A Part Made of Sentences
This is an advocacy document. It is the case for the prosecution against Congress governments in power. The selection of material is one-sided on purpose. The evidence in it is not. Every serious claim carries a grade showing how well it is proved, the other side is given its strongest argument, and where the other side wins, this book says so. A second series applying the same test to the government of 2014 onwards is planned.
Sixteen parts of this series have argued about what documents establish. This one is about words that were spoken out loud, usually in front of a crowd, often on camera.
That sounds easier and it is not. A document sits still. A sentence gets translated, shortened, moved to a different date, put in a different mouth, and quoted for forty years by people who have never seen the original. By the time a political statement is famous it has usually stopped being a statement at all and become a slogan about somebody.
So this part has a rule, and it is the one the whole series has been built on since Part One.
Nothing appears here without a date, a venue, and the operative wording.
If this book cannot say when a thing was said, where it was said, and what the words were, the statement does not go in — however useful it would be.
Where a statement was made in Hindi and circulates in English, the translation is treated as evidence about the translator and not about the speaker, and the competing versions are given.
Where a statement is famous and did not happen, it goes in Chapter Two, which is the first chapter of substance in this part and the most important one.
That rule is easy to state and it has a consequence worth naming before anybody turns a page.
Everybody assumes that a collection of damaging statements is a collection of evidence. It looks like the easiest kind of proof there is: he said it, here it is, in his own voice.
But a statement proves only that a sentence was spoken. It does not prove what the speaker believed, what the government then did, or whether anybody was harmed. Sixteen parts of this series have been about acts — a demolition, an allocation, an arrest, a file that was not released. This part is about none of those.
Which makes it the weakest chapter of the prosecution and the one most likely to be quoted. That combination is exactly what makes it dangerous, because a series that has graded its evidence for sixteen parts can lose all of it in the seventeenth by treating a sentence as though it were an act.
The way out is not to leave the statements out. It is to say, for each one, what it is evidence of — and in most cases the honest answer is: of what a government thought it could say in public and get away with. That is a real thing to know about a government. It is a smaller thing than a crime, and this part will not inflate it.
1How a Statement Is Verified
A political quotation in India usually arrives with none of the three things that would let you check it. It has a name attached and nothing else — no date, no place, no context, and frequently not the language it was spoken in.
This chapter sets out what this book required before a sentence went on the page, because the rest of the part is unreadable without it.
1.1 — The three tests
Date. Not a year. A day, or at worst a session. A statement whose date cannot be fixed usually turns out to be a statement whose existence cannot be fixed either.
Venue. Where, and to whom. The same words in Parliament, at a rally and to a reporter are three different acts. Parliament is recorded verbatim. A rally is recorded by whoever was filming. A remark to a reporter exists only in that reporter’s copy.
Wording. The operative sentence, in the language it was spoken in where that is known.
Most famous political statements are one sentence long, taken out of a speech of several thousand words.
This book calls that sentence the operative one: the part that does the damage and gets repeated. Everything around it is context, and context can be summarised.
The reason for the distinction is practical. A speech is somebody’s work and reproducing it whole is neither necessary nor right. The operative sentence is what is actually in dispute, and it is short.
The best record of any of them is the one Parliament keeps, and even that has a known hole in it.
Parliament keeps a verbatim record — a word-for-word transcript of everything said in the chamber, published afterwards. Nothing else in Indian public life is recorded that way. A rally has whoever was filming; a press conference has whoever was there.
The Chair can order words expunged — struck from the transcript. Those words were still said, and people in the room heard them. They are simply not in the record.
So the best source in this part has a known gap in it, and the gap is in exactly the direction that matters: the most offensive things said in Parliament are the ones most likely to be missing from the record of Parliament.
1.2 — What counts as said
Four levels, in descending order of how much weight this book will put on them.
| Source | What it establishes | Weakness |
|---|---|---|
| Video or audio of the speaker | That the words were spoken, in that order, in that tone | Can be clipped. Requires the surrounding minutes to be checked too |
| Parliamentary record | Verbatim wording, date, House, and who else spoke | Members may have words expunged; the record is of the House, not of a rally |
| An official release from the speaker’s own office | What the speaker’s office says the speaker said | Written afterwards, sometimes in response to a controversy |
| Contemporary newspaper report | That a journalist present recorded these words that day | Paraphrase and translation are routine and usually invisible |
Anything below the fourth row — a quotation circulating without any of these, however often repeated — is not used in this part except as a subject of Chapter Two.
1.3 — The translation problem
Most of the statements in this part were made in Hindi and are famous in English. That is not a technicality. It is the single largest source of dispute about what Congress leaders have said.
Take the most famous sentence in this part, which Chapter Four is about. In Hindi it is nine words. It circulates in at least four English versions, and they are not the same sentence.
One version has the earth shaking. Another has it shaking a little. A third has it only natural that the earth shakes. A fourth drops the qualifier entirely.
The strongest of those four, the one that does the most damage to the speaker, appears on the Congress party’s own website. Chapter Four sets out why, and it is not what anybody would guess.
This book’s rule is therefore: give the original where it is known, give the competing English versions, and treat a translation as evidence about whoever made it.
There is a serious objection to that rule and it should be put before the rule is applied to anybody.
The objection to this chapter’s rule: that it is too strict, and that strictness here is a kind of dishonesty.
Most of what a government says is said at rallies, in interviews and at press conferences, none of which is systematically recorded. A rule that admits only what can be pinned to a day and a room therefore admits mainly what was said in Parliament — which is the most guarded speech any politician produces, and the least revealing.
So the rule filters for caution. The unguarded remark, which is the one that tells you what somebody actually thinks, is precisely the one least likely to survive the three tests. A book applying this rule will systematically under-report how politicians talk, and will do so while congratulating itself on rigour.
That objection is correct and this book accepts it. Chapter Ten sets out what the rule cost: roughly four statements excluded for every one admitted, and one of the excluded ones is named.
Why the rule stays anyway: because the alternative is not a fuller picture, it is an unfalsifiable one. A statement that cannot be checked cannot be defended against either, and a document built from such statements cannot be corrected by anybody — including by its author, who has no way of finding out that he is wrong. Under-reporting is a cost. Being uncorrectable is a different kind of thing altogether.
1.4 — What a statement is evidence of
The front matter said this and it is worth putting once more in plain terms, because it governs every verdict in this part.
A statement establishes that a sentence was spoken. It does not establish intent, and it does not establish that anything followed. Where something did follow, this part says so and cites the part of this series that covers it. Where nothing followed, it says that too.
Of the eleven statements set out in this part, two were followed by a change in what the government did.
One of them, in Chapter Six, changed things in the opposite direction from what its speaker claimed. The other, in Chapter Two, killed a Cabinet decision within days and is the only statement here that achieved what the speaker intended.
The rest changed nothing at all. They were sentences, they caused offence, they were quoted for years, and the machinery described in the previous sixteen parts carried on exactly as before.
That is the honest measure of this part’s weight, and it is given at the front rather than buried at the back.
Which is why the chapters that follow are ordered as they are, with the fabrications before the facts.
Nothing in this part appears without a date, a venue and the operative wording. If those three cannot be established, the statement is not used — however useful it would be.
Most of these sentences were spoken in Hindi and are famous in English. A translation is evidence about the translator, and this part gives the competing versions rather than choosing one.
And a statement proves that words were spoken. It does not prove intent, and it does not prove that anything followed. Usually nothing did.
2What Was Never Said
This chapter goes first, before any real statement, and the reason is the argument Part One made and this series has followed for sixteen parts.
Every fabricated quotation in circulation is a defence. It lets a documented statement be waved away as more of the same, by a person who has just watched a false one be debunked and reasonably concluded that the whole genre is unreliable. Killing the fakes is what makes the real ones impossible to dismiss. Nobody else writes this chapter, in either direction.
2.1 — The one about first claim on resources
This is the most quoted political sentence in modern India and it is the clearest case in this book of a real statement turned into a different one.
What happened. On 9 December 2006, at the fifty-second meeting of the National Development Council in New Delhi, the Prime Minister, Manmohan Singh, set out the government’s spending priorities. He listed them: agriculture, irrigation and water, health, education, rural infrastructure, general infrastructure, and programmes for Scheduled Castes and Tribes, other backward classes, minorities, and women and children. He then said that plans would be needed so that minorities, particularly the Muslim minority, could share equitably in development, and followed it with a nine-word sentence.
“They must have the first claim on resources.”
Nine words. Spoken in English. Not disputed by anybody, including the speaker’s own office.
The entire argument of the last twenty years is about the word “they”.
What was made of it. Within a day the sentence was reported as a statement that Muslims have the first claim on the country’s resources. It has been repeated in that form in election campaigns ever since, including at the highest level.
What the record shows. On 10 December 2006 the Prime Minister’s Office issued a clarification calling the reading a deliberate and mischievous misinterpretation, and reproduced the surrounding paragraph. Its position was that “they” referred back to the whole list of priority groups — Scheduled Castes and Tribes, backward classes, minorities, women and children — and not to Muslims alone.
The case that the popular reading is fair: the sentence immediately before names the Muslim minority specifically. In ordinary English, “they” refers to the nearest preceding subject. A Prime Minister who says a group must have first claim on resources, one sentence after naming that group, cannot complain when people take him to have meant it.
And this is not a hostile invention. It was how several major newspapers and news agencies reported it on the day, before any political fight had started.
The reply: the paragraph is a list, and the list is in the speech before the sentence about Muslims. The clarification came the next day, not years later under pressure. And an approach paper for the Five Year Plan set out the same priorities in the same order.
Where this leaves it — and this book will not pretend it is comfortable: the sentence was badly constructed and the ambiguity is genuine. What is not genuine is the version now in circulation, in which a Prime Minister announced that Muslims come first. Fabricated as to that meaning; accurate as to the words.
This book’s verdict is that the sentence is real, the popular reading is not supportable on the full paragraph, and the drafting was careless enough that the confusion was foreseeable.
When a statement causes trouble, the speaker’s office issues a written note saying what was meant. In India this is normally a press release from the ministry or the Prime Minister’s Office.
A clarification is evidence of two things and not of a third. It establishes what the office says the speaker meant, and it establishes the date on which the office decided that mattered. It does not establish what the speaker meant when speaking, because it is written afterwards by people whose job is to reduce the damage.
