(4), the replacement of Article 19(2), the state trading amendment, Articles 31A and 31B, and the Ninth Schedule. The committee stage did substantial work on the free speech clause, tightening the drafting and carrying the reasonableness requirement through the whole of the replaced provision, which is the single most consequential difference between the measure as proposed and the measure as enacted. Six members recorded minutes of dissent, which indicates that the scrutiny was real rather than formal. The committee worked and reported within roughly a fortnight, so the depth of that scrutiny was necessarily limited by the pace the government had set.

Q: Why did President Rajendra Prasad hesitate before giving assent?

He had misgivings about whether a body sitting as the Provisional Parliament, elected on a restricted franchise for the purpose of framing a constitution, was the right legislature to narrow the fundamental rights it had itself enacted, and he raised the question of his own position as the person who would have to assent. He was advised that the head of state acts on the advice of ministers on a bill duly passed and that he must give assent, and he did so on 18 June 1951. The episode is an early and instructive demonstration of a principle the Constitution left implicit: the President has an ample right to be consulted and to warn, and no power to substitute his judgment for the government’s on a bill that has passed.

Q: How many entries did the Ninth Schedule originally contain?

Thirteen, all of them agrarian. The original list carried statutes from Bihar, Bombay, Madhya Pradesh, Madras, Uttar Pradesh and Hyderabad, dealing with the abolition of intermediary tenures such as zamindari and with land tenures more generally. That composition is the strongest argument available to the amendment’s defenders, because it shows a device drafted narrowly around a specific programme rather than a general-purpose immunity chamber. The schedule grew substantially through later amendments, first into adjacent land legislation and then well beyond it into trading, industrial regulation, urban property and other subjects, with the largest single expansion occurring during the Emergency. The criticism that survives is structural: nothing in the design limited what could be listed.

Q: Which parts of the First Amendment remain in force?

Almost all of it, which is the fact that disposes of any suggestion that the amendment was later repudiated. Article 15(4) remains the constitutional basis for special provision in education. The replaced Article 19(2) governs every free speech case decided in India, and its reasonableness limb is the most litigated phrase in the provision. The amended Article 19(6) continues to protect state trading and nationalisation. Article 31A remains, as does Article 31B and the Ninth Schedule, though the immunity conferred on statutes added after the dividing line the courts fixed is no longer absolute. The only part of the architecture the amendment touched that has gone is Article 31 itself, repealed later when the right to property was moved out of Part III.