Where We Left Off
Every step described here was taken under a power the constitution provided. That is the finding of this part and it is worse than the alternative, because nothing had to be broken for any of it to happen.
Part Eight stopped on the night of 25 June 1975 and said that by then everything in this part had already been built. This is the demonstration.
Four things carry into this part, and the first is the last debt this series owes to Part Two.
Detention without trial. Part Two recorded it as the emergency measure of 1950 and said the test of an emergency power is what happens when the emergency ends. Part Three tracked it through a decade in which the justifying conditions disappeared and the power was renewed. Part Eight found it continuously available for twenty-one years and re-enacted as MISA in 1971. Part Two said Part Nine owed the account of what it became. Chapter Two is that account, and it closes the last of Part Two’s four promises.
The machinery already existed. Part Eight’s Chapter Seven took an inventory on 11 June 1975 and found eight instruments on the shelf, of which six predated 1966 and four predated independence. Part Eight’s test was: name an instrument used after 25 June that did not exist before 12 June. There is none.
The basic structure doctrine had survived. Kesavananda was decided seven to six in April 1973 and stood. Part Eight called it the only check in that part that was tested and held. Chapter Four is the attempt to remove it and Chapter Eight is what happened to that attempt.
The Shah Commission. Part Two used it as the worked example of a state that counts, writes the number down, and then removes the number. This part uses it as a primary source, because it is the only systematic official investigation of these months that exists — and Chapter Eight returns to what became of it.
One note about tone. This is the part of the series where the temptation to write with the volume turned up is strongest, and where doing so would cost the most. The material does not need it. Almost every figure in this part comes from a body appointed by an Indian government to investigate an Indian government, and the plainest possible statement of what it found is the most damaging thing available.
How to read this book
The Six Boxes
Six coloured boxes run through the series, each doing one job. Here they are with examples from the material ahead.
A Word Box explains a hard word the moment it first appears.
Habeas corpus: Latin for “you shall have the body”. A court order requiring whoever is holding a person to bring them before the court and justify the detention.
It is the oldest and simplest protection in the common law and it is the one everything else rests on. Every other right depends on being able to ask a court why you are in a cell — because a person who cannot ask that question cannot enforce any right at all.
Why it matters here: Chapter Five is about the case in which the Supreme Court of India held that during an emergency this question could not be asked, even if the detention was illegal.
An In Real Terms box turns a number too big to picture into something with a body.
The Shah Commission put the number of people arrested and detained without trial at about 110,806.
That is a large town. It is more people than most Indian districts had in any single settlement in 1975. Held without charge, without trial, and in many cases without being told why — some for up to eighteen months.
Alongside them, 25,962 government employees were prematurely retired, in many cases for declining to carry out orders they thought unlawful.
Those two figures come from the same source, and that source was appointed by an Indian government to investigate an Indian government.
A How We Actually Know This box shows the physical evidence and then says what it cannot prove.
This part rests more heavily on one source than any other in the series, so the source needs stating up front.
The Shah Commission of Inquiry was appointed in May 1977 under Justice J.C. Shah, a former Chief Justice of India, to examine the period from 25 June 1975 to 21 March 1977. It held around a hundred sittings, examined roughly forty-eight thousand papers, and reported in three parts between 1978 and August 1978.
Its strengths: it had powers to summon, it took evidence on oath, it worked within two years of the events, and its authors were not politicians.
Its limits, which Part Two already set out: it questioned officials who knew they were being questioned, by a body appointed by the successor government of their opponents. People being investigated shade their answers, and people investigating a defeated government have an audience.
And the fact that governs everything: after 1980 the report became unobtainable in India. Copies were withdrawn. It was reconstructed decades later from a set that had survived abroad and in private hands. Chapter Eight returns to this.
An Argument box appears where serious people disagree, each side at its strongest.
Everything in this part turns on this and the defence is stronger than it is usually allowed to be.
Inflation above twenty per cent. A railway strike that had threatened the movement of food. Two states in open agitation, one of which had brought down its government. A movement that had shifted from demanding policy changes to demanding the resignation of a government elected four years earlier with a landslide. And on 25 June a national leader publicly urging the police and the armed forces to decide for themselves which orders to obey. No government anywhere treats that last item as ordinary politics, and a country with India’s fragility in 1975 had every reason to treat it as less ordinary than most.
Every one of those conditions existed in May 1975 and produced no proclamation. What changed on 12 June was a judgment against one person, and what changed on 24 June was a Supreme Court stay that let her stay in office. The proclamation came the night after that stay. If the emergency was the country’s, it should have been declared when the country’s condition warranted it. If it was declared thirteen days after a court unseated the Prime Minister and one day after a court let her stay, the burden is on the government to explain the coincidence, and it never has.
What would settle it: the cabinet record of 25 and 26 June 1975. Not released.
Why people care so much: because a genuine emergency badly handled and a personal emergency dressed as a national one are different offences, and Indian politics has needed different answers at different times.
A Hidden Assumption box digs out a belief sitting underneath an argument that both sides accept without discussing.
Almost everybody who writes about these twenty-one months assumes that the Emergency was unconstitutional.
It is in the standard vocabulary. The constitution was suspended. Democracy was overthrown. The rule of law was set aside. Critics say it and defenders answer within the same frame, arguing that the measures were lawful responses to a lawful proclamation.
Almost nothing in this part was unconstitutional. That is the finding.
The proclamation was made under Article 352, which the constitution provides. The suspension of Article 19 came under Article 358, which the constitution provides. The suspension of the right to move courts for Articles 14, 21 and 22 came under Article 359, which the constitution provides. Detention without trial was authorised by Article 22, which the constitution provides, and legislated by a Parliament elected in 1971. The amendments were passed under Article 368 by the required majorities. And when the detentions were challenged, the Supreme Court of India held in Chapter Five that the challenge was not maintainable.
Every instrument was in the document. Every threshold was met. Every procedure was followed.
This is worse than the alternative and it is important to see why. If the Emergency had required somebody to break something, then the lesson would be that India needs better guardians of its constitution. Because it required nobody to break anything, the lesson is about the document — that a constitution containing an emergency power with no external threshold, a detention power with no expiry, and an amendment procedure reachable on a minority of the vote does not need to be violated in order to produce this.
Both sides avoid the finding for opposite reasons. Critics need a violation, because a violation has villains and can be prevented by vigilance. Defenders need the legality to be a defence, and it is not one — it is the charge, and it is the charge against the drafters as much as against the government.
And every chapter closes with a Remember This box: the chapter in the plainest words available.
Part Two’s last debt is paid in Chapter Two: what detention without trial became. About 110,806 people were detained, and 25,962 government employees prematurely retired.
This part leans on the Shah Commission — appointed by an Indian government to investigate an Indian government, and withdrawn from circulation after 1980.
Almost nothing here was unconstitutional. Every instrument was in the document, every threshold was met, every procedure was followed. That is worse than the alternative, and it is a finding about the document.
1The Night of 25 June
The order in which things happened that night tells you almost everything about what kind of decision it was.
1.1 — The sequence
The proclamation of emergency was issued under Article 352 on the ground of internal disturbance. It was signed by President Fakhruddin Ali Ahmed on the night of 25 June 1975, on the advice of the Prime Minister.
The Union Cabinet was informed afterwards. It met the following morning and ratified a decision that had already taken legal effect.
That sequence is not a technicality. Under the constitution as it then stood, the President acted on the advice of the Prime Minister, and there was no requirement that the Cabinet as a body advise a proclamation. The gap was in the document, and it was used.
It was also closed afterwards. The Forty-Fourth Amendment of 1978 required that the President act only on the written advice of the Cabinet, precisely because of this night. Chapter Nine returns to what was and was not repaired.
1.2 — Two in the morning
Before the arrests, the electricity.
At about two o’clock on the morning of 26 June the power supply to the newspaper presses on Bahadur Shah Zafar Marg in Delhi was cut. The Shah Commission established the chain: an instruction from the Lieutenant Governor of Delhi, conveyed to the General Manager of the Delhi Power Supply Corporation.
Not a censorship order. Not a legal notice. A switch.
The effect was that on the morning of 26 June, when Indians would have read about a proclamation made overnight, most Delhi papers did not appear.
1.3 — Before dawn
The arrests began the same night. Something in the order of 676 people were taken in the first hours.
They included Jayaprakash Narayan, whose rally that evening is described in Part Eight’s Chapter Eight; Morarji Desai, a former Deputy Prime Minister; Atal Bihari Vajpayee and L.K. Advani; and Charan Singh — men who would between them furnish India with three future Prime Ministers.
On 27 June two further orders followed. Article 358 suspended the protections of Article 19 — speech, expression, assembly and movement. Article 359 suspended the right to move any court for the enforcement of Articles 14, 21 and 22 — equality before the law, life and personal liberty, and protection against arbitrary detention.
Read those two sentences together and the position on the morning of 27 June is complete. You could be detained without charge, you could not publish, and you could not ask a court why.
Article 352: the provision allowing the President to proclaim a national emergency where the security of India is threatened by war, external aggression or — as it read in 1975 — internal disturbance.
The two words matter. “Internal disturbance” has no definition, no threshold, and nobody outside the government to certify it. Any level of political trouble can be described that way by somebody who wants to.
The Forty-Fourth Amendment replaced it in 1978 with armed rebellion, which is a much higher and far more checkable bar. That single change is the most valuable thing done in the aftermath.
Why it matters here: the ground for the proclamation was a phrase with no content, and everything else in this part follows lawfully from it.
Before going further it is worth being clear about how much of that sequence is established and how firmly, because the chapter’s argument depends on the order of events rather than on anybody’s characterisation of them.