The clarification of 10 December 2006 is unusually strong for the genre, because it did not paraphrase. It reproduced the paragraph and let readers see the list for themselves.
2.2 — The one where the words are real and the act is not
The second worked example is better than the first, because here the words are established beyond any argument and it is the action that was invented.
What happened. On 24 September 2013 the Cabinet approved an ordinance to reverse the Supreme Court’s judgment of that July, which had disqualified convicted legislators immediately. Part Sixteen sets out the constitutional background.
On 27 September 2013, at the Press Club of India in New Delhi, the Congress general secretary Ajay Maken was holding a “Meet the Press” session. The party’s vice-president, Rahul Gandhi, arrived unannounced, took the microphone, and said what he thought of the ordinance.
“It is complete nonsense, it should be torn up and thrown away. It is my personal opinion.”
And, in the same remarks: that what the government had done was wrong, and that if corruption was to be fought — by his party or any other — these small compromises could not continue.
Established to the highest standard available outside Parliament: on camera, at a named venue, on a fixed date, in English, and posted by the party itself the same day.
What was made of it. That he tore up the ordinance.
He did not. He said it should be torn up. A photograph of him tearing a piece of paper, which circulates attached to this episode, is from a rally in 2012 and has nothing to do with it.
The fabrication here is unusually well documented because of who repeated it.
Major news organisations reported the tearing as fact; at least one has since published a correction saying he did not tear the ordinance but called for it to be torn up. The footage posted by the Congress on 27 September 2013 shows no tearing.
And in February 2023, defending the episode, the party’s own chief spokesman referred to it as Rahul Gandhi’s act of tearing a copy of the ordinance — describing something that had not happened, on his own leader’s behalf, ten years later.
This is Chapter Two’s second pattern operating on an action rather than a sentence, and it demonstrates the thing this chapter exists to say: a fabrication does not need a hostile author. It needs only a vivid image and nobody checking.
2.3 — And it is the one statement here that worked
Chapter One said that of the statements in this part, one changed what the government did. That was the press conference of January 2011, and it changed things against the speaker.
This is the second, and it is the only one in the part that achieved what the speaker intended.
The ordinance had been approved by the Cabinet. The Prime Minister was in the United States. The ordinance was withdrawn.
The judgment it would have reversed still stands, and legislators of every party have lost their seats under it since — including, in 2023, the man who made this statement.
Part Sixteen recorded that the Prime Minister, on that trip, asked a colleague whether he should resign. A statement by a man holding no government office killed a Cabinet decision and nearly took the Prime Minister with it.
Whether that is to anybody’s credit is harder than it looks, and the case against is strong.
The case that this is not to anybody’s credit: the ordinance was killed by the right argument delivered in the worst possible way — at a press conference, by a person outside the government, against a Cabinet decision, while the head of that Cabinet was abroad and could not answer.
Part Fifteen’s opening box asked what happens when the office of Prime Minister is separated from the authority that goes with it. Part Sixteen called this episode the clearest illustration in nineteen parts, and it stands by that.
The reply, which this book also stands by: the ordinance was indefensible, the Supreme Court has been vindicated, and somebody stopped it. The method was constitutionally ugly and the outcome was right, and a prosecution document does not get to keep only the half it prefers.
What it cannot be called is disrespect for the Prime Minister and nothing more. It was a public repudiation of a decision the Prime Minister’s Cabinet had taken, and it worked.
2.4 — The general problem
The 2006 sentence is unusual only because the original text survives and can be checked. Most circulating quotations cannot be.
Try it yourself with any political quotation you are confident about. Ask three questions: what day, what building, and what were the words either side of it.
For most of the quotations that circulate about Indian politicians of any party, you will not be able to answer any of the three, and neither will the person who sent it to you.
That is not a fact about Congress. It is a fact about how political speech travels, and it works identically in every direction.
Three patterns account for almost all of it, and they are worth naming because once you can name them you stop being fooled by them.
| Pattern | What is done | How to catch it |
|---|---|---|
| The clipped referent | A pronoun is separated from what it referred to, as in 2.1 | Find the paragraph, not the sentence |
| The travelling quote | A real sentence by one person is attributed to a more famous one | Ask for a date. It is almost never available |
| The improved translation | A Hindi sentence is rendered in English more strongly than it was said | Find the original audio. Chapter Four |
All three are used against politicians of every party. This book has been sent examples of each about the leaders it is defending in this chapter and about the ones it prosecutes in the next four.
Anybody who circulates a damaging quotation assumes that if it turns out to be false, the worst outcome is that one claim fails — a single bad item among many looks like an acceptable rate of error. It is not, and the reason is arithmetic rather than morality.
A reader who checks one quotation and finds it fabricated does not discard that quotation. He discards the source, and with it every accurate statement the source ever made. One false quotation does not cost you one item. It costs you the list — and the loss is retrospective, because pages already read stop counting.
Part One made this argument about charges and Part Sixteen made it about audit figures. It is sharpest here, because a statement is the cheapest thing in the world to fabricate and the easiest to check. Which means anybody who fabricates one has chosen to be caught, and anybody who repeats one without looking has chosen not to.
Which is the whole reason this chapter comes before the ones containing the real material.
The most quoted sentence in modern Indian politics is nine words long, was said on 9 December 2006 at the National Development Council, and turns entirely on the word “they”.
The words are real. The version in circulation — that a Prime Minister said Muslims come first — is not supportable on the full paragraph, and a clarification was issued the next day.
One fabricated quotation does not cost you one item. It costs you the list. That is why this chapter comes before the real ones.
3What Was Said Well
Parts Fourteen, Fifteen and Sixteen all placed the credit chapter early, before the charge, and said why: at the end it reads as a sweetener produced late. The same applies here. Three statements, all real, all verifiable, all to the credit of the person who made them.
3.1 — The apology
On 11 August 2005, in the Rajya Sabha, during a debate on the report of the Nanavati Commission into the killings of November 1984, the Prime Minister, Manmohan Singh, apologised.
He said he had no hesitation in apologising to the Sikh community and to the whole nation, and that what had happened in 1984 was a negation of the concept of nationhood set out in the Constitution.
Verbatim on the parliamentary record. Date, venue and wording all establishable to the highest standard available in this part.
It is the only occasion in this series on which a Prime Minister apologised, in Parliament, for something his own party had done.
Part Eleven set out what followed it, which was: the compensation package of January 2006, and not much else. An apology is not a remedy and this book has said so. It is also not nothing, and Chapter Four is what the alternative sounds like.
Where it was said matters as much as what was said, and the setting needs a word of explanation.
A commission of inquiry is appointed by government to investigate a matter. Part Fourteen described one that took seventeen years, and Part Eleven counted four of them on the events of November 1984 alone.
When such a commission reports, the government must lay the report before Parliament with a memorandum saying what it intends to do. That triggers a debate on the report — one of the few occasions on which a House is obliged to discuss something a government would usually rather not.
The apology of 11 August 2005 was made in that setting: not at a press conference, not on an anniversary, but in the upper house during the debate on the Nanavati Commission’s report, where the Prime Minister had to speak and was going to be answered.
What it was worth is a separate question and this book has already answered it elsewhere.
The apology came in August 2005 — twenty years and nine months after the killings.
What followed it: a compensation package in January 2006, of which around ₹200 crore was never distributed because of disputes over entitlement. Part Eleven has the detail.
What did not follow it: the communal violence law that would have imposed duties on officials, promised in 2004, drafted in 2005 and 2011, and never passed in ten years of office. Parts Fifteen and Sixteen.
So the apology is real, it is on the record, and it is the only one in this series — and Part Eleven’s finding stands unchanged around it: establishing a record has functioned as a substitute for accountability rather than a step towards it.
An apology is a record. It is the purest example of one.
3.2 — The letters
Part Six of this series was severe about Jawaharlal Nehru’s China policy and stands by all of it. This section is about what he said when the policy had already failed.
After the fighting of 1962, in correspondence with the Chinese premier Zhou Enlai, Nehru rejected the proposal that both sides pull back twenty kilometres from a line China had drawn. He asked what that line was, and observed that it was a line created by aggression since the beginning of September. He said that the demand that India accept the Chinese line of 1959 was one to which India would never submit, whatever the consequences and however long the struggle.
These are letters, not speeches. They exist as documents, they were exchanged between two governments, and both sides have published them.
That makes them the strongest kind of source in this part — stronger than video, because there is no question of clipping, and stronger than a newspaper report, because there is no translator between the reader and the words.
Part Six’s charge was that the Forward Policy was adopted against the professional advice of the officers who would have to execute it, and that the warnings about the Aksai Chin road were ignored for years. None of that is affected by the letters, and the letters are not affected by it. Both go in.
3.3 — The best sentence in this part
On 5 December 1961 Nehru told Parliament that the disputed ground in Aksai Chin was worthless — a sentence Chapter Seven examines at length.
A Congress member of that Parliament, Mahavir Tyagi, pointed at his own bald head and asked whether it followed that his head had no value.
Consider what that took. Tyagi was a Congress MP. The man he was contradicting was the Prime Minister, the leader of his party, and by some distance the most powerful individual in the country.
He did it in the House, in public, on the record, to his face — and he did it with a joke, so that the House laughed and the Prime Minister laughed with it, and the point landed anyway.
It is the most effective single sentence spoken by anybody in this part, and it was spoken by a Congressman against his own leader.
Part Three found that the restraint on power through the 1950s was a handful of individuals with independent standing, and that all of them were gone by 1954. Tyagi is a late survivor of that species, and there is no equivalent anywhere in Parts Fourteen to Sixteen.
One qualification, because this book’s own rule requires it. Accounts of that exchange differ on whether Tyagi was answering Nehru directly or adding to an opposition member’s point, and on which House it happened in. Chapter Ten records the discrepancy. What is not disputed is that the words were his, that they were said in Parliament, and that they were said to Nehru’s face.
3.4 — The self-assessment
One more, and it belongs in this chapter although it is not obviously to anybody’s credit until you look at what it took to say.
Date: January 2014. Venue: a press conference in New Delhi — one of only a handful the Prime Minister gave in ten years, and among his last. Speaker: Manmohan Singh.
Asked repeatedly about the charge that he had been a weak and indecisive Prime Minister, he answered it directly.
“I do not believe that I have been a weak Prime Minister.”
“I honestly believe that history will be kinder to me than the contemporary media, or for that matter the Opposition in Parliament.”