Three claims in this chapter need separating by strength.
Solid: the proclamation itself, its date, its constitutional basis and the orders of 27 June. All are gazetted instruments and their texts are published.
Solid: the power cut. The Shah Commission traced the chain of instruction on oath, from the Lieutenant Governor of Delhi to a named officer of the electricity undertaking, and the fact that most Delhi papers did not appear on 26 June is verifiable from the papers themselves.
Documented: that the Cabinet was informed after the fact rather than before. This rests on the Commission’s findings and on the accounts of ministers, and it is not disputed by any of them.
What none of it can show: who was on the list, or who made it. Some 676 arrests in a night require a list prepared in advance, which means the operation was planned some time before it was executed. How far before is not established, and it is the question that would most illuminate Chapter Nine’s argument about the night of 25 June.
What that looked like from the inside is worth reconstructing, because nothing about it resembled the way such nights are usually imagined.
Imagine going to sleep in a country with a free press, courts you can petition, and an opposition, and waking up in one without any of them — with no tanks in the street, no announcement interrupting the radio, and no soldier anywhere.
What had happened overnight was a signature, a telephone call to an electricity office, and a list.
For most Indians the first sign was that the newspaper did not arrive. For about six hundred and seventy-six of them it was a knock.
Which points at the way that night is almost universally described, and at what the description gets wrong.
Everybody who describes that night assumes that something was seized.
The language is always of capture. Power was grabbed. The constitution was suspended. Democracy was overthrown. Even sympathetic accounts describe an extraordinary assertion of authority.
Nothing was seized because nothing had to be. Every element of that night was an existing power exercised by the person entitled to exercise it. The proclamation was advised by a Prime Minister who held office lawfully and signed by a President who was obliged to act on her advice. The detentions were made under a statute passed by an elected Parliament in 1971. The suspensions of 27 June were made under two articles written into the constitution in 1950 for exactly this purpose.
The one act of that night with no legal clothing at all was the electricity.
Hold that. In a night that removed a free press, an independent judiciary’s reach and the liberty of thousands of people, the only step that required somebody to do something the law did not authorise was cutting off a power supply — and it was done by telephone, by a civil servant, on a Lieutenant Governor’s instruction.
Which is the whole argument of this part in one image. The constitutional machinery did not have to be broken into. It had a door, the door had a handle, and the only thing anybody had to improvise was the light switch.
One clarification before the summary. Nothing in this chapter argues that the proclamation was invalid. It was made under a power that existed, by people entitled to exercise it, on a ground the constitution then permitted. The chapter’s finding is about how little had to be done, not about whether it was lawful.
The proclamation was signed on the night of 25 June 1975 under Article 352 on the ground of internal disturbance — a phrase with no definition and no external threshold. The Cabinet was informed afterwards.
At two in the morning the power to the Delhi newspaper presses was cut, on an instruction traced by the Shah Commission to the Lieutenant Governor. Around 676 people were arrested before dawn.
On 27 June, Articles 358 and 359 removed free speech and the right to go to court. The only step that night with no legal authority behind it was the electricity.
2The Detentions
Part Two recorded a power created for an emergency in 1950 and asked what it became. This is the answer, and it is the last of the four promises that part made.
2.1 — The debt
Part Two’s Chapter Nine established three things. Preventive detention was written into the Indian constitution itself, at Article 22, as an express exception to the protections in the same article. The Preventive Detention Act was passed in February 1950, weeks after the constitution came into force. And the first people held under it included political opponents.
Part Two’s argument was not that the power was unjustified in 1950. It was that a power justified by an emergency should have carried an expiry, that making it permanent cost nothing to avoid, and that the test would be what happened when the emergency ended.
Part Three found the justifying conditions gone by the end of the 1950s and the power renewed. Part Eight found it re-enacted as the Maintenance of Internal Security Act in 1971 and continuously available for twenty-one years.
This chapter closes the account.
Preventive detention: holding somebody to stop them doing something in future, rather than punishing them for something already done.
The difference from an arrest is total. An arrest is followed by a charge, evidence, a trial and a chance to answer. Preventive detention has none of those, because there is nothing to answer — the case is about what an official believes you might do.
Which means there is no way to be innocent of it. You cannot disprove a prediction, and a person held on one has no argument available except that the official is wrong about the future.
Why it matters here: Part Two established that India wrote this into its constitution in 1950, as an express exception inside the article protecting people from arbitrary arrest.
2.2 — The numbers
The Shah Commission’s figure for people arrested and detained without trial across the twenty-one months is about 110,806.
Of these, something in the region of 35,000 were held under MISA specifically, and the remainder under other preventive detention provisions including the Defence of India Rules. Detentions ran for periods up to about eighteen months.
Amnesty International reported over a hundred and ten thousand arrests in the first year and at least twenty-two deaths in custody.
Separately, 25,962 central and state government employees were prematurely retired — a category that included officials who had declined to carry out instructions they regarded as unlawful.
2.3 — Who was held, and why
The Commission’s finding on this is the most important sentence in the chapter and it is not about numbers.
It found that detention orders were frequently not based on any genuine security concern. They were issued to silence political opponents, to prevent public mobilisation, and to create fear. It found that decisions to arrest and to release were often made on political grounds rather than on legal merit.
Those held included opposition politicians, student leaders, trade unionists, journalists, lawyers, and members of the governing party who had objected. People were detained for opposing the sterilisation programme of Chapter Six and the demolitions of Chapter Seven.
The composition is the answer to the argument box in the front matter. An emergency declared because the state faced collapse would produce detentions of people threatening the state. This produced detentions of people criticising the government, which is a different category and was recognised as such by the body that examined it.
2.4 — One case
Numbers of this size stop meaning anything, so one case.
P. Rajan was an engineering student in Kerala. He was taken into police custody in 1976. He died there. His body was never recovered.
His father spent the rest of his life pursuing the case. The habeas corpus petition he brought became one of the defining legal proceedings of the period in that state, and the government of Kerala fell over it afterwards.
It is worth being exact about what makes that case representative rather than exceptional. Not the death — at least twenty-two are recorded and the true number is unknown. What makes it representative is the position his father was in: a man trying to find out where his son was, at a time when Chapter Five’s judgment meant that a court could not require anybody to tell him.
The detention figures come from the Shah Commission, and the reason they are the best available is worth restating: they were compiled by a body with the power to summon officials and take evidence on oath, within two years of the events, from the records of the administration that had done it.
Amnesty International’s figures are an independent check made at the time from outside, and they broadly agree, which is the strongest form of corroboration available for a number like this.
What this cannot show: how many detentions were never recorded as detentions. A person picked up, held for some days and released without an order entering any register does not appear in either figure. Every estimate here is therefore a floor.
And one further limit, which Part Two set out and which applies with full force: the Commission was appointed by the successor government of the detainees’ own political allies. That does not make its documents false — most of what it reports is drawn from the previous government’s own files — but it means the framing was not neutral, and this part does not pretend otherwise.
A figure that size stops registering, so it is worth converting into something a reader can place.
A hundred and eleven thousand people is roughly one in every five thousand Indians alive in 1975.
Put differently: in an average district of that period, several hundred people. In a large town, dozens. Enough that in most parts of northern India, somebody in any given neighbourhood knew somebody who had gone.
That is the point of a number that size. It is not large enough to touch everybody directly. It is exactly large enough that everybody knows it could reach them — which is the effect a detention programme is for, and it is achieved at a fraction of the cost of detaining everybody.
Which brings the last of Part Two’s four promises due.
This box pays the last of Part Two’s four promises. The question is whether the criticism of 1950 is sustained by what happened in 1975.
India in 1950 faced an armed communist insurrection, an assassination conspiracy, millions in camps and organised communal violence. Every democracy has emergency detention powers and most used them in comparable circumstances. Nobody in 1950 could have known that a government twenty-five years later would use them against journalists and students, and holding the framers responsible for that is holding them responsible for the future.
Part Two’s charge was never that the power was wrong in 1950. It was that it was written into the constitution rather than into a statute with a sunset clause — a choice that required no emergency, cost nothing, and would have forced every subsequent government to justify the power afresh instead of merely renewing it. The Constituent Assembly debates record members arguing for exactly that limit. They were overruled, and 1975 is what the absence of the limit permitted.
What would settle it: a comparison with democracies whose emergency detention statutes carried expiry provisions, on the question of whether renewal debates ever defeated one. Some did.
Why people care so much: because successor powers to MISA have existed in India continuously since, under different names, and every government of every party has used them.
That closes the last of the four promises Part Two entered into the record. The privy purses were settled in Part Eight, the Kashmir plebiscite and Article 370 in Part Seven, and detention without trial here. Nothing further is outstanding from that part.
About 110,806 people were arrested and detained without trial, roughly 35,000 of them under MISA, for periods up to eighteen months. At least 22 died in custody. 25,962 government employees were prematurely retired.
The Shah Commission found that detention orders were frequently not based on any genuine security concern, and that arrests and releases were decided on political grounds.
Part Two’s verdict stands. The objection was never to the power in 1950 but to writing it into the constitution without an expiry — and by 1975 nobody had to argue for it, because nobody ever had.
3Censorship
From 26 June every newspaper in India submitted its pages to a government officer before printing. The interesting question is not what was cut but what stopped being written.
3.1 — The mechanism
Pre-publication censorship was imposed from 26 June 1975. Editors were required to submit articles, and in many cases photographs, to government-appointed censors before publication.
The legal basis was Article 358, which had suspended the protections of Article 19 the previous day — the same clause Part Three’s Chapter One examined, in the form the First Amendment of 1951 left it.
Around the censorship sat a set of supporting measures: the Press Council was abolished, the news agencies were merged into a single body, and government advertising, newsprint allocation and accreditation were all available as instruments against a paper that did not comply.
That last group matters more than the censor’s pencil. A newspaper in 1975 India depended on the state for its newsprint quota and for a large share of its advertising revenue. A government does not have to close a paper it can starve.
3.2 — What resistance looked like
Some of it was very good and it deserves recording by name.
The Indian Express is remembered for printing an editorial column that was simply blank. Some papers ran verses from Sanskrit texts or from Tagore whose meaning any literate reader could supply. Obituary notices appeared for the death of Democracy, aged twenty-six. The Statesman and the Indian Express resisted most consistently, and both paid for it.
Journalists were detained, including senior figures whose arrests were widely known.
And some of it was not resistance at all. A great deal of the Indian press complied, and some of it complied enthusiastically. The line usually attributed to a senior figure of the period — that the press was asked to bend and chose to crawl — was directed at that, and it was not unfair.
Pre-publication censorship: a requirement that material be shown to an official and approved before it is printed, as distinct from being punished afterwards.
The distinction is the whole of it. Punishment after publication means somebody has to decide the thing was illegal, in public, with the material available for everybody to judge. Approval before publication means the decision is made in an office, on material nobody else will ever see, by a person who never has to explain.
Nothing suppressed by a pre-publication censor leaves any trace. There is no banned edition to point at, because the edition was never printed.
Why it matters here: this is why the historical record of what Indians were not told between 1975 and 1977 is so thin, and why it will stay thin.
All of which is usually assessed by adding up what was stopped. That is the wrong measure and it understates the effect by a very large margin.
Everybody who writes about the censorship of these months assumes that it is measured by what was forbidden.
The accounts are lists: the stories killed, the editions stopped, the journalists jailed, the blank columns. Defenders count them and say the total was smaller than the legend. Critics count them and say it was larger. Both are counting interventions.
The censor’s real product is not the suppressed article. It is the article nobody proposes.
Think about the working life of a reporter under this system. You have a story. Pursuing it takes a fortnight, costs money the paper does not have, and will end at a desk where a man you have never met will decide whether it exists. He does not have to explain himself and there is no appeal. After the third time, you do not stop being brave. You start having different ideas about what is worth a fortnight.
None of that appears in any count, because a story that was never begun leaves no record of any kind. And it does not stop when the censorship stops, because the habits of a profession are learned in about eighteen months and unlearned much more slowly.
This is why the question “how much was actually censored” is the wrong question, and why both sides of the Indian argument about the press in this period keep answering it. The measurable interventions are the smallest part of the effect, and the largest part is permanently unmeasurable — which is exactly the property that makes pre-publication censorship the instrument of choice for anybody who understands it.
Which leaves a charge that is made freely in India and is usually made too broadly.
The charge is usually made against the press as a whole. It should be made more carefully than that.
Most Indian newspapers complied without significant resistance, and some editorialised in the government’s favour. Newsprint quotas, advertising and accreditation were all commercial pressures rather than physical ones, and a proprietor deciding that his business mattered more than his masthead is making a choice, not submitting to force. The papers that did resist — and there were several — demonstrate that resistance was possible, which removes the excuse from the rest.
An editor who resists loses his newsprint, his advertising and possibly his liberty, and his paper closes — which removes even the compromised journalism his readers were getting. The Emergency did not present the press with a choice between honour and dishonour but between a censored paper and no paper, and reasonable people differed about which served readers better. The ones who resisted were heroic; that does not make the others collaborators.
What would settle it: a paper-by-paper account of what each did — how many submissions, how many refusals, how much favourable comment. The material exists in the editions themselves. Nobody has assembled it.
Why people care so much: because the Indian press has claimed the resistance of 1975 as part of its self-description ever since, and the claim belongs to a minority of it.
One consequence for this book. Everything in Chapters Six and Seven happened while the country could not be told about it — which is why the Shah Commission matters so much, and why a reader in 2026 knows more about the sterilisation programme than an Indian newspaper reader did in 1976.
Pre-publication censorship from 26 June 1975, under Article 358 — the clause the First Amendment of 1951 left in the form Part Three examined.
The censor’s pencil mattered less than newsprint quotas, government advertising and accreditation. A government does not have to close a paper it can starve.
The real product of a censor is the story nobody proposes. It leaves no record, appears in no count, and outlasts the censorship by many years.
4The Forty-Second Amendment
The largest single alteration ever made to the Indian constitution, passed by a Parliament whose opposition was in jail. One clause in it was the point of the whole exercise.
4.1 — The one that came first
Before the Forty-Second there was the Thirty-Ninth, and it is the more revealing of the two because it is so much narrower.
Passed in August 1975, some six weeks after the proclamation, it did two things.
It placed the election of the Prime Minister, the President, the Vice-President and the Speaker beyond the reach of the ordinary courts, providing that such disputes would go to a body Parliament would establish rather than to a High Court, and that pending proceedings would abate.
And it inserted MISA into the Ninth Schedule — Part Three’s Chapter Two device, which places a law beyond challenge on fundamental rights grounds. The detention statute of Chapter Two was thereby put where no court could reach it, in the same weeks it was being used on a hundred thousand people.
The first of those was retrospective and personal. Part Eight’s Chapter Eight described a judgment of 12 June 1975 and an appeal pending before the Supreme Court. The Thirty-Ninth Amendment was drafted to dispose of that appeal by removing the jurisdiction under which it was being heard.
4.2 — What happened to it, on 7 November 1975
The appeal was heard anyway, and this is the passage of this part that most complicates the whole of it.
On 7 November 1975, with the Emergency running, the press censored and the opposition in detention, a five-judge bench of the Supreme Court struck down the election provisions of the Thirty-Ninth Amendment as violating the basic structure of the constitution. Free and fair elections, the Court held, were part of that structure and Parliament could not place them beyond judicial reach.
The bench was presided over by Chief Justice A.N. Ray — appointed in April 1973, as Part Eight’s Chapter Six described, over three senior judges who had established the basic structure doctrine, having himself argued against it.
The Court used the doctrine he had opposed, in a case about the person whose government had elevated him, at the height of her power.
It also upheld her election on the merits, which is the part usually left out. She won the appeal and lost the amendment.
This is the single strongest fact available to any defender of the Indian constitutional order in this period, and Part Eight flagged it forward for exactly this reason. Whatever else was happening in those months, that judgment was delivered, and it stood.
4.3 — The scale
The Constitution (Forty-Second Amendment) Act, 1976 amended the Preamble, altered around forty articles, added fourteen new articles and two new Parts, and changed the Seventh Schedule.
It is known as the Mini-Constitution, and the name is not rhetoric. No other amendment before or since comes close to it in extent.
It was steered through Parliament by the Law Minister, H.R. Gokhale, and drew on the work of a party committee under Swaran Singh.
It was passed by a Parliament sitting on a term that this same amendment extended, with a large part of the opposition detained under Chapter Two.
4.4 — What it did
| What it changed | Effect |
|---|---|
| The Preamble | Added “Socialist” and “Secular”; changed “unity of the Nation” to “unity and integrity of the Nation”. |
| Article 368(4) and (5) | No constitutional amendment could be questioned in any court on any ground, and Parliament’s amending power was declared to have no limitation. Directed squarely at Kesavananda. |
| Article 31C | Extended protection from challenge under Articles 14 and 19 to laws implementing any Directive Principle, not only two — reversing the priority between rights and principles. |
| Terms of legislatures | Lok Sabha and State Assemblies extended from five years to six. |
| Article 356 | President’s Rule in a state extended from six months to one year at a time. |
| Article 352 | An emergency could be proclaimed for part of the country as well as the whole. |
| Article 31D | New. Parliament empowered to legislate against “anti-national activities”, a term the Act itself defined broadly. |
| Article 257A | New. Enabled the centre to deploy armed forces in a state. |
| High Court powers | Judicial review by High Courts curtailed through a set of new articles restricting their jurisdiction. |
| The Seventh Schedule | Education, forests, administration of justice, weights and measures and protection of wildlife moved from the State List to the Concurrent List. |
| Part IVA | New. Ten Fundamental Duties in Article 51A. |
| Directive Principles | Added Article 39A (free legal aid), 43A (worker participation in management) and 48A (protection of the environment). |
| Articles 323A, 323B | New. Administrative and other tribunals, removing categories of dispute from the ordinary courts. |
| Article 74 | Made the President explicitly bound by the advice of the Council of Ministers. |
Imagine a club whose rules say the committee may change the rules, and the members may go to a magistrate if the committee changes them unfairly.
The committee now passes a rule saying no rule change may ever be examined by a magistrate.
Nothing else has to happen. Every future rule is now whatever the committee says it is, including the rules about who sits on the committee and when the next election is held. The rule book still exists, is still printed, and still says a great many admirable things.
It is no longer a rule book. It is a record of the committee’s current preferences.
4.5 — The clause it was for
Read that table again and notice that most of it is arguable. Reasonable people can defend fundamental duties, environmental directives, free legal aid and even tribunals. Several of those provisions are still in force and nobody proposes removing them.
One row is not arguable, and it is the row the amendment existed for.
Article 368(4) and (5). No amendment questionable in any court, on any ground, and no limitation on the power to amend.
Consider what that means as a piece of machinery. Parliament may amend the constitution. Parliament has just declared that no amendment may be reviewed. Therefore Parliament may amend anything at all, including the provisions that constrain Parliament, and no court may look at it.
A constitution with that clause in it is not a constitution. It is a set of arrangements the legislature may rewrite at will, and the word for the document then is not law but minutes.