And, on the same occasion: that given the political compulsions, he had done the best he could.
Established. On camera, at a press conference, in English, in answer to a hostile question.
Four months later his party had forty-four seats.
Twelve years on, some of that has held and some has not.
Part Fourteen credits him, as Finance Minister, with executing the 1991 reforms. Part Fifteen credits his government with the transparency law and the employment guarantee. Part Sixteen credits it with education and food as rights, and with the coal auctions it legislated in 2010.
Part Sixteen also finds a six-year delay on those auctions, a Lokpal legislated and not constituted, and the least productive full-term Parliament in Indian history.
Both halves of that are in this series and neither cancels the other. Whether history has been kinder is not something this book can settle. What it can say is that the sentence was a prediction, it was checkable, and it has been partly borne out — which is more than can be said for any other statement in this part.
There is an obvious objection to treating any of that as creditable, and it deserves stating.
The objection: this is a politician asking to be judged later because he is being judged badly now. Every leader in trouble says some version of it and it costs nothing.
The reply: mostly true, and the sentence would be unremarkable from almost anybody. What makes it worth recording is the second half — than the contemporary media, or for that matter the Opposition in Parliament — because it names the two institutions whose job is to hold a government to account and says, in effect, that both had got it wrong.
That is a serious claim and a Prime Minister making it at a press conference is answerable for it. Part Sixteen tested the largest version of it — the audit figures — and found the government substantially right about the number and wrong about zero. On the record so far the claim is neither vindicated nor absurd, which is an unusual place for a political sentence to end up.
Four statements, then, in a chapter this book expected to be able to fill with two.
Three real statements, all verifiable, all to somebody’s credit.
11 August 2005, Rajya Sabha: a Prime Minister apologised to the Sikh community for 1984. The only such apology in this series.
After 1962: Nehru told Zhou Enlai that India would never accept a line created by aggression, whatever the cost.
5 December 1961: a Congress MP told his own Prime Minister, in the House, that worthless ground is still ground. The best sentence in this part came from inside the party it prosecutes.
4The Big Tree
On 19 November 1984 the Prime Minister, Rajiv Gandhi, addressed a very large crowd on the Boat Club lawns near India Gate in Delhi. The occasion was the first birth anniversary of Indira Gandhi since her assassination nineteen days earlier.
Part Ten described the five days that followed that assassination. Part Eleven described the forty-one years since. This chapter is about nine words spoken at the end of the third week.
4.1 — What was said
The speech was in Hindi. Its structure, in summary, was this: that Indira Gandhi should be remembered; that the reason for her killing should be remembered; that those who might have been behind it should be remembered; that after her killing there had been disturbances in the country; that the anger of the people had been very great; and that for some days it had seemed that India was shaking.
Then the operative sentence.
In Hindi: “jab bhi koi bada ped girta hai, to dharti thodi hilti hai”
Nine words. Spoken to a crowd of thousands. Filmed. The footage was released publicly in November 2015 by the lawyer H. S. Phoolka, who had spent decades on the 1984 cases, and the audience can be heard applauding.
Established to the highest standard in this part: video, date, venue and the original language all fixed.
4.2 — The four translations
Here is where this chapter earns its place, because that nine-word Hindi sentence circulates in English in at least four forms, and they say different things.
| Version | Where it is used | What it does to the meaning |
|---|---|---|
| When a big tree falls, the earth shakes | The most common form in circulation | Drops “thodi” — a little. Makes the shaking total rather than slight |
| When a big tree falls, the earth shakes a little | Closest to the Hindi | Keeps the qualifier. Still describes mass killing as a tremor |
| Whenever a big tree falls, the ground shakes a little | Used with the 2015 video | “Whenever” makes it a general law rather than an observation |
| When a mighty tree falls, it is only natural that the earth around it does shake a little | The Indian National Congress’s own account | Adds “mighty”, adds “only natural”, and adds “around it” |
The fourth row is not a hostile translation. It is the version published in the party’s own material, by a party worker who says he was present.
Look at the fourth row again, because it is the most remarkable thing in this part.
A party’s own version of a damaging quotation is normally the mildest one available. That is what a defence is for. Here it is the harshest.
The Hindi says the earth shakes a little. The Congress’s own English says it is only natural that it does. “A little” is a description of scale. “Only natural” is a statement about causation and expectation — it says the shaking follows from the falling and that nobody should be surprised. It converts an observation into a justification, in the mouth of a defender.
Why would anyone do that? Because the paragraph it sits in is defending the timing — the point that the speech came sixteen days after the killings had stopped and was not an instruction. To make that argument, the defender had to make the sentence sound like a reflection rather than a command, and the word that does that is “natural”. A defence constructed to answer one charge routinely concedes a larger one, because the defender is looking at the charge and not at the concession.
4.3 — What the metaphor says
Set the translations aside for a moment and look at the image, because that is where the content is and almost nobody examines it.
A big tree falls. The earth shakes. Two things in the sentence, and the speaker’s mother is one of them.
The image is doing more work than it appears to, and it is worth taking apart.
The sentence assumes, without ever saying so, that the people who were killed are the ground.
That is not a rhetorical flourish; it is the grammar of the image. The tree is a person, named and grieved and identified with the nation. The shaking is what happens to everything else — undifferentiated, unnamed, and by the nature of ground, not something to which anything is done. Ground does not have a case. It does not have a name. Nobody is responsible to it.
So the metaphor does three things at once, and it does them so economically that they pass unnoticed. It makes one death a felling and three thousand a tremor. It makes the killing a natural event rather than an act. And it removes any actor from the second half of the sentence entirely — earth shakes; nobody shakes it.
Part Eleven found that for most of those killed in November 1984 there has never been a document naming anybody. Part Ten found 587 reports for 2,733 deaths and around 240 closed as untraced.
Nineteen days after those killings, the head of government described them in a form of words with no agent in it. Forty-one years later the files have no names in them either. This book is not claiming the first caused the second. It is saying that the sentence and the record have the same shape, and that this is the most disturbing thing in this part.
4.4 — The defence, at full strength
There is a real defence of this speech and it deserves to be put properly, because the popular version of it is unfair in one specific way.
First, the timing. The killings described in Part Ten ran from 1 to 4 November. This speech is from the nineteenth. It cannot have caused what had already stopped a fortnight earlier, and the widespread description of it as an incitement is simply wrong on the dates.
Second, the surrounding words. The same speech contains an appeal against anger and anguish. Mani Shankar Aiyar, writing in 2011, argued that Rajiv Gandhi had made an appeal for peace and had been seen rebuking Congress workers who were shouting the slogan associated with the killings.
Third, what the speaker did afterwards. He appointed a commission of inquiry in 1985.
Where the defence wins: on incitement, completely. This book does not say and has never said that the speech caused any death. Part Eleven declined to name individuals against whom no body had found and the same discipline applies here.
Where it fails: the charge is not incitement. It is that nineteen days after roughly three thousand citizens were killed, the head of government described it to a cheering crowd as the ground shaking when a tree falls — and that the sentence sits between a reminder of who might have been behind the assassination and a call to remember. The timing defence answers a charge nobody serious is making.
4.5 — What it is evidence of
Chapter One said this part would state, for each statement, what it actually establishes. Here is the answer for this one, and it is narrower than the sentence’s reputation.
The crowd applauded.
That is the finding. Not that a Prime Minister said a terrible thing — politicians say terrible things — but that he said it in public, from a platform, on his mother’s birth anniversary, to a very large audience, and correctly predicted how it would be received.
A statement is evidence of what a speaker believed he could say and be applauded for. On 19 November 1984, in Delhi, this is what that was.
Part Eleven found that for most of those killed there has never been a document naming anybody. This chapter finds the other half of the same fact: there was, at the time, no political cost to describing them that way.
4.6 — And it did not go away
One more date belongs here, because it establishes that the sentence was not a slip that the party spent forty years regretting.
On 21 May 2022, the anniversary of Rajiv Gandhi’s death, the Leader of the Congress in the Lok Sabha, Adhir Ranjan Chowdhury, published a tribute graphic carrying that quotation as the thing to remember him by. It was taken down shortly afterwards and replaced with a different quotation.
And the published version is not an old document either. The account carrying the “mighty tree” wording — the one that adds only natural — appears in the party’s own material as a defence of the speech, written by somebody who says he was in the crowd that day, and it is still there.
So the position, forty-one years on, is this.
The party has never disowned the sentence. What it disputes is the timing — that the speech came after the killings had stopped — and to make that argument it publishes a version of the words that is stronger than the Hindi.
In 2022 a senior figure offered the sentence as a tribute and withdrew it within hours.
A party that had decided the sentence was indefensible would have said so once, in one place, and stopped explaining it. What has happened instead is forty-one years of explanation, and the explanation keeps producing a worse sentence than the one being explained.
The 2022 episode is the sharpest evidence of that, and it is evidence the party produced itself.
The 2022 post is documented, dated and was published by the person who then led the party in the lower house of Parliament. It was deleted within hours and replaced.
The deletion is what makes it evidence. Somebody in 2022 chose that sentence as a tribute, and somebody else in 2022 recognised within hours that it could not stand as one.
Both of those facts are about the party’s own relationship to the sentence, thirty-eight years later, and neither depends on anything a critic says.
Which brings this chapter to the only summary of it that the evidence supports.
19 November 1984, Boat Club lawns, Delhi. Nine words in Hindi, on video, to a crowd that applauded.
The speech came nineteen days after the killings and did not cause them — this book says so plainly and the incitement charge is wrong on the dates.
The charge is that it was said at all, and that it was received the way the speaker expected. And of the four English versions in circulation, the harshest is the Congress party’s own — because the word it added to defend him was “natural”.
5Contempt
Three statements, from 2014, 2017 and 2019, made by two men who were not ministers when they made them and who had both begun their political careers alongside Rajiv Gandhi.
They belong together because they are all about the same thing: who is fit to be in the room.
5.1 — The tea
Date: 17 January 2014. Venue: the All India Congress Committee’s pre-election plenary session, Talkatora Stadium, New Delhi, in an interview given in the convention hall to a video agency. Speaker: Mani Shankar Aiyar, a former Union minister.
Asked about the man the opposition was expected to nominate for Prime Minister, he said that Narendra Modi would never become Prime Minister in the twenty-first century, and that if he wanted to come and sell tea at the venue, space would be made for him.
There is a wrinkle here that matters, and this book’s own rule requires it to be stated.
Aiyar has written since — in a book published in 2024 — that he never used the word chaiwala, which is the word the remark is universally known by.
On the record he appears to be right about the word and wrong about the substance. What he offered was space to sell tea, at a party conference, to a man who had sold tea. The insult did not require the noun.
This is a small illustration of Chapter Two’s third pattern working in the opposite direction from usual: a remark made more quotable by the addition of one word that was not said. Fabricated as to the word; accurate as to the offer.
What the remark cost is measurable, and it is not measured in offence.
Modi was sworn in as Prime Minister on 26 May 2014 — four months and nine days after the remark.
The tea-selling became a central image of the campaign that followed and was used deliberately and to great effect.
This book is a prosecution document and will say the obvious thing: the remark was not merely offensive. It was, as political judgement, one of the worst single sentences uttered by anybody in modern Indian politics, and it was uttered by a man whose party then lost 162 seats.
Where a remark is made turns out to matter more here than anywhere else in this part.
The All India Congress Committee is the party’s national body. A plenary is its full conference — thousands of delegates, held in a stadium or a convention hall, with the leadership on a platform and the national press in attendance.
It is the most formal setting a party has. Statements made there are made in the party’s own house, to its own people, with cameras present.
That matters for what follows. A remark to a reporter in a corridor is one thing. The same remark in the convention hall of a pre-election plenary is a statement about what the party’s senior figures think is acceptable to say among themselves.
5.2 — The word
Date: 7 December 2017. Venue: during the campaign for the Gujarat assembly election. Speaker: Mani Shankar Aiyar.
He described the Prime Minister as a neech kism ka aadmi.
In ordinary Hindi neech means low, base, vile — a comment on somebody’s character.
It is also, and this is why the remark exploded, a word with a caste register. Neech jaati means low caste. A phrase that means “a base sort of man” in one reading means something else entirely in another, addressed to a man who had made his origins a central part of his public identity.
Aiyar’s defence at the time was that his Hindi was poor and he had meant only the first sense. Whether that is believable is a matter of judgement. What is not in doubt is that both readings were available and that the second was the one that travelled.
What the party then did with him is the point, and it turns on a distinction most readers have no reason to know.
Suspension puts a member’s rights on hold. It is reversible by the same authority that imposed it, usually without any public process, and it does not require the member to withdraw anything.
Expulsion ends membership. Readmission requires a fresh decision.
The difference is the whole of Chapter Nine’s second charge. A party that suspends a member in the middle of a campaign has taken a visible action at the moment of maximum attention. Whether it was a punishment depends entirely on what happens eight months later, when nobody is looking.
He was suspended from the Congress the same day. The suspension was revoked on 18 August 2018.
In May 2019, in an article published during the general election, he referred back to the remark and asked whether he had not been prophetic. Asked about it by reporters, he swore at them.
5.3 — Hua to hua
Date: 9 May 2019. Venue: a press interaction during the general election campaign. Speaker: Sam Pitroda, a technology adviser who had worked with Rajiv Gandhi and who then headed the Congress’s overseas organisation.
Asked about the killings of November 1984, he said, in Hindi: what is there about ‘84 now — “'84 mein hua to hua” — and went on to ask what the present government had done in five years.
“hua to hua” — three words. It means, roughly: if it happened, it happened.
Established. On camera, dated, and circulated by a news agency the same day.
Part Eleven’s subject is the forty-one years of non-accountability that followed November 1984. Part Ten’s is the five days themselves. Chapter Three of this part records a Prime Minister apologising for them in the Rajya Sabha in 2005.
Fourteen years after that apology, a senior figure of the same party disposed of the same events in three words.
What happened next is genuinely to the party’s credit and this book will say so.
The defence says: the party did not defend it. Rahul Gandhi condemned the remark directly and without the usual formula about sentiments being hurt, Pitroda apologised, and the whole thing was over in days. Compare the handling of the 2017 remark, where a suspension was quietly reversed eight months later.
And that comparison is correct. It is the clearest example in this part of a party learning something between one incident and the next, and it should be recorded as such.
Where the defence fails: a condemnation is a statement about a statement. Part Eleven’s finding was that the machinery which processed 1984 has produced almost nothing in four decades, and that nothing in the six ordinary laws that would have changed it has been passed. The apology of 2005 and the condemnation of 2019 are the two ends of a period in which nothing was done, and they are both made of words.
5.4 — Why these two men
Three statements, two speakers, and the obvious question is one this part has not yet asked: why were either of them anywhere near a microphone during an election?
Neither held government office when they spoke. Neither commanded a bloc of seats or a state unit. Neither had won a national election recently. Both had begun in politics alongside Rajiv Gandhi in the 1980s and had remained close to the family for thirty years.
What they had was proximity, and proximity had lasted longer than anything else about their careers.
A party leader who staffs the front rank by electoral weight ends up surrounded by people who can survive without him. A party leader who staffs it by loyalty ends up surrounded by people who cannot — which is safer, and which means the people speaking for the party in an election are chosen for a quality that has nothing to do with speaking for the party in an election.
Both men were extremely good at the thing they had been selected for. Neither was selected for this.
That is not a fact about two individuals and it is not confined to one party. It is worth naming because this series has found the same shape before, in a much graver setting.
The test is what happened afterwards, and it is unusually clean.
The speaker of December 2017 was suspended and readmitted eight months later. In May 2019 he published an article defending the remark, and the party condemned that too, and nothing further happened.
The speaker of May 2019 apologised within days after the party leadership condemned him directly. He remained in his position.
In neither case did the remark affect the speaker’s standing with the leadership. That is the evidence that the selection criterion was never public performance — because a criterion that had been violated would have produced a consequence, and none did.
Part Eight found the Congress split of 1969 was about candidate selection rather than ideology, and Part Twelve found a party organisation answering to a leader rather than to an electorate. Chapter Five is what that produces sixty years later at the level of who is standing in front of a camera in January.
5.5 — What the three have in common
None of the three speakers held government office when they spoke. All three were close to the party leadership. All three remarks were made during or immediately before an election campaign, and all three were used against the party by its opponents within days.
Three remarks. 2014, 2017, 2019. Two speakers, both of whom had begun in politics under Rajiv Gandhi in the 1980s.
Nobody was speaking for the government. Nobody was announcing a policy. In no case did anything change as a result except the tone of a campaign.
And yet these are, by a distance, the most widely known Congress statements of the last fifteen years — better known than anything in Parts Fifteen and Sixteen, which are about six hundred deaths and three and a half lakh crore rupees.
Chapter Eight is about why that is, and the answer is not flattering to anybody, including the reader.
Three sentences, then, and between them they did more visible damage to this party than anything in Parts Fifteen or Sixteen.
17 January 2014, Talkatora Stadium: space would be made if he wanted to come and sell tea. He was Prime Minister four months later. The word chaiwala was added afterwards and was not said.
7 December 2017: neech kism ka aadmi. Suspended that day; readmitted eight months later; described as prophetic in 2019.
9 May 2019: hua to hua — three words for November 1984, fourteen years after a Prime Minister had apologised for it in Parliament. This one the party condemned within days, and that is to its credit.
6Denial
On 7 January 2011, at a press conference in Delhi, the Minister for Communications, Kapil Sibal, said that the allocation of 2G spectrum had caused no loss to the exchequer.
Part Sixteen is about what that allocation was and what came of it. This chapter is about the press conference, and it is the only statement in this part that was followed by a change in what anybody did.
6.1 — What was said
Two things were said that day and only one of them is ever quoted.
The one everybody quotes: that the audit figure of one lakh seventy-six thousand crore rupees was, with respect to the Comptroller and Auditor General, “so utterly erroneous” — and that the loss to the exchequer was zero.
The one almost nobody quotes: that prima facie there had been something wrong in the procedures adopted for the allocation.
Established. Both were said at the same press conference, by the same man, in the same hour.
Now set those against what Part Sixteen established.
| What was said on 7 Jan 2011 | What happened | Verdict |
|---|---|---|
| The ₹1.76 lakh crore figure is utterly erroneous | The CBI’s own charge sheet used ₹30,984 crore. The November 2012 auction largely failed. The Finance Minister called the figure a myth in 2013. | Substantially right about the figure — see Part Sixteen, Chapter Three |
| The loss was zero | The CBI Director told the Public Accounts Committee on 15 February 2011 that saying zero loss was wrong. Spectrum auctions raised about ₹1.10 lakh crore in 2015 alone. | Wrong, and known to be a stretch at the time |
| Prima facie something was wrong in the procedures | On 2 February 2012 the Supreme Court cancelled all 122 licences, holding the method arbitrary. That finding has never been disturbed. | Correct, and it was the minister’s own government’s position |
Row three is the government’s own accurate assessment of its own conduct, delivered thirteen months before the Supreme Court reached the same conclusion, and it has been almost entirely forgotten.
Everybody who remembers this press conference remembers it as a denial, and assumes that the minister’s problem was that he was wrong.
He was not wrong about the thing he is mocked for. The figure really was soft; Part Sixteen spent a chapter establishing that, and the prosecuting agency’s own number was a sixth of it. What he was wrong about was the smaller claim — zero — which he did not need to make and which destroyed the credibility of the larger one.
And the sentence that was right about his own government’s misconduct is the one nobody kept.
So the pattern is not “government denies wrongdoing”. It is: a defence containing an accurate technical point and an indefensible absolute claim, in which the absolute claim consumed the accurate one.
Part Sixteen’s Chapter Three argued that both sides of Indian politics used presumptive-loss figures when convenient and called them myths when not. Here is the same failure inside a single hour, by a single speaker — and it is why that whole argument was lost for a decade by the side that had the better of it.
Two terms in that table need explaining before the chapter goes further, because the argument rests on both.
Prima facie is Latin for “at first sight”. In legal use it means: on the material available so far, and before anything has been tested, this appears to be so. It is a real finding and a provisional one.
The Public Accounts Committee is the parliamentary committee that examines the Comptroller and Auditor General’s reports. By long convention its chairman is a member of the opposition — which makes it the one committee in the Indian Parliament not controlled by the government, and the reason the CBI Director’s evidence in February 2011 became public at all.
Part Sixteen found that a committee composed to reflect the House has never established anything against the party running it. The Public Accounts Committee is the exception, and it exists because of a convention rather than a law.