It was aimed at one judgment. Part Eight’s Chapter Six described Kesavananda Bharati — thirteen judges, seven to six — holding that the amending power could not damage the basic structure. This was the answer, and it was written to be unanswerable.
4.6 — What happened to it
The Forty-Third Amendment of 1977 and the Forty-Fourth Amendment of 1978, passed by the successor government, undid most of the dangerous material. Judicial review was restored. Terms went back to five years. Article 31D went. “Internal disturbance” became armed rebellion. The President was required to act on the written advice of the Cabinet.
Then in Minerva Mills in 1980 the Supreme Court struck down the two provisions that mattered most: the expansion of Article 31C, and clauses (4) and (5) of Article 368.
It struck them down using the basic structure doctrine — the very doctrine those clauses had been written to abolish.
What survived: the Preamble changes, the Fundamental Duties, and the new Directive Principles. All three remain in the constitution today and none is seriously contested.
This is the most straightforwardly verifiable chapter in the entire series. The Forty-Second Amendment is a statute. Anybody can read it, section by section, and check every row of the table against the text.
The same is true of the Forty-Third and Forty-Fourth Amendments, and of Minerva Mills, which is a reported judgment.
There is no interpretation here, no memoir, and no inference. A reader who suspects this chapter of overstatement should test it directly, which takes an afternoon.
What it cannot show: what was said in the drafting, or which provisions were included as cover for the two that mattered. That the amendment bundled uncontroversial improvements with clauses removing judicial review is a fact about its structure; whether the bundling was deliberate is not established.
Which leaves the question of whether a formally valid amendment can be illegitimate, and if so on what basis.
It was passed by the required majorities under the correct procedure. Whether that makes it legitimate is the question.
Parliament holds the amending power and used it. Much of the content is genuinely good: fundamental duties exist in many constitutions, free legal aid and environmental protection are unarguable, and the primacy of Directive Principles reflects a serious position about whether property rights should obstruct redistribution. On the judicial review clauses, there is a real democratic argument — an unelected court had assumed the power to strike down amendments passed by elected representatives, on a doctrine found nowhere in the text, by a single vote.
Take the democratic argument at its strongest and it still fails on its own terms. A Parliament claiming democratic authority to remove judicial review had, at that moment, extended its own term beyond the mandate it was elected on, was sitting with a large part of the opposition in detention, and was legislating under a censorship regime that prevented the country being told what it was doing. Whatever those conditions produce, it is not a democratic mandate. And clauses (4) and (5) do not merely limit a court; they place the amending power beyond any review by anybody, which is a claim no legislature in a constitutional system can coherently make.
What would settle it: the drafting record. Not released.
Why people care so much: because the words “socialist” and “secular” in the Preamble come from this amendment, and arguments about them recur in Indian politics without either side wanting to discuss where they came from.
One thing to carry forward. The Preamble of the Indian constitution today reads as it does — sovereign, socialist, secular, democratic republic — because of an amendment passed during the Emergency, by a Parliament that had extended its own term, with the opposition in detention. Whatever one thinks of the words, that is where they come from, and Indian arguments about them are almost always conducted without it.
The 39th Amendment of August 1975 put the Prime Minister’s election beyond the ordinary courts and MISA into the Ninth Schedule. On 7 November 1975 the Supreme Court struck the election provisions down under the basic structure doctrine — presided over by the Chief Justice appointed to oppose it.
The 42nd Amendment altered forty articles, added fourteen, added two Parts and rewrote the Preamble. It was passed by a Parliament that had extended its own term, with much of the opposition detained and the press censored.
The clause it existed for was Article 368(4) and (5): no amendment questionable in any court, and no limit on the power to amend. A document with that clause in it is not a constitution.
The 44th Amendment undid most of it and Minerva Mills struck down the rest — using the basic structure doctrine those clauses were written to abolish.
5ADM Jabalpur
The Attorney General told the Supreme Court that if a person’s life were taken unlawfully, the courts could do nothing. Four of the five judges agreed.
5.1 — The question
Thousands of people were detained under Chapter Two. Their families did what families do: they went to the High Courts with petitions for habeas corpus, asking the state to produce the detainee and justify the detention.
Several High Courts held that such petitions were maintainable — that even with Article 21 suspended, a court could still ask whether a detention was authorised by the law it claimed to rely on.
The government appealed, and the appeals were heard together as Additional District Magistrate, Jabalpur v. Shivkant Shukla. The question was whether, during the Emergency, a detained person could go to a court at all.
The bench of five was the five most senior judges: Chief Justice A.N. Ray, and Justices M.H. Beg, H.R. Khanna, Y.V. Chandrachud and P.N. Bhagwati.
Part Eight’s Chapter Six established how the first of those names came to be sitting where he was.
5.2 — The exchange at the bar
During argument, Justice Khanna put the question to the Attorney General, Niren De, in its starkest form. If a person were deprived of life unlawfully during the Emergency, would there be any remedy?
The Attorney General’s answer, in substance, was that the courts would be helpless.
He was not being reckless. He was stating the government’s position accurately, and the government’s position followed from the Presidential order under Article 359. If the right to move a court for the enforcement of Article 21 is suspended, and Article 21 is the source of the right to life, then there is nothing to enforce and no court to enforce it in.
The logic is sound. That is what makes the case what it is.
Locus standi: standing — the right to bring a case before a court at all.
It is a threshold question decided before any court considers whether you are right. A person without standing does not lose the argument; the argument is never heard.
This is why the holding in this case is worse than it first appears. The Court did not decide that the detentions were lawful. It decided that the question could not be asked, which forecloses every possible detention, lawful and unlawful alike, and leaves nothing behind for anybody to appeal.
Why it matters here: a right that cannot be brought before a court is not a weakened right. It is not a right.
5.3 — Four judges
The judgment was delivered on 28 April 1976. By four to one, the Court held that during the Emergency no person had any locus to move any court for enforcement of the right to personal liberty, that habeas corpus petitions were not maintainable, and that this held even if the detention was illegal or made in bad faith.
Ray wrote the majority. Beg observed that liberty is a gift of law and may be limited by law. Bhagwati acknowledged that the position was unfortunate and said it could not be helped.
The Court also upheld the provision of MISA barring disclosure of the grounds of detention.
5.4 — One judge
Justice H.R. Khanna dissented.
His position was that the right to life and personal liberty is not the creation of the constitution and cannot be extinguished by suspending a remedy; that the state may not deprive a person of life or liberty without the authority of law even in an emergency; and that what was at stake was whether a court could be silenced entirely.
He met the argument that the detentions were legal by observing that legality alone settles nothing — that a regime can be entirely lawful and entirely monstrous, and that the most notorious example within living memory had been meticulous about its paperwork.
Three days later the New York Times published an editorial saying that if India found its way back to freedom, somebody would erect a monument to Justice Khanna.
5.5 — What it cost him
Khanna was the senior-most judge after the Chief Justice. By the convention Part Eight described — broken once already, in 1973 — the Chief Justiceship would come to him.
He knew what he was doing. He is recorded as having told a family member before delivering the dissent that it would cost him the office.
On 28 January 1977 he went to court as usual. At five o’clock that evening he heard on the radio that M.H. Beg, junior to him and in the majority, had been appointed Chief Justice of India.
He typed his resignation the same evening.
It is worth adding the ending. In 2017, in K.S. Puttaswamy, the Supreme Court overruled ADM Jabalpur outright. Among the judges who did so was D.Y. Chandrachud — the son of Y.V. Chandrachud, who had been in the majority in 1976 — and he said so explicitly.
Establishing what that judgment was and was not is unusually easy, and it matters for the argument that follows.
ADM Jabalpur is a reported judgment. The bench, the split, the reasoning of each judge and the dissent can all be read in full by anybody.
The exchange with the Attorney General is recorded in the judgment and in contemporaneous accounts by lawyers who were in court.
The consequences are equally documented: the date of Beg’s appointment, the date of Khanna’s resignation, and the New York Times editorial two days after the judgment.
What it cannot show: what the four judges believed. Two of them later described the decision as wrong, which tells us something about their view afterwards and nothing reliable about their view at the time. This chapter’s argument does not require an answer to that, and the hidden assumption box explains why.
What the five of them were actually choosing between is worth stating without any legal vocabulary at all.
Four men in that room had lifetime security, the highest professional standing available in India, and no personal danger of any kind. Nobody was going to detain a Supreme Court judge.
What was on offer for agreeing was an office. What was on offer for disagreeing was the loss of it.
One of them wrote the dissent knowing the price, and paid it nine months later while listening to the radio.
The others went on to hold the highest offices in the Indian judiciary. Two of them later described the judgment as a mistake.
Whether that makes the four blameworthy is argued about seriously by people who have spent careers on it, and the defence of them is better than it is usually allowed to be.
This is argued about in Indian legal circles to this day and it is not the easy question it looks.
A Presidential order under Article 359 had suspended the right to move courts for Article 21. That order was validly made under a provision the constitution contains. A judge’s task is to apply the law, not to invent a remedy the text has removed, and the majority reasoning follows from the premises with uncomfortable rigour. Khanna’s position, whatever its moral standing, required finding a source for the right outside the constitution — which is precisely the kind of judicial improvisation that critics of activist courts object to in every other context.
Courts choose their premises constantly, and this Court had done so three years earlier in Kesavananda, where it found a limit on the amending power nowhere stated in the text. A bench willing to derive a basic structure from what a constitution is could have derived the same about a right to life, and one of its members did. The narrower point is that the majority went further than the premises required: they held the position unchallengeable even where a detention was illegal or in bad faith, which no reading of Article 359 compelled.
What would settle it: nothing further. This is a question about a published text and it has been argued for fifty years by people who have read it more carefully than I have.
Why people care so much: because two of the four went on to be Chief Justice, and one of them became the most celebrated judicial reformer of the following decade — which makes the case a permanent difficulty for any account of Indian judicial heroism.
What both of those positions share is an assumption about the circumstances, and it does not survive the fact that one judge in the same room decided otherwise.
Everybody who writes about ADM Jabalpur assumes that the judges were under pressure.
The critical version: a cowed court capitulated to a government that had already superseded three of its members. The sympathetic version: judges in an authoritarian moment did what they could and cannot be blamed for the times. Both explain the judgment by reference to the conditions the judges were working under.
The conditions were real. They were also identical for all five of them, and one of them dissented.
That single fact removes the explanation. Pressure that four men yield to and a fifth does not is not what determined the outcome, because if it were, the fifth man could not have done what he did. Khanna sat in the same room, faced the same government, knew the same history of April 1973, and correctly predicted the same consequence for himself.
So the four were not coerced. They were choosing under known conditions, with a demonstration standing beside them of what the other choice looked like and what it cost.
That is a much harder thing to say and it is the only thing the evidence supports. It also explains the discomfort this case still produces in Indian legal culture: two of the four later said publicly that they had been wrong, which is not what people do about decisions they were forced into.
The general rule is worth carrying forward, because this series keeps meeting institutions that failed at the moment they were needed. When one person in a room does the difficult thing, the conditions stop being an explanation for everybody else in it. The conditions determine the price, not the choice.
Set this chapter beside Part Eight’s Chapter Six and the sequence is complete. In April 1973 three judges were passed over for deciding against the government, and resigned. In April 1976 four of the five judges hearing the most important liberty case in Indian history decided for the government. In January 1977 the one who had not was passed over, and resigned. Three years, two supersessions, and a bench that had been shown exactly what the terms were.
On 28 April 1976, by four to one, the Supreme Court held that during the Emergency no detained person could approach any court, even if the detention was illegal or in bad faith. The Attorney General had told the Court that if life were taken unlawfully, the courts would be helpless.
Justice H.R. Khanna dissented, holding that the right to life is not the creation of the constitution and cannot be extinguished by suspending a remedy. He knew it would cost him the Chief Justiceship, and on 28 January 1977 he heard on the radio that it had.
The conditions were identical for all five and one dissented. When one person in a room does the difficult thing, the conditions stop being an explanation for everybody else in it.
6The Sterilisations
The number tripled in twelve months. Nothing about a population’s willingness to have surgery changes that fast, and that single fact is the whole of this chapter.
6.1 — The figures
Family planning was not invented in 1975. India had run sterilisation programmes since the 1950s, they were voluntary, and international opinion at the time was strongly in favour of population control by almost any means.
Here is what happened to the numbers.
In 1975–76, about 26.42 lakh sterilisations were performed — around 2.6 million.
In 1976–77, about 81.32 lakh — around 8.1 million.
Over the two years, something in the order of 1.07 crore, or about 10.7 million operations.
Look at those first two figures again. The annual number roughly tripled in a single year.
There is no version of a voluntary programme that does this. Public willingness to undergo irreversible surgery does not triple in twelve months. Something else changed, and the Shah Commission established what.
6.2 — How it was done
Two mechanisms, and they worked together.
Quotas. Targets were set and passed down through the administration. Officials — teachers, clerks, revenue staff, police, railway employees — were given numbers to produce and their pay, promotion and postings depended on producing them. A quota does not require anybody at the top to order coercion. It requires somebody at the bottom to find a way of hitting a number.
Linkage. Access to things people could not do without was tied to a sterilisation certificate. Rations. Housing allotments. Government employment. Medical treatment. Loans. Licences. Irrigation water in some places. Salary release in others.
The Shah Commission recorded that people were denied rations, housing, jobs, healthcare and loans for having more than two or three children and declining the operation.
Note what the second mechanism does to the word “voluntary”. A man who signs a consent form to get his family’s ration card has consented in exactly the sense that a man who hands over his wallet to a person holding a knife has made a gift.
Before the mechanisms, it is worth registering what a figure of that size means operationally.
Eight million operations in a year is around twenty-two thousand a day, every day, including the day of every festival and every day of the monsoon.
Consider what that requires of a health system that in 1976 could not reliably deliver vaccines to a village. Camps set up in schoolyards and bus stations. Staff working through lists. Surgery performed by people meeting a number, on people produced by other people meeting a number.
The 1,774 recorded deaths are what a rate like that does to a system not built for it. And that figure counts the deaths somebody recorded, in a year when Chapter Three had removed the mechanism by which anybody else would have found out.
6.3 — What it cost
1,774 deaths were recorded as resulting from these procedures.
That is the official figure, from operations performed at speed, in camps, by staff under quota pressure, often without the conditions or the follow-up that surgery requires.
The Commission also recorded 548 complaints of the forced sterilisation of unmarried people — a category for whom the programme’s own stated rationale does not apply at all.
And there was a further cost that is not in any table. The programme was concentrated in northern India, and it attached to the Indian state a suspicion of public health campaigns that persisted for decades. Immunisation drives in the affected regions were still meeting resistance traceable to 1976 twenty and thirty years later.
Target-driven administration: a system in which officials are set numerical goals and rewarded or punished on whether they meet them.
It is not inherently wrong and it is used everywhere. Its known failure mode is exact: where the target is easy to measure and the quality of the outcome is not, officials optimise for the number.
The people who design the target rarely instruct anybody to cheat, coerce or falsify. They do not have to. The instruction is the number, and everybody below understands what happens if it is not met.
Why it matters here: this is why “nobody ordered coercion” is true and is not a defence. The quota was the order.
Set those two facts together — the quota and the linkage — and the vocabulary usually applied to this programme starts to look wrong. The evidence for all of it comes from an unusual direction, and its strength and its limits are both worth stating.
The sterilisation figures are the government’s own. They were compiled by the health administration as performance data — the whole point of the programme was that the numbers went upward to people checking them.
That makes them unusually reliable for the fact of the operations, and it is why this chapter argues from the annual totals rather than from testimony. A government that inflates a figure to please a superior produces a number that is too high; there is no plausible reason for anybody in that chain to have reported more operations than took place at the very moment the target was the thing being measured — and if they did, the inflation is itself the finding.
The deaths, the complaints about unmarried people, and the linkage of entitlements come from the Shah Commission’s examination of those same records and of officials on oath.
What it cannot show: consent. Every one of these operations has a signed form behind it. A programme that documents consent for coerced procedures destroys the evidence that would allow anybody to audit consent afterwards, permanently.
Which is why the argument about proportion has to be conducted from the shape of the numbers rather than from the paperwork.
The defence of this programme is not that it was gentle. It is about proportion and about what the alternative was.
Ten point seven million operations in a country of over six hundred million is not a national campaign of forced surgery. A very large number of them were performed on people who wanted them — sterilisation was and remains India’s dominant method of contraception, chosen voluntarily by millions. India’s population was growing at over two per cent a year, Part Four showed food output struggling to keep pace, and every serious international body of the period regarded population growth as India’s central problem. A government that did nothing about it would have been indicted for that instead.
Nobody disputes that voluntary sterilisation existed before and after. The question is what the extra six million operations in 1976–77 consisted of, and the answer cannot be a sudden change of heart in one year. What changed was quotas and linkage, and both are documented. Note also which direction the coercion ran: overwhelmingly at poor men, in northern India, by officials with numbers to meet — not at anybody with the standing to refuse.
What would settle it: district-level records of operations against quota assignments. Much of this material was compiled by the Shah Commission and much has since been dispersed.
Why people care so much: because population control is still argued about in India, and every argument for a coercive version has to answer 1976, which is the strongest available evidence about what such programmes do when handed to an administration under pressure.
Which leaves one word to deal with, and it is the word almost every account of this period reaches for.
Almost everybody who writes about these months — critics included — calls what happened in them the excesses.
The word is universal. Emergency excesses. The excesses of 1976. Even the Shah Commission’s terms of reference used the language of excess and abuse of authority.
Part One’s Chapter Two warned about exactly this word. An excess is a legitimate thing done too much. It implies a proper programme, pursued with insufficient restraint, by people who went further than intended. It locates the fault in enthusiasm.
Apply it here and see what it hides. The quota was not an excess; it was the design. The linkage of ration cards to sterilisation certificates was not enthusiasm; it was a policy instrument, adopted deliberately, because it worked. Nobody exceeded anything. Officials met the numbers they were given, using the tools they were given, and were rewarded for it.
The same applies across this whole part. Detaining a hundred and eleven thousand people was not an excess of an emergency; it was what the emergency was for. Censoring the press was not zeal; it was the instrument. The Forty-Second Amendment was not an over-reaction; it was drafted, printed and voted on.
Which is why the vocabulary matters so much. “Excesses” allows an entire period to be regretted without anything being conceded about its design, and it permits the standard formulation still used in Indian politics — that mistakes were made and lessons learned — which names no instrument and requires no change.
Part One said this series would name the word every time it appeared. It appears here more than anywhere else, and this is what it is doing.
Part One promised to name that word every time it appeared in this series. It has now been named at the place where it does the most concealing, and it will be named again in Part Ten.
26.42 lakh sterilisations in 1975–76, then 81.32 lakh in 1976–77 — roughly 10.7 million over two years. The annual number tripled in twelve months, and no voluntary programme does that.