Which raises the question of how much weight a press conference can bear at all.
A press conference is the weakest of the four sources in Chapter One’s table, and this one is an exception for a specific reason.
It was televised live and was the government’s own set-piece — a minister with charts, taking questions, for the purpose of being quoted. Both statements in the box above come from that broadcast and neither has ever been denied by the speaker.
What makes the second one recoverable at all is that a journalist wrote it down at the time. It appears in a long magazine profile published in 2012, in a passage explaining that the minister was not the first person to say the loss was notional — and it is preserved there almost incidentally, in a parenthesis.
Chapter Nine’s recommendation exists because of parentheses like that one. The accurate half of a government’s most-discussed press conference survives in Indian public memory only because one writer happened to include it in a bracket.
6.2 — The one statement that changed something
Chapter One said that two of the eleven statements in this part were followed by a change in what the government did. Chapter Two covered the other one — the ordinance of September 2013, killed within days of being denounced.
This is it. The zero-loss claim did not close the argument; it opened it. The Public Accounts Committee summoned the CBI Director, who contradicted the minister on 15 February 2011. The demand for a joint parliamentary committee, which the government had resisted through an entire lost parliamentary session, became harder to refuse. Part Sixteen’s Chapter Seven records what that session cost.
A minister said in January 2011 that the loss was zero.
Within six weeks his own government’s investigating agency had told a parliamentary committee that this was wrong. Within thirteen months the Supreme Court had cancelled every licence. Within three years his party had 44 seats.
No statement in this part was more consequential and none was more self-inflicted. The technical case was available, it was largely correct, and it was thrown away for a word.
There is a real defence of the man who said it, and it is stronger than his reputation on this suggests.
The defence says: a minister defending his government at a press conference is doing his job, and the substance of what he said has been substantially vindicated. Part Sixteen concedes that the presumptive figure was soft, that the CBI used a much smaller number, and that everyone accused was acquitted in 2017. If this book grades evidence honestly, the minister should get the benefit of it.
The reply, which concedes most of that: he does get the benefit of it, and Table 4’s first row says so. The charge is not that he defended the government. It is the word zero.
Where this is different from ordinary political overstatement: the same speaker, in the same hour, said that something had prima facie been wrong with the procedure. He had the accurate position in his hand and said it out loud. Choosing to lead on “zero” instead was not ignorance. It was a decision about what would carry a news cycle, and it cost his side an argument they were winning on the facts.
Which is the difference between being wrong and being careless, and this was the second.
Which leaves one question about the man rather than the statement.
6.3 — What the speaker did next
One coda, because it bears on how much weight a defence of this kind carries.
Kapil Sibal left the Congress in 2022 and entered the Rajya Sabha with the support of a different party. He has since appeared in court against positions his former party’s opponents hold, and for several of the figures this series has discussed.
That is not a charge. Lawyers act for whoever instructs them and politicians change parties. It matters here for one narrow reason.
The most-mocked defence of the UPA’s largest scandal was delivered by a man who was a professional advocate before he was a minister and after he stopped being one.
That explains something about the press conference of 7 January 2011 that “he was lying for his party” does not. Chapter Six found a defence containing an accurate technical point and an indefensible absolute claim, delivered in the same hour.
That is what advocacy looks like when it is done badly: the strongest available argument and an unnecessary overreach, presented together, because the advocate has decided which one will lead.
This book is an advocacy document and says so on its cover. Chapter Six is therefore about a failure this book could commit on any page — and Part Sixteen’s author’s note records the moment it nearly did, in deciding where to put the chapter on presumptive loss.
Which is the most useful thing this chapter has to offer, and it is a warning rather than a charge.
7 January 2011, New Delhi. Two statements in one hour. One is famous and wrong: the loss was zero. One was correct and is forgotten: prima facie something was wrong in the procedures.
Thirteen months later the Supreme Court cancelled all 122 licences on exactly that ground. The minister had described his own government’s misconduct accurately and then buried it under a word he did not need.
This is the only statement in this part that changed anything, and it changed it against the speaker.
7Indifference
The oldest statement in this part is also the one whose consequences are still on a map.
7.1 — What was said
Date: 5 December 1961. Venue: Parliament, during a debate on the border with China. Speaker: the Prime Minister, Jawaharlal Nehru.
Describing the disputed ground in Aksai Chin, he said that not a blade of grass grew there; that it was some seventeen thousand feet high and uninhabitable; and that India had not even known where it was.
“Not a blade of grass grows there.”
Six words. Spoken in Parliament, in English, ten months before the Chinese attack that Part Six describes.
Established as to substance and date. Contested as to which House — accounts place it in both — and this book cannot resolve that. Chapter Ten records it.
The reply came at once, from a Congress member, and Chapter Three has already given it: Mahavir Tyagi pointed at his own bald head and asked whether it followed that the head had no value. The House laughed, including the Prime Minister.
This is the oldest statement in the part and it has the weakest sourcing, which is the opposite of what a reader might expect.
Parliament kept a verbatim record in 1961 and still does. But the record of that day is not what most people are working from. The exchange survives principally through histories written decades later and through the retelling of Tyagi’s reply, which is much better remembered than the sentence that provoked it.
Those accounts differ. One respected history has an opposition member taunting Nehru and a Congress member adding the retort; other accounts have Tyagi answering the Prime Minister directly. Sources also place the exchange in different Houses.
What is common to every version is the substance of what Nehru said, the date, the fact that a Congress member made the reply, and the terms of it. Everything beyond that this book flags rather than resolves — and Chapter Ten lists it.
7.2 — Why it is in this part
Not because it was callous. It was not — it was, on its own terms, an accurate description of the terrain. That is precisely the difficulty.
A high desert plateau of roughly thirty-eight thousand square kilometres between Ladakh and Xinjiang. Almost no rainfall, almost no vegetation, effectively no population.
It is also the ground the Xinjiang–Tibet highway crosses. Part Six set out the sequence: about a hundred and sixty kilometres of that road runs across Aksai Chin, it was completed in the second half of 1957, and the Indian government was told about it by its own Foreign Secretary in early 1958.
Its value was never agricultural. It was that somebody else needed to drive across it.
Which is why the sentence is in this part at all, and the reason has nothing to do with tone.
The sentence assumes that the value of territory is what grows on it. Almost nobody notices that assumption, because for most of human history it was true — land was worth what it produced.
By 1961 it had stopped being true, and the specific thing that had made it stop being true was sitting on the ground under discussion. A road across a plateau where nothing grows connects two regions of a rival state to each other. That is the value, and it is invisible to any test involving grass.
Part Six’s charge against this policy was about warnings ignored and a Forward Policy adopted over professional objection. This chapter finds something underneath that: a frame of mind in which a place is worth what it yields, so a barren place is worth nothing, so losing it is not a loss.
And the correction came from inside the House, in one sentence, from the Prime Minister’s own benches. Tyagi’s joke is not a joke about hair. It is the observation that a thing’s worth is not the same as its produce — which is the whole of what was wrong with the original sentence, delivered in eleven words and received with laughter.
7.3 — The year that moved
One more thing about this statement belongs here, and it is the reason Chapter Seven and Chapter Two are really the same chapter.
Most versions of this remark in circulation date it to 1962 — after the war, or during it. The date is 5 December 1961: ten months before the Chinese attack.
One year. It sounds like a detail and it reverses the meaning.
Said in 1962, the sentence is a Prime Minister explaining away a defeat that has already happened. It is shabby, and it is retrospective.
Said in 1961, it is a Prime Minister telling Parliament that a piece of ground under active dispute — with a road across it, built four years earlier, which his own Foreign Secretary had told him about in 1958 — was not worth anything.
The second is far worse, and it is the true one.
The misdating makes the statement less damaging, and it is the version almost everybody repeats — including people using it to attack the speaker.
How an error like that survives sixty years is itself the argument of this part.
This is the third pattern from Chapter Two’s table, working in an unexpected direction.
Chapter Two described the travelling quote and the improved translation, both of which make a statement worse than it was. Here the movement runs the other way: an error that has been repeated for sixty years softens the charge, and it has been repeated most enthusiastically by the speaker’s critics.
Nobody did this deliberately. A war in 1962 is the thing everyone remembers; a parliamentary debate in December 1961 is not. The date drifted towards the memorable event, which is how dates drift.
Which is the argument for Chapter Nine’s archive, stated as sharply as this part can state it: the error was not ideological and it fooled everybody equally.
7.4 — What followed
Part Six gives the account in full and this chapter will not repeat it. In summary: China attacked on 20 October 1962; a ceasefire was declared unilaterally on 21 November; India withdrew in the east and China retained Aksai Chin.
Roughly thirty-eight thousand square kilometres — larger than Kerala, and about the area of Switzerland.
It has been administered by China for sixty-three years. Indian maps show it as Indian territory. There is a road across it.
Part Six also recorded that on 25 October 1962 China offered a ceasefire with mutual withdrawal twenty kilometres behind the line of actual control, and that India rejected it — a fact most Indian accounts omit.
The sentence about grass did not lose Aksai Chin. The road was built in 1957 and the government knew in 1958. What the sentence records is how the loss was being explained to Parliament three years later, and the explanation was that there was nothing there worth having.
There is a serious defence of the remark and it has rarely been put properly, so it is put properly here.
The defence says: the remark is quoted without its context, which was a Prime Minister explaining to a House demanding military action that the ground in question could not sustain a supply line, was at seventeen thousand feet, and had never been surveyed or administered by India in any meaningful way. Every one of those things was true and all of them bore on whether a war was winnable.
Chapter Three of this part records that the same man, after the fighting, told the Chinese premier that India would never accept a line created by aggression whatever the cost. A man who did not care about the territory does not write that letter.
Where the defence wins: on indifference as a personal charge, largely. The letters are real and Chapter Three gives them full weight.
Where it fails: the sentence was not a private assessment. It was the reason offered in Parliament, and reasons offered in Parliament are how a country is taught what to care about. A government that explains a loss by saying the thing lost was worthless has not prepared anybody for what follows — and ten months later something followed.
Which is the charge this chapter makes: not coldness, but a way of explaining a loss that left a country unprepared for it.
5 December 1961, Parliament. Six words about Aksai Chin: not a blade of grass grows there.
It was accurate. Nothing does grow there. What grew there was irrelevant — the value was a road, built in 1957, which the government had been told about in 1958.