Two mechanisms: quotas passed down to officials whose pay and postings depended on the number, and linkage of rations, housing, jobs, healthcare and loans to a certificate. 1,774 deaths were recorded.
The word “excesses” implies a legitimate programme pursued too far. The quota was not an excess. It was the design.
7What Was Done to Cities
The second programme of these months was carried out on the instructions of a man who held no office of any kind, and officials obeyed him.
7.1 — Turkman Gate
In April 1976 a demolition drive reached the area around Turkman Gate in old Delhi — a dense, poor, largely Muslim quarter of narrow lanes inside the walled city.
Housing was cleared. Residents resisted. Police opened fire. People died.
The official death toll was around a dozen. Claims made afterwards were very much higher, and the true figure is not established — which, as Part One’s rule states, is itself a finding: a state that counted sterilisations to the individual did not produce a reliable count of the people killed by its own police in its own capital.
The survivors were moved across the Yamuna to resettlement colonies on the eastern edge of the city, several kilometres from the work, markets and relationships their lives had been built around.
And the two programmes met. In the same weeks, in the same neighbourhoods, sterilisation quotas were being enforced, and the connection between demolition and the family planning drive was widely understood by the people it happened to. That is a substantial part of why Turkman Gate became the single most politically consequential event of the Emergency.
The distance from the walled city to the resettlement colonies across the Yamuna is a few kilometres on a map. In 1976 it was a bus ride that could take an hour each way, on a service that did not reliably exist.
A tailor whose customers lived in the next lane now has no customers. A woman who cooked for a household four streets away now has no work. A boy at a school ten minutes’ walk away is now at no school.
None of that appears in a compensation calculation, because compensation is assessed on the structure and the structure was worth very little. What was worth something was where it stood, and that is the thing nobody was paid for.
7.2 — Demolition described as improvement
The programme had a respectable name: beautification, slum clearance, urban renewal.
Some of the diagnosis was real. Old Delhi was severely overcrowded, sanitation was poor, and there was a genuine case for rebuilding. Cities do have to be rebuilt and somebody always loses.
What distinguishes this from ordinary urban policy is procedure. There was no consultation, no hearing, no compensation process worth the name, and — because of Chapter Five — no court a resident could go to. A demolition programme in a country with a functioning judiciary produces litigation and delay, which is precisely the mechanism by which the people affected obtain terms. Here that mechanism had been switched off two months earlier.
Resettlement colony: land on a city’s edge to which people cleared from central areas are moved, usually with a plot and little else.
The arithmetic is what matters. A family in a central slum is poor and is within walking distance of the work, the customers, the school and the relatives that make its income possible. Moved fifteen kilometres out, it has a plot it may own and no way to earn.
This is why clearance without consent impoverishes people who were already poor, and why the compensation question in urban policy is never really about the value of the structure demolished. It is about the value of the location.
Why it matters here: the residents of Chapter Seven were moved across a river, and the thing they lost is not visible in any valuation of what was knocked down.
7.3 — The man with no office
Now the structural fact, which matters more than any individual episode.
Sanjay Gandhi held no office. He was not a minister. He was not a member of either House of Parliament. He held no post in the government of India or of Delhi. Through the whole of the period covered by this part he was, constitutionally speaking, a private citizen.
He also had a programme — a set of priorities including family planning, slum clearance, tree planting, adult literacy and the abolition of dowry — which officials across northern India treated as instructions.
The Shah Commission’s finding on this was that directives emanated from the Prime Minister’s household and office and were implemented by a compliant bureaucracy and by party functionaries.
Consider what that means as a question about the state rather than about a family. A senior civil servant receiving an instruction from a person with no office has three choices: refuse and be transferred or retired, comply, or ask for the instruction in writing from somebody entitled to give it. Chapter Two records that 25,962 government employees were prematurely retired.
The system’s protection against unauthorised instruction is that officials can demand authority. That protection is worth exactly as much as an official’s ability to survive demanding it, and by 1976 the answer was known.
7.4 — The ledger for this part
Part Five committed every subsequent part to carry the running record of which places were affected by its material.
| Where | What was done |
|---|---|
| Delhi | The power cut of 26 June. Turkman Gate and the demolitions of 1976, with residents moved across the Yamuna. The centre of both the detention machinery and the parallel authority. |
| Uttar Pradesh, Haryana, Bihar, Madhya Pradesh, Rajasthan | The sterilisation programme was concentrated in the north. The political consequence arrived in March 1977 in exactly these states, where Congress was almost eliminated. |
| Kerala | The Rajan case. A student died in police custody and his body was never recovered; the pursuit of the case brought down the state government afterwards. |
| The southern states generally | The Emergency was administered less harshly in much of the south, and the electoral verdict of 1977 reflected it — Congress held up in the south while collapsing in the north. That divergence is the clearest single piece of evidence that the 1977 result was a judgment on these months and not a general swing. |
The demolition programme is documented in the Shah Commission’s records, in contemporaneous municipal and Delhi Development Authority papers, and in the accounts of residents collected afterwards.
The finding about instructions emanating from the Prime Minister’s household is the Commission’s own, made after examining officials on oath about where their orders came from.
What it cannot show: the death toll at Turkman Gate. The official figure and the claims made afterwards differ by an order of magnitude, no independent count was made at the time, and the conditions for making one — a free press, an accessible court, an opposition — had all been removed in June 1975. The number is not recoverable and this part does not supply one.
That is the second time in this part that the absence of a number is itself the evidence. The first was the detentions never entered in any register.
The ledger entry above is short and the last row of it does the most work. Where the Emergency was administered lightly, Congress survived the election of 1977. Where it was administered hardest, Congress was destroyed. That correlation is the closest thing this series has to a controlled experiment on the political consequences of these months.
At Turkman Gate in April 1976 police fired on residents resisting demolition. The official toll was around a dozen and the real figure is not recoverable — a state that counted sterilisations individually did not count the people its police killed in its own capital.
The demolitions had a real diagnosis and no procedure: no consultation, no hearing, no compensation and no court, because Chapter Five had closed the last of those two months earlier.
Sanjay Gandhi held no office of any kind and officials across northern India treated his priorities as instructions. The protection against unauthorised orders is an official’s ability to demand authority, and by 1976 the price of doing that was known.
8The Election She Did Not Have to Call
She had extended Parliament’s term, held the opposition in jail and censored the press. In January 1977 she announced a general election, lost it, and left. Explaining that is harder than explaining anything else in this part.
8.1 — What happened
In January 1977 the government announced elections to the Lok Sabha. Detainees began to be released. Censorship eased.
Polling was held between 16 and 20 March 1977. The Emergency was formally lifted on 21 March.
Congress was defeated. The Janata coalition formed a government on 24 March 1977 under Morarji Desai, who had spent much of the previous twenty-one months in detention.
Indira Gandhi lost her own seat at Rae Bareli — the constituency whose election petition had produced the judgment of 12 June 1975. Sanjay Gandhi lost at Amethi.
And she left office.
8.2 — Why she called it
This is the genuinely hard question in this part and there are four serious answers.
She thought she would win. Reports reaching her said so — from the party, from the administration, from the intelligence services. This is the explanation with the most support and section 8.4 is about why it should be believed.
She wanted a mandate. An Emergency ratified by an election ceases to be an emergency and becomes a programme, and that would have resolved the legitimacy problem permanently.
External pressure. India’s standing had suffered and its self-description as the world’s largest democracy had become difficult to sustain.
Conviction. She had been raised inside the independence movement by a man for whom elections were the point of the whole exercise, and there is a reading in which the Emergency was always understood by her as temporary.
These are not exclusive and the honest position is that the mix is not knowable. What can be said is that the decision was hers, that nothing compelled it, and that the alternative — extending Parliament again under a power the Forty-Second Amendment had just supplied — was fully available.
Transfer of power: the handing over of government from a defeated party to its opponents after an election.
It sounds procedural and it is the hardest thing a political system does. Everything else about a democracy is preparation for this one moment, when the people holding the police, the army, the treasury and the broadcasting apparatus give them to people who have promised to investigate them.
The test is not whether an election is held. Authoritarian states hold elections routinely. The test is what happens on the morning after a government loses one.
Why it matters here: on 24 March 1977 that happened in India, three days after the end of twenty-one months in which the losing government had held the winners in prison.
The shape of the defeat is worth setting out, because it is the clearest evidence in this part about what the country thought it was voting on.
Congress had won around 352 seats in 1971. In 1977 it came back with something in the region of 150.
The geography is the striking part. Across the northern states where Chapter Six’s sterilisation drive and Chapter Seven’s demolitions had been hardest — Uttar Pradesh, Bihar, Haryana, Delhi, Punjab — the party was very nearly eliminated, losing every seat or close to it in several of them.
In the southern states, where the Emergency had been administered far less harshly, it held up.
An electorate voting on the economy or on a general mood does not produce a map like that. A map like that is a verdict on specific things done in specific places, and it arrived from people who had spent twenty-one months unable to read about any of it in a newspaper.
8.3 — The part that is genuinely remarkable
Something here deserves stating without qualification, because a prosecution brief that will not say it is not honest.
She held an election she did not have to hold, lost it, and handed over power.
The machinery of state — the police, the administration, the army, the Election Commission — conducted a free poll against the government that controlled it, counted the votes accurately, and transferred office to men who had been in its jails weeks earlier.
Very few countries that go as far as India went in these twenty-one months come back. Most do not have the election. Of those that do, most do not report the result. Of those that do, most do not hand over.
Whatever else is in this part, that happened, and it is the single most important fact about the Indian republic in the twentieth century.
8.4 — Where the commission went
The Shah Commission of Inquiry was appointed in May 1977 under Justice J.C. Shah. It held around a hundred sittings, examined roughly forty-eight thousand papers, and delivered its findings in three parts, the last in August 1978.
Congress returned to power in 1980.
Copies of the report were then recalled, and it ceased to be obtainable in India. It survived because a small number of copies existed in foreign libraries and private hands, and it was reconstructed and republished decades later by a former member of parliament working from those.
Part Two used this as the worked example of its rule: when a state that counts everything has no count of something, somebody decided that. Here the state had produced the count, in three volumes, and then removed it.
Grade: Solid. The report exists, its recall is a matter of record, and the reconstruction is published.
The events of Chapter Eight are as well documented as anything in this series: the announcement, the polling dates, the results constituency by constituency, the revocation, and the swearing-in.
The Election Commission’s returns are the strongest single source in this part, and for a reason worth noting. They were produced by an institution of the same state, during the same weeks, and they record the defeat of the government that appointed its officials. An electoral machinery that reports its own government’s destruction accurately is telling you something about itself that no commission of inquiry could establish afterwards.
What none of it can show: why she called it. Section 8.2 gives four explanations and declines to choose. The intelligence assessments that would settle it are not released, and the accounts of participants were written afterwards by people with reputations to arrange.
What can be examined, though, is not her reasoning but the machinery that produced the information she reasoned from.
Everybody explaining March 1977 assumes that the Emergency ended because she called an election.
Her defenders treat the calling as proof of democratic conviction. Her critics treat it as a miscalculation that democracy was lucky to survive. Both accept that the decision was the cause of the ending, and argue about what the decision reveals about her.
Ask instead what made the decision look safe.
She called it because she was told she would win. And she was told she would win because every institution capable of telling her otherwise had been dismantled by the machinery of this part. The press could not report discontent, because Chapter Three. The opposition could not organise or demonstrate it, because Chapter Two. Officials who had objected had been retired, because Chapter Seven. Party workers whose careers ran through one office had no incentive to carry bad news upward, because Part Eight’s Chapter Two. Courts could not surface grievances, because Chapter Five.
Every one of those was a channel by which a government finds out what is happening in the country. All of them were closed, deliberately, and every closure was defensible in its own terms at the time.
So the ending is not a story about conviction or about miscalculation. It is that the apparatus built to prevent the country from telling her anything worked, and the last person it blinded was her.
This is the most important finding in this part and it generalises beyond India and beyond 1975. A government that suppresses information does not acquire a country that is quiet. It acquires a country it can no longer see — and it loses the one instrument every government needs most, which is the ability to know when it is wrong.
Part Six found a warning delivered and not received. Part Four found a diagnosis on the file and no action. This is the same failure at maximum extension, and the difference is that here the government had built the blindness itself, on purpose, and then walked into it.
One last observation about the commission. It was appointed in May 1977 and delivered its final report in August 1978 — about fifteen months to investigate twenty-one. That is fast, and it is fast because the successor government wanted it fast. Chapter Ten is about what happened to the result.
Elections were announced in January 1977, held 16–20 March, and the Emergency was lifted on 21 March. Congress lost. She lost Rae Bareli. Janata took office on 24 March.
She held an election she did not have to hold, lost it, and handed over power. Very few states that go this far come back, and this is the single most important fact about the Indian republic in the twentieth century.
She called it because she was told she would win — and every channel that could have told her otherwise had been closed, deliberately, one at a time. The last person the blindness blinded was her.
9What Would Have Prevented It
Some of it was repaired within three years and the repairs were good. This chapter is about what was fixed, what was not, and what is still on the shelf.
9.1 — Four arguments for the defence
These are the strongest arguments available. Two of them are conceded in full.
One. The provocation of 25 June was real. A national leader publicly urging the police and the armed forces to decide which orders to obey is not ordinary opposition. Part Eight said so and this part repeats it.
Two. Everything was constitutional. Not a defence, as the front matter argued, but it is a true statement and it means the charge falls partly on the drafters of 1950 and not only on the government of 1975.
Three. It ended, and she ended it. Chapter Eight. Conceded without qualification.
Four. The system corrected itself and the correction has held for forty-nine years. This is the strong one. Within three years the worst provisions were repealed. Within five the Supreme Court had struck down the rest. No Indian government since — of any party, across many changes of power — has proclaimed a national emergency on internal grounds. Whatever the Emergency proved about the fragility of Indian institutions, the half-century since has proved something about their recovery, and no honest account can print the first without the second.
9.2 — What the Forty-Fourth Amendment fixed
The repairs of 1978 were serious and specific, and they deserve listing because they are the most useful constitutional work done in India since 1950.
| The hole | What was done about it |
|---|---|
| “Internal disturbance” as a ground for emergency — undefined, no threshold | Replaced with armed rebellion, which is a far higher and more checkable bar. |
| A proclamation on the advice of one person | The President may act only on the written advice of the Cabinet as a body. |
| A proclamation that ran until revoked | Must be approved by both Houses within a month, by a special majority, and renewed every six months. The Lok Sabha may revoke it by resolution. |
| Articles 20 and 21 suspendable | They can no longer be suspended, even during an emergency. This is the direct answer to Chapter Five, and it means ADM Jabalpur could not be decided the same way today. |
| Amendments beyond judicial review | Article 368(4) and (5) repealed; struck down in any case by Minerva Mills in 1980. |
| Six-year legislative terms | Returned to five. |
Look at the fourth row in particular. The specific legal proposition of Chapter Five — that the right to life cannot be enforced during an emergency — was made impossible by constitutional text within two years of the judgment. That is a functioning system correcting a catastrophic error, and it should be said plainly.
9.3 — What is still on the shelf
And now the other half, which is why this chapter is not a happy ending.
Preventive detention was not touched. Article 22 stands as it was written in 1950. MISA was repealed in 1978 and the National Security Act was passed in 1980 — by the Congress government that returned to office that year. Successor statutes have existed continuously since, under a series of names, and governments of every party have used them.
Part Two’s charge was never about MISA. It was about the constitutional authorisation, and the constitutional authorisation is unchanged.
Article 356 remains. Its use was constrained by the Supreme Court in 1994, which is real and valuable, but the provision is where it was.
The Ninth Schedule remains. Entries made after April 1973 became reviewable in 2007, which leaves the earlier ones where they were.
So the honest accounting is this. The instruments used to seize the moment were repaired. The instruments used to hold the country for twenty-one months were not.
9.4 — What I would have to retract
Three things would break this part.
If the cabinet record of 25 and 26 June 1975 showed that a genuine, specific and imminent threat was assessed that night — something beyond the state of the agitation everybody could see — the front matter’s argument box tips and Chapter One with it.
If district records showed that the sterilisation increase of 1976–77 tracked something other than quota enforcement, Chapter Six’s central inference fails.
If it were shown that the intelligence reaching her in January 1977 accurately predicted defeat, Chapter Eight’s hidden assumption collapses and the calling of the election becomes an act of conviction rather than an artefact of blindness.
The first and third depend on files the Indian government holds. That is now the fifth part in a row ending on that sentence, and Chapter Ten has something to say about the pattern.
Which produces the question every Indian discussion of this period ends on.
Every Indian discussion of the Emergency arrives here, and both answers are held sincerely by serious people.
The Forty-Fourth Amendment shut the door that was used. “Armed rebellion” is not a phrase a government can apply to a protest movement. Cabinet advice must be written and collective. Parliament must approve within a month and renew every six. Articles 20 and 21 cannot be suspended, which removes the legal basis of every detention in Chapter Two. And a state with a hundred television channels, an internet and a Supreme Court that has overruled ADM Jabalpur is not the state of 1975.
The 1975 route is closed and nothing else is. Preventive detention remains constitutionally authorised and statutorily available, and is used continuously without any emergency at all. Article 356 remains. The financial instruments Chapter Three described — advertising, licences, allocations — never required an emergency and never will. The lesson a competent government takes from 1975 is not that these things are impossible but that declaring them is the mistake, because the declaration is what created a date, a name, and something to repeal.
What would settle it: nothing settles a question about the future. What can be tracked is the use of preventive detention year by year, which is published and which nobody compiles.
Why people care so much: because every Indian government since 1977 has been accused of an undeclared emergency by its opponents, and the accusation is unfalsifiable precisely because no declaration is required.
That is the case against this part, and two of the four defences I think simply win. What none of them reaches is section 9.3, because the accounting there is not about 1975 at all — it is about which powers a country decided to keep after it had seen exactly what they could do.
Four defences: the provocation of 25 June was real; everything was constitutional; she ended it; and the system corrected itself and the correction has held for forty-nine years. The third and fourth are conceded in full.
The 44th Amendment replaced “internal disturbance” with armed rebellion, required written Cabinet advice and parliamentary renewal, and made Articles 20 and 21 unsuspendable — which means ADM Jabalpur could not be decided the same way today.
The instruments used to seize the moment were repaired. The instruments used to hold the country for twenty-one months were not. Article 22 stands as written in 1950.
10An Honest List of What We Do Not Know
The central document of this part has never been officially published. That is not an accident of archiving and it is the fifth time this series has ended a part on the same sentence.
10.1 — Genuinely unknown
Six things in this part cannot be established, and the shape of the list is unusual: most of them are counts that a state which was counting everything declined to keep.
How many people were actually detained. The Shah Commission’s figure of about 110,806 is the best available and it is a floor.
Why it is unknown: a person picked up, held for days and released without an order entering any register does not appear in it. There is no way to estimate that category, and it was the category most likely to include the poorest detainees.
How many died in custody. At least twenty-two are recorded.