A Congress MP corrected the Prime Minister on the spot, in the House, with a joke about his own bald head. Ten months later China attacked.
8What They Have in Common
Eleven statements, spread across fifty-three years, by seven different people, in five different settings. This chapter asks what — if anything — they are collectively evidence of.
8.1 — The ledger
| Date | Speaker and venue | Grade | Did anything follow? |
|---|---|---|---|
| 5 Dec 1961 | Nehru, Parliament — Aksai Chin | Established; House contested | No. Corrected in the House and ignored |
| After Oct 1962 | Nehru to Zhou Enlai — the letters | Established | To his credit. Chapter Three |
| 19 Nov 1984 | Rajiv Gandhi, Boat Club — the big tree | Established; video | No. Applauded at the time |
| 9 Dec 2006 | Manmohan Singh, NDC — first claim | Words established; meaning misrepresented | A clarification the next day. Nothing else |
| 11 Aug 2005 | Manmohan Singh, Rajya Sabha — the apology | Verbatim record | Compensation package, Jan 2006. Part Eleven |
| 7 Jan 2011 | Kapil Sibal, press conference — zero loss | Established | Yes — against the speaker. Chapter Six |
| 27 Sep 2013 | Rahul Gandhi, Press Club — the ordinance | Established; the tearing fabricated | Yes — the ordinance was withdrawn. Chapter Two |
| 17 Jan 2014 · 7 Dec 2017 · 9 May 2019 | Aiyar and Pitroda — tea, neech, hua to hua | Established; one word added | No policy change. Large electoral effect |
Two of the eight rows are to the credit of the speaker. One changed government behaviour, and it changed it against the person who spoke. The rest changed nothing except how people felt.
8.2 — The shape
Read down the last column. It is the finding of this chapter and it is uncomfortable for the genre this part belongs to.
Statements do not do things. Not one of these sentences moved a policy, freed a prisoner, released a file, passed a law or stopped a killing. The only one with a measurable consequence produced the opposite of what its speaker intended.
So what are they for?
This series assumes — and so does everybody who compiles a list like this one — that what a government says reveals what it is. That is the premise on which a whole part of a nineteen-part audit has been built.
Test it against the ledger above. The most callous sentence in this part, in 1984, came from a government that then appointed a commission of inquiry. The most honourable, in 2005, came from a government that then did almost nothing for four decades of victims, as Part Eleven established. The most accurate technical judgement, in 2011, was delivered by a man in the act of saying something false.
Words and acts do not correlate. A cruel sentence does not predict a cruel policy. A decent sentence does not predict a decent one.
Which means this part is evidence of something narrower than it looks: not what these governments were, but what they believed they could say and survive. That is a fact about the audience as much as the speaker — and on the evidence of 1984, when a crowd applauded, and 2014, when a remark about tea cost 162 seats, the audience changed far more over these fifty-eight years than the party did.
Before that conclusion is allowed to stand, the size of the thing it rests on has to be stated.
A finding drawn from eight statements is a finding drawn from eight statements, and this book is not going to dress it up.
The sample is small, it was chosen by one person, and it was filtered by a rule — Chapter One’s three tests — that Chapter One itself concedes selects for guarded speech. Chapter Ten records that roughly four in five candidates were excluded.
What the sample can support is a claim about these eight: that none of the six damaging ones changed a policy, and that the two which changed anything were the two whose speakers were arguing about a decision rather than describing an event.
What it cannot support is a general law about political speech in India. If somebody assembles a hundred statements to the same standard and finds otherwise, this chapter is wrong, and that is the correct way for it to be wrong.
8.3 — The two that worked, and what they had in common
Nine of the eleven changed nothing. Two did, and they are the most useful rows in the table because they are so obviously unlike the rest.
| 7 January 2011 | 27 September 2013 | |
|---|---|---|
| Subject | A decision already taken and being defended | A decision already taken and being attacked |
| Speaker’s position | A minister, defending his own government | Not in government, attacking his own government |
| What it asked for | That an audit figure be disbelieved | That an ordinance be withdrawn |
| What followed | The CBI contradicted him within six weeks; the argument was lost | The ordinance was withdrawn |
The other nine statements in this part describe events, characterise opponents, or explain the past. None of them asks for anything.
That is the distinction, and it is not about tone, party or decade.
A statement changes something when it is about a live decision that has not yet become irreversible.
The ordinance of September 2013 had been approved by the Cabinet and not yet promulgated. There was still something to stop, and a sentence stopped it.
The audit figure of 2010 was already public and the argument about it was already running. There was something to win, and a sentence lost it.
By contrast: nothing said on 19 November 1984 could have altered anything, because everything it referred to had already happened. The same is true of Aksai Chin in 1961, of the apology in 2005, and of all three remarks in Chapter Five.
Which means the famous statements in this part are famous precisely because they were consequence-free. A sentence that stops a Cabinet decision produces a news cycle and then a withdrawn ordinance. A sentence about people who are already dead produces forty years of quotation.
8.4 — What changed over the fifty-three years
Something did change across this part, and it is not the party.
In 1961 a Prime Minister said in Parliament that a piece of national territory was worthless, was corrected by a member of his own party, and the House laughed. Nothing followed. In 1984 a Prime Minister described the killing of three thousand citizens as the ground shaking and the crowd applauded. Nothing followed. In 2014 a former minister offered a man space to sell tea and his party lost a hundred and sixty-two seats.
Three sentences of roughly comparable carelessness, in 1961, 1984 and 2014.
The first two cost nothing. The third cost an election.
What is between them is not a change in the party’s manners. It is video, and then it is video that anybody can forward.
The footage of 1984 existed from 1984 and reached the public in 2015. The remark of January 2014 was on every phone in the country within a day.
That is the mechanism, and it produces a conclusion that cuts against the natural reading of this part.
The natural reading of this part is that the Congress became careless with its words. The statements cluster after 2010 and the worst electoral consequences are all recent.
The evidence does not support that. The two most serious statements here are from 1961 and 1984. What clusters after 2010 is not the carelessness but the cost of it.
A party that governed for most of fifty years without any of this being recoverable, and then found every sentence recoverable within hours, did not change. Its environment did — and a habit formed under one set of conditions was carried into another.
Which is a defence of a sort, and this book gives it: the leaders in Chapter Five were doing what leaders in Chapters Four and Seven had done without consequence. It is not a defence of the sentences. It is an explanation of why they kept being said.
8.5 — The one pattern that does hold
There is a shape, and it is not about cruelty.
In five of the nine rows, the speaker made a claim that was true in a narrow technical sense and false as an account of the situation. Nothing grows in Aksai Chin — and its value was a road. The earth does shake when a tree falls — and three thousand people had been killed. The audit figure was soft — and the loss was not zero. The list did include several groups — and the sentence was constructed so that it need not have been read that way.
Five statements. Each one defensible on a technicality and indefensible as a description of what was happening.
That is not lying. Lying would be easier to answer, because a lie can be disproved. A statement that is true in the narrow sense cannot be disproved at all — it can only be answered by somebody pointing out what it leaves out, which takes longer, is less quotable, and usually does not happen.
Mahavir Tyagi is the only person in this part who did it, and he did it in eleven words. That is why Chapter Three calls his the best sentence here.
This series has met that shape before, and it is worth naming the connection. Part Nine found the Emergency was constitutional — every instrument in the document, every threshold met — and found that this was worse than the alternative. Part Sixteen found presumptive-loss figures that were legitimate audit technique and useless as evidence of theft. A statement that is technically accurate and substantively false is the verbal form of the same thing, and it is the most characteristic single move in this entire series.
Eleven statements, fifty-three years, seven speakers. Nine of the eleven changed nothing at all. The two that did are the two whose speakers were arguing about a decision rather than describing an event.
Words and acts do not correlate. The cruellest sentence here came from a government that then appointed a commission; the most honourable from one that then did nothing for decades.
What does hold: five of the nine are true on a technicality and false as a description of what was happening. That is not a lie. It is harder to answer than a lie, and it is the most characteristic move in this series.
9What Would Have Prevented It
Every part of this series ends with this chapter, and this one is the strangest of the seventeen, because the honest answer to its title is: almost nothing, and most of the things that would have worked should not be done.
9.1 — The defence
One. They are sentences. Nothing in this part is an act of government. Nobody was jailed, no file was withheld, no asset was allocated, no election was affected by anything except the tone of a campaign. Sixteen parts of this series are about what governments did; this one is about what individuals said, mostly when they were not in office.
Two. Two of the eight are to the party’s credit, and this part put them in Chapter Three rather than at the end.
Three. One of the most famous is not what it is said to be. Chapter Two established that the 2006 sentence does not bear the meaning universally attached to it, and Chapter Five that a word central to the 2014 remark was not spoken.
Four. The party did learn. Compare the handling of December 2017 — suspension, quiet readmission eight months later, and the speaker calling himself prophetic — with May 2019, when the leadership condemned the remark within hours and an apology followed.
Five, and this is the strongest. A collection of damaging statements can be assembled about any political party in any country over fifty-eight years. The genre proves nothing about its subject that it would not equally prove about the alternative, and a series that presents one without acknowledging that has misled its reader.
The fifth point is testable and this book has tested it on itself.
While assembling this part I collected statements by leaders of the party this series is not auditing, applying the same three tests. Date, venue and wording were establishable for a comparable number of them, of comparable severity, across a comparable period.
Those statements are not in this part, because this part is not about that party and Part Eighteen is where the comparison belongs. But a reader is entitled to know that the exercise was done and what it produced, and it produced the conclusion that a list of ugly sentences is available about anybody who has been in politics long enough.
Which is why Chapter Eight’s finding is about what an audience will applaud rather than about what a party is.
9.2 — Where the defence loses
On two things, and they are narrow.
The 1984 speech is not equivalent to the others. A remark about tea is a discourtesy. Three words about the killings of 1984 are a discourtesy. Nine words on a public platform nineteen days after three thousand people were killed, delivered by the head of government to an applauding crowd, describing those deaths as the ground shaking, is not in that category. The others cost votes. This one told a country what its government thought had happened.
And the readmission. The suspension of December 2017 was reversed in August 2018 without any retraction, and in May 2019 the speaker described the remark as prophetic. A party is entitled to discipline its members as it sees fit. What it is not entitled to is the credit for a punishment it withdrew.
The objection to Chapter Nine’s second charge — that a suspension quietly reversed earns no credit — deserves an answer, because there is one.