Why it is unknown: the mechanisms that produce such a count — inquests reported in a free press, habeas corpus proceedings, an opposition asking questions in Parliament — had all been removed by the events of Chapters One to Five.
How many died at Turkman Gate. The official figure and the claims differ by an order of magnitude.
Why it is unknown: the same reason, in one afternoon in one city. Chapter Seven treats the absence as the finding rather than as a gap.
What share of the sterilisations were coerced. Chapter Six establishes the mechanism and the tripling. It does not establish a proportion.
Why it is unknown: the consent forms were all signed. A programme that produces documentation of consent for coerced procedures cannot afterwards be audited on consent, which is why Chapter Six argues from the annual figures instead.
What was decided on the night of 25 June, and why. Cabinet papers closed.
What she was actually told in January 1977. Chapter Eight’s central claim is that she called an election because the reports reaching her said she would win.
Why it is unknown: intelligence assessments of the period are not released. The claim rests on the accounts of participants and on the shape of the outcome, and Chapter Eight’s argument is structural rather than documentary — it says the channels were closed, which is established, and infers that what came through them was wrong.
10.2 — Solid
The chronology. The proclamation of 25 June, the orders of 27 June, the amendment of 1976, the judgment of 28 April 1976, the announcement of January 1977, the poll of 16 to 20 March, the revocation of 21 March, and the change of government on 24 March.
The Shah Commission’s figures. About 110,806 detained; 25,962 government employees prematurely retired; 26.42 lakh sterilisations in 1975–76 and 81.32 lakh in 1976–77; 1,774 deaths recorded from those procedures; 548 complaints concerning unmarried people.
Everything in the Forty-Second Amendment. It is a statute and can be read.
ADM Jabalpur. The bench, the four-to-one split, the holding, Khanna’s dissent, Beg’s appointment on 28 January 1977 and Khanna’s resignation the same evening. All reported or on the record.
The repairs. The Forty-Third and Forty-Fourth Amendments and Minerva Mills. All published.
The recall of the Shah Commission report after 1980, and its later reconstruction from surviving copies.
10.3 — The fifth part in a row
Something has now happened often enough to be reported as a finding in its own right.
Part Five ended by naming a calculation nobody has done. Part Six ended by naming a report the government has held closed for sixty-three years. Part Seven ended by naming two counts anybody could make and nobody has. Part Eight ended by naming three closed files. This part ends by naming the same thing again.
The Shah Commission is the most extreme case in the series, because here the state did not merely fail to investigate. It investigated thoroughly, published in three volumes, and then withdrew the volumes.
Set that beside Part Six. The Henderson Brooks report on the defeat of 1962 was never published and remains closed after sixty-three years. The Sunderlal report on Hyderabad was not public for sixty-four years. The Shah Commission report was published and then recalled.
Three national catastrophes, three official investigations, and in each case the finding of the investigation is harder to obtain in India than in a foreign library.
I do not think this is a conspiracy, for the reason Part Six’s Chapter Eight gave: it does not need to be. Withholding requires nobody to decide anything and releasing requires somebody to sign. But the effect is the same however it is produced, and after five parts the effect is the pattern.
Genuinely unknown: how many were detained, how many died in custody, how many died at Turkman Gate, what share of the sterilisations were coerced, what was decided on 25 June, and what she was told in January 1977.
Solid: the chronology, the Shah Commission’s figures, the whole text of the 42nd Amendment, ADM Jabalpur, the repairs of 1978 and 1980, and the recall of the report after 1980.
Three national catastrophes, three official investigations, and in each case the finding is easier to obtain abroad than in India. That is now the pattern rather than the exception.
Sources & further reading — Part 9
Timeline, June 1975 to March 1977
Dates only. The arguments are in the chapters.
| Date | What happened |
|---|---|
| 12 Jun 1975 | The Allahabad judgment. Congress loses Gujarat the same day. Part Eight’s Chapter Eight. |
| 24 Jun 1975 | The Supreme Court grants a conditional stay allowing her to remain Prime Minister. |
| 25 Jun 1975 | A rally in Delhi calls on the police and armed forces not to obey illegal orders. That night the President signs a proclamation of emergency under Article 352 on the ground of internal disturbance. The Cabinet is informed afterwards. |
| 2 a.m., 26 Jun | Power to the Delhi newspaper presses is cut, on an instruction from the Lieutenant Governor. Around 676 arrests are made before dawn. |
| 26 Jun 1975 | Pre-publication censorship imposed. Most Delhi papers do not appear that morning. |
| 27 Jun 1975 | Article 358 suspends the protections of Article 19. Article 359 suspends the right to move any court for Articles 14, 21 and 22. |
| 1975–76 | About 26.42 lakh sterilisations performed. |
| Aug 1975 | The 39th Amendment places the Prime Minister’s election beyond ordinary judicial challenge and puts MISA in the Ninth Schedule. |
| 7 Nov 1975 | A five-judge bench strikes down the election provisions of the 39th Amendment as violating the basic structure, while upholding her election. The doctrine holds under the Chief Justice appointed to break it. |
| Apr 1976 | Demolitions reach Turkman Gate. Police fire on residents. Survivors are moved across the Yamuna. |
| 28 Apr 1976 | ADM Jabalpur. Four to one: no detained person may approach any court, even where the detention is illegal. Justice H.R. Khanna dissents. |
| 30 Apr 1976 | The New York Times publishes an editorial on the dissent. |
| 1976 | The Forty-Second Amendment: forty articles altered, fourteen added, two new Parts, the Preamble rewritten, terms extended to six years, and amendments placed beyond judicial review. |
| 1976–77 | About 81.32 lakh sterilisations — roughly three times the previous year. 1,774 deaths are recorded from the procedures. |
| Jan 1977 | Elections announced. Detainees begin to be released and censorship eases. |
| 28 Jan 1977 | M.H. Beg is appointed Chief Justice over Justice Khanna, who hears it on the radio and resigns the same evening. |
| 16–20 Mar 1977 | Polling. Congress is defeated. She loses Rae Bareli; Sanjay Gandhi loses Amethi. |
| 21 Mar 1977 | The Emergency is revoked. |
| 24 Mar 1977 | A Janata government takes office under Morarji Desai, who had spent much of the period in detention. |
| May 1977 | The Shah Commission of Inquiry is appointed under Justice J.C. Shah. |
| 1977–78 | The 43rd and 44th Amendments repeal most of the 42nd. “Internal disturbance” becomes “armed rebellion”. Articles 20 and 21 are made unsuspendable. |
| Aug 1978 | The Shah Commission delivers the last of its three reports. |
| 1980 | Congress returns to power. Copies of the report are recalled. The National Security Act is passed. |
| 1980 | Minerva Mills strikes down the expansion of Article 31C and clauses (4) and (5) of Article 368, using the basic structure doctrine those clauses had been written to abolish. |
| 2017 | ADM Jabalpur is overruled in K.S. Puttaswamy, by a bench including the son of one of the judges who decided it. |
Part Ten starts here. It goes to Punjab, from 1978 to 1993 — a state Part Five’s ledger already records as both the largest beneficiary and the largest casualty of central policy.
Glossary
Every hard word used in this part, in plain English.
| Word | What it means |
|---|---|
| Article 352 | The power to proclaim a national emergency. In 1975 the grounds included “internal disturbance”, an undefined phrase; since 1978 the ground is “armed rebellion”. |
| Article 358 | Suspends the protections of Article 19 — speech, expression, assembly, movement — automatically during an emergency. |
| Article 359 | Allows suspension of the right to move a court for enforcement of specified fundamental rights. Since 1978 it cannot reach Articles 20 and 21. |
| Basic structure doctrine | The rule from Kesavananda Bharati (1973) that Parliament may amend the constitution but may not alter its essential features. The 42nd Amendment tried to abolish it and was struck down by it. |
| Habeas corpus | An order requiring whoever holds a person to produce them before a court and justify the detention. The protection every other right depends on. |
| Locus standi | The right to bring a case before a court at all — standing. ADM Jabalpur held that detainees had none during an emergency. |
| MISA | The Maintenance of Internal Security Act 1971. Detention without charge or trial on the executive’s satisfaction. Repealed 1978; the National Security Act followed in 1980. |
| Pre-publication censorship | A requirement to show material to an official before printing, as opposed to punishment afterwards. Nothing suppressed this way leaves any trace. |
| Preventive detention | Holding a person to stop them doing something, rather than for something they have done. Authorised by Article 22 of the constitution since 1950 and never removed. |
| Target-driven administration | Setting officials numerical goals and rewarding them on the number. Its known failure is that where the target is measurable and the quality is not, the number wins. |
What Comes Next
Part Ten — Punjab, 1978 to 1993
Part Ten takes fifteen years in one state. Part Five’s ledger recorded Punjab as both the largest beneficiary of central policy and, later, one of its largest casualties. It asks:
- Part Five found Punjab’s river waters allocated by notification in 1976 and by an agreement in 1981 that ended a live Supreme Court case. What did that do to the politics of the state in the years immediately after?
- The Anandpur Sahib Resolution is described in Indian politics as a secessionist document. What does it actually say, and when did the description change?
- A figure was built up in Punjab politics in the late 1970s. Who built him, why, and what is the evidence for that beyond the assertions of people who later had reason to make it?
- Operation Blue Star was conducted in June 1984. What were the alternatives on the table, who proposed them, and why were they rejected?
- Part Two’s rule about counting applies to November 1984. How many people were killed in Delhi, how do we know, and how many convictions followed in the four decades since?
- Between 1984 and 1993 the state was governed under emergency provisions for most of the period. What was done under them, and what has been established since?
Questions, not topics. If any of them is answered badly, you will be able to tell.