A political party is a voluntary association. It is entitled to discipline members as it chooses, to change its mind, and to readmit somebody who has served eight months out of the party during which he lost his platform, his position and his standing. Eight months of that is not nothing, and a permanent expulsion for a sentence would be a harsher rule than most institutions in India apply to far worse conduct.
And the party’s behaviour has changed. In May 2019 it condemned a remark within hours and produced an apology; Chapter Five records that as genuine.
Where the objection fails: not on the readmission, which is defensible, but on what followed it. In May 2019 the readmitted member published an article referring back to the remark and asking whether he had not been prophetic — and the party condemned that too, and nothing happened.
A punishment that is reversed and then publicly repudiated by its subject, with no consequence, has been converted into a statement about the party rather than about him. That is the charge, and it is narrower than “the party did not care”.
9.3 — The rules, and why most of them should not be made
Sixteen parts of this series have ended with a table of ordinary laws that would have caught the failure and were never passed. This part cannot do that, and it matters to say why.
| The proposal | What it would do | Verdict |
|---|---|---|
| Criminalise the statement | Prosecute politicians for offensive speech about communities or individuals | No. Part Three traced what happened when the grounds for restricting speech were widened in 1951, and Part Nine what a state does with such powers |
| Expand the Election Commission’s power to bar candidates for speech | Disqualification for remarks made during a campaign | No. It places the decision with an appointed body and invites exactly the sanction problem found in Parts Thirteen, Fifteen and Sixteen |
| A parliamentary code of conduct with enforcement | Sanctions for members inside the House | Probably not. Part Sixteen found that a majority-composed committee has never established anything against its own side |
| A public verbatim record | Audio and transcript of every public statement by a minister or party officeholder, published within a fixed period, in the language spoken | Yes. It creates no offence and restricts nobody. It removes the ability to deny, to improve a translation, or to invent |
Only the fourth row is a recommendation. The first three are listed because they are the things people actually propose after an episode like these, and each of them would do more damage than the statements did.
The fourth row is worth a paragraph, because it is the only remedy in seventeen parts that solves the problem by adding information rather than by adding a penalty.
Every problem in this part is a problem of record.
The 1984 sentence circulates in four English versions because there was no authoritative one. The 2006 sentence was misread because the paragraph was not to hand. The 2014 remark acquired a word that was never said. The 2011 press conference is remembered for one of its two claims.
Every one of those is fixed by a published transcript and none of them is fixed by a punishment.
The video of the 1984 speech existed from 1984. It reached the public in 2015, released by a lawyer rather than by any archive.
9.4 — What an archive would actually fix
The recommendation in the fourth row of that table is unglamorous and it is worth being specific about what it would and would not do.
It would not stop anybody saying anything. That is the point. Every other proposal on that list works by making speech riskier, and this series has spent four parts describing what Indian governments do with powers over speech — the First Amendment of 1951 in Part Three, pre-publication censorship in Part Nine, the special statutes in Parts Ten and Fifteen.
What it would do is remove four specific abuses, all of which appear in this part.
| The abuse | Where it appears here | How a transcript stops it |
|---|---|---|
| Denying that words were said | The word chaiwala, disputed for a decade | The audio settles it in one minute |
| Improving a translation | Four English versions of nine Hindi words | The original language is published alongside |
| Clipping a referent | “They” in the 2006 speech | The paragraph is one scroll away from the sentence |
| Losing the half that mattered | The procedural concession of 7 January 2011 | Everything said in the hour is in one place |
Each of those cost this part real work, and in one case — the statement named in Chapter Ten — it cost the part a statement altogether.
Parliament already does it. The verbatim record exists, it is published, and there is almost no argument in Indian public life about what was said in the chamber.
Every statement in this part that is disputed was made outside Parliament. Every statement that is not disputed — the Aksai Chin remark’s substance, the apology of 11 August 2005 — was made inside it.
The correlation is complete across this part’s whole sample, and the institution that produces it has existed since 1952. The proposal is not to invent something. It is to do outside the building what has been done inside it for seventy-four years.
9.5 — Who would run it
A recommendation without an owner is the thing this series has criticised in five previous parts. So: who.
Not a ministry, and not a body appointed by one. Parts Thirteen, Fifteen and Sixteen all found the same failure — a decision handed to somebody political and answering accordingly — and an archive of what politicians said, curated by a government department, would be the purest possible example of it.
Who: the Election Commission, which already regulates campaign speech, already has a presence in every district during a poll, and is not a department of the government of the day.
What: audio, and a transcript in the language spoken, of any public statement by a minister, a chief minister, or an officeholder of a recognised party, at any event to which the press was admitted.
When: published within a fixed period — a week is enough — with a permanent public identifier and no power to remove anything once posted.
What it does not do: create an offence, restrict what anybody may say, or give anybody a power over anybody. It is a filing obligation.
Two objections are worth answering because both are real.
Both of the serious objections to it are worth stating and only one has a good answer.
First: it would chill speech. A politician who knows every word is permanently recorded will say less and mean less, and political language will become as guarded as parliamentary language already is. Chapter One’s own argument box conceded that the guarded register is the least revealing.
That is probably right, and it is the real cost. This book thinks it is worth paying, and a reader may disagree. What should not be in dispute is that the cost exists.
Second: nobody would enforce it. Part Sixteen found a Lokpal legislated in 2013 and not constituted until 2019. A filing duty with no penalty is a suggestion.
The answer to that one is better. An archive does not require enforcement to work, because the gap is itself the finding. If a party’s statements are systematically missing from the record, that is visible, dated and attributable — which is more than can be said for any of the five other recommendations in this part.
What makes the whole proposal unglamorous is that none of it needs inventing.
Everything proposed above already exists somewhere. The Election Commission exists. Recording equipment exists. The video of 19 November 1984 existed in 1984.
What does not exist is anybody whose job it is to put those things in one place.
Part Fifteen found seventeen intelligence alerts forwarded and never assessed, because assessment was nobody’s job. Part Thirteen found every component of a national displacement figure existing and the addition never made. This is the same shape a third time, and here the missing job is a filing clerk’s.
9.6 — The question turned on this book
This part assumes that collecting a government’s worst sentences is a legitimate part of auditing it. The author of this series agreed to write this part before thinking hard about whether that is true.
Chapter Eight found that words and acts do not correlate. Chapter Nine’s fifth defence point concedes that a comparable list is available about anybody. Put those together and the conclusion is uncomfortable: this part does less work than any other in the series, and it will be quoted more than all of them.
That is not an accident of this book. It is what statements are for. A sentence travels because it is short, needs no background, and can be felt rather than assessed. Part Sixteen’s finding about a six-year delay in legislating coal auctions is the most substantial charge in this entire series against any government, and it will never be as widely repeated as three words said in 2019.
So the honest thing is to say what this part is: the most quotable and least probative chapter of a nineteen-part audit — and to say it here, in the middle, where a reader who takes only one thing from these pages might take this.
Saying that in the middle of the part, rather than in a footnote at the end, is the only honest place to put it.
The defence is strong: these are sentences, not acts; two are to the party’s credit; one famous one was never said in the form quoted; and a comparable list is available about any party anywhere.
It loses on two: 1984 is not in the same category as the rest, and a suspension that is quietly reversed earns no credit.
And of the four remedies people propose after episodes like these, three would do more damage than the statements did. The one that works adds a transcript, not a penalty.
10An Honest List of What We Do Not Know
This part’s honest list is longer than most, because its rule is stricter than most. Everything below failed one of the three tests and was therefore left out.
10.1 — Left out for want of a date, a venue or a wording
This book began with a working list of roughly forty statements attributed to Congress leaders across seventy-nine years. Eleven survived.
Around forty statements in, eleven out. Roughly three in four failed.
They did not fail because they were shown to be false. Most of them failed because no date could be established, or no venue, or because every available version was a translation of a translation.
That is the single most useful number in this part, and it is a fact about the genre rather than about any party. Four out of five political quotations in circulation in India cannot be pinned to a day and a room.
The categories that failed, so that a reader can recognise them:
Statements attributed to Indira Gandhi during the Emergency. Several circulate. Part Nine’s account of that period rests on the Shah Commission, on court records and on the amendments themselves, all of which are documentary. The remarks attributed to her personally could not be fixed to a date and a venue and are therefore absent from both parts.
Remarks about the 26 November 2008 attacks and their investigation. A number of statements by Congress figures about who was responsible circulate widely. Part Fifteen covers the investigations in detail on the basis of charge sheets and judgments. The statements did not meet this part’s test.
Statements reported only in a memoir. Several of the most striking things attributed to Prime Ministers in this series exist only in somebody’s recollection of a private conversation. Part Twelve used a memoir for what its author saw in a courtroom and Part Fourteen for what a lawyer was told by his own client, and both said so. A private remark recalled years later by one participant is not a statement in the sense this part uses, and none is here.
Everything on social media without a source. The largest category by far.
10.2 — One that was left out, named
Describing a rule is cheap. Here is the rule costing this book something.
During the Gujarat assembly campaign of 2007 the Congress president, Sonia Gandhi, described her opponents in terms that became one of the best-known political phrases of the decade — maut ka saudagar, merchants of death. It was widely reported at the time, it was used against the Congress in that election and in every Gujarat election since, and there is no serious dispute that something of the kind was said.
It is not in this part.
This book could not establish, to the standard set in Chapter One, the day it was said, the town it was said in, or the wording either side of it.
Those things certainly exist. A campaign speech in a state election was covered by every newspaper in the country. But what is easily available now is the phrase, detached, repeated for eighteen years — which is precisely the condition Chapter Two describes.
A statement this book is confident happened, and cannot pin down, does not go in. If it went in on the strength of everybody knowing it, then Chapter Two’s whole argument collapses, because that is the identical basis on which every fabrication in circulation travels.
The same applies to many remarks about the investigations in Part Fifteen, to several statements attributed to Indira Gandhi during the Emergency, and to almost everything that circulates as an image with a name on it.
A reader who wants to test this book can do it in ten minutes and should.
Take the single most damning Congress quotation you are personally confident about. Do not ask whether it is true. Ask only: what day, what town, and what were the two sentences either side of it.
If you can answer all three, it belongs in a document like this one and this book has probably missed it. If you cannot, you have just discovered what four-fifths of this part’s working list looked like.
10.3 — Where this part may be wrong
Which House heard the Aksai Chin remark. Accounts place it in both. One respected history describes an opposition member taunting Nehru and a Congress member adding the retort, which does not match the version in which Tyagi answered the Prime Minister directly. Chapter Seven states the date and substance and flags the rest, and this book could not resolve it.
The exact wording of the 1984 speech in English. There is no such thing. It was spoken in Hindi and Chapter Four gives four English versions rather than choosing one. A reader who wants the sentence should watch the footage.
The 2014 remark. This part accepts the speaker’s own account that the word chaiwala was not used, and finds that the substance was worse than the word. A reader may reasonably think this book has been too generous to him, or not generous enough.
This book’s own translations. Chapter One says a translation is evidence about the translator. Several passages in this part summarise, in English, what was said in Hindi — the structure of the 19 November 1984 speech, the sense of hua to hua, the two readings of neech.
Every one of those renderings is a choice made by the author of this book, using exactly the discretion Chapter Four criticises the Congress for using in its own version of the big-tree sentence.
This part has tried to limit the damage in two ways: by giving the operative words in Hindi wherever they are known, so a reader who has the language can check; and by setting out the competing English versions rather than picking one.
Neither of those makes the problem go away. A reader who does not read Hindi is reading this author, and there is no arrangement of a book that fixes that.
It is the sharpest single instance of a difficulty running through all seventeen parts: a document that grades its evidence still has to be trusted about the grading. Chapter Nine’s archive is the only proposal in this series that would let a reader dispense with trusting the author, which is why it is the one recommendation this part makes.
The selection. Eleven statements out of seventy-nine years is a very small number and every one of them was chosen by the author. A different writer applying the same three tests would produce an overlapping but different eleven. There is no method that makes a selection objective, and this part does not claim one.
10.4 — What a reader can do with this
This is the only chapter in seventeen parts where the remedy is available to the reader rather than to a legislature, so it is worth setting out.
Every failure described in this part is a failure of checking, and every one of them is checkable by anybody with a phone in about two minutes.
Three questions, in order, applied to any political quotation:
One: what day? Not a year. If no day is available, stop — roughly four in five fail here, on the evidence of this part’s own working list.
Two: what room? Parliament, a rally, a studio, a press conference. Each carries different weight and each has a different record.
Three: what language? If the answer is Hindi and you are reading English, you are reading somebody’s choice, and Chapter Four shows what those choices do.
Nothing on that list requires expertise, a subscription or an archive. It requires being willing to lose a quotation you liked, which is the only genuinely difficult part and is the reason it is so rarely done.
This book lost about twenty-nine of them that way, and it lost the best one first.
10.5 — What this part owes Part Eighteen
Part Two of this series put four forward promises on the record and named the parts that owed them, and all four were paid. This part makes one, and it is the largest in the series.
Chapter Nine’s fifth defence point said that a collection of damaging statements can be assembled about any party in any country over fifty-three years, and that a series presenting one without acknowledging that has misled its reader. The How We Know box under it recorded that the exercise had been done — that statements by leaders of the party this series is not auditing were collected against the same three tests, and that date, venue and wording were establishable for a comparable number of comparable severity.
Part Eighteen will carry that list. Not as a defence of anything in this part, and not as a comparison of who said worse things, but subject to the identical rule: date, venue and operative wording, or it does not appear.
If Part Eighteen does not carry it, then Chapter Nine’s fifth defence point was a rhetorical concession rather than a real one, and this part is exactly the thing it claims not to be.
A reader who reaches Part Eighteen and finds no such chapter is entitled to conclude that, and to discount this part accordingly.
The reason for putting that in writing, in this chapter, rather than leaving it as an intention, is the whole method of this series.
Part Two of this series made four forward promises and named the parts that owed them. Part Seven paid two, Part Eight paid one, Part Nine paid the last.
Part Twelve promised the ending of the village-council story and Part Fourteen delivered it. Part Thirteen named the 1894 land acquisition Act and Part Sixteen recorded its replacement.
A promise recorded in a numbered chapter is a promise somebody can check. An intention stated in an author’s note is not.
This is the seventeenth part of nineteen. There are two left in which to be caught.
10.6 — What would settle it
India has no public archive of political speech.
Parliamentary debates are recorded and published. Everything else — rallies, press conferences, party sessions, interviews — exists in the holdings of broadcasters, in the notebooks of reporters, and in whatever somebody happened to upload.
The state broadcaster filmed the speech of 19 November 1984. That footage reached the public thirty-one years later, released by a private lawyer.
A searchable public archive of what people in office said, with date, venue and original language, would end four-fifths of the arguments in Indian political life at a stroke. It requires no new law, creates no offence, and restricts nobody. The material largely already exists in institutional hands.
Part Five ended by naming a calculation nobody has done. Parts Fourteen, Fifteen and Sixteen each named another. This is the fifth, and it is the only one that is not arithmetic — it is a filing job.
One closing observation, and it is the one this part exists to make.
Sixteen parts of this series have ended by naming something the state was holding back: a report from 1963, nine days of file notes from 1992, ninety pages from 2009, an appeal pending since 2018. This part ends by naming something nobody is holding back at all.
The words were spoken in public. Thousands of people heard them. Cameras were running. And the reason a country argues for forty years about what its Prime Minister said to a crowd on the Boat Club lawns is not that anybody hid it. It is that nobody wrote it down.
Around forty statements went in and seven came out. Four in five failed — not because they were disproved, but because no date, no venue or no reliable wording could be established.
The exact English wording of the most famous sentence in this part does not exist, because it was said in Hindi.
And nothing here was ever hidden. India has no public archive of political speech, which is why the footage of 19 November 1984 reached the public in 2015, from a lawyer.
Sources & further reading — Part 17
The State Ledger
Part Five committed every subsequent part to carrying this table. This part is thematic and has no geography of its own, so the ledger records where each statement was made and which part of this series covers what surrounded it.
| Place | What was said there | Covered in |
|---|---|---|
| Parliament, New Delhi | Aksai Chin, 5 Dec 1961. The apology of 11 Aug 2005. Tyagi’s reply. | Parts Six and Eleven |
| Boat Club lawns, New Delhi | 19 Nov 1984. | Parts Ten and Eleven |
| National Development Council, New Delhi | 9 Dec 2006. | Part Fifteen (the Sachar Committee reported that year) |
| Press conference, New Delhi | 7 Jan 2011. | Part Sixteen |
| Talkatora Stadium, New Delhi | 17 Jan 2014. | Part Sixteen (the 2014 result) |
| Gujarat | 7 Dec 2017, during the assembly campaign. | Part Eighteen |
| Punjab · Delhi | Not where anything was said, but what the 1984 and 2019 statements were about. | Parts Ten and Eleven |
Ten of the eleven statements were made within a few kilometres of each other in New Delhi. That is not a coincidence and it is not a finding about Congress: it is where the platforms are.
Timeline
| Date | What was said, and where |
|---|---|
| 5 Dec 1961 | Nehru, in Parliament: not a blade of grass grows there. Mahavir Tyagi replies from the Congress benches. |
| Late 1957 | The Xinjiang–Tibet highway across Aksai Chin is completed. Part Six. |
| Early 1958 | The Foreign Secretary tells Nehru there is little doubt the road crosses Indian-claimed territory. Part Six. |
| 20 Oct – 21 Nov 1962 | The Chinese attack and the unilateral ceasefire. Part Six. |
| After Nov 1962 | Nehru to Zhou Enlai: India will never accept a line created by aggression. |
| 31 Oct – 4 Nov 1984 | Indira Gandhi assassinated; the killings that followed. Parts Ten and Eleven. |
| 19 Nov 1984 | Rajiv Gandhi at the Boat Club lawns. The crowd applauds. |
| 11 Aug 2005 | Manmohan Singh apologises in the Rajya Sabha during the Nanavati debate. |
| 16 Jan 2006 | The ₹717 crore package for the victims of 1984. Part Eleven. |
| 9 Dec 2006 | The National Development Council speech. “They must have the first claim on resources.” |
| 10 Dec 2006 | The Prime Minister’s Office issues its clarification. |
| Nov 2010 | The CAG report on spectrum. Part Sixteen. |
| 7 Jan 2011 | Kapil Sibal’s press conference. Zero loss — and the procedural concession. |
| 15 Feb 2011 | The CBI Director tells the Public Accounts Committee that zero loss is wrong. |
| 2 Feb 2012 | The Supreme Court cancels 122 licences on the procedure. Part Sixteen. |
| 24 Sep 2013 | The Cabinet approves the ordinance reversing Lily Thomas. |
| 27 Sep 2013 | Rahul Gandhi at the Press Club of India: complete nonsense, torn up and thrown away. The ordinance is withdrawn. He does not tear it. |
| Jan 2014 | Manmohan Singh’s press conference: not a weak Prime Minister; history will be kinder. |
| 17 Jan 2014 | Mani Shankar Aiyar at the AICC plenary, Talkatora Stadium. |
| 26 May 2014 | Narendra Modi sworn in as Prime Minister. |
| Nov 2015 | Footage of the 19 November 1984 speech released publicly by H. S. Phoolka. |
| 7 Dec 2017 | Neech kism ka aadmi. Suspension the same day. |
| 18 Aug 2018 | The suspension is revoked. |
| 9 May 2019 | Sam Pitroda: hua to hua. Condemned by the party leadership; an apology follows. |
| May 2019 | Aiyar publishes an article asking whether he had not been prophetic. |
| 21 May 2022 | The Leader of the Congress in the Lok Sabha posts the big-tree quotation as a tribute, then deletes it. |
| Feb 2023 | The Congress chief spokesman defends “Rahul Gandhi’s act of tearing a copy of the ordinance” — an act that did not happen. |
| 2024 | Aiyar writes that he never used the word chaiwala. |
| 26 Dec 2024 | Manmohan Singh dies. The January 2014 sentence circulates again, this time approvingly. |
Words Used in This Part
| AICC | All India Congress Committee, the party’s national body. Its plenary sessions are large public conferences. |
| Aksai Chin | A high desert plateau of about 38,000 sq km between Ladakh and Xinjiang, administered by China since 1962. Part Six. |
| Expunged | Words removed from the parliamentary record by order of the Chair. They were said; they are not in the transcript. |
| Hua to hua | Hindi: roughly, “if it happened, it happened.” |
| National Development Council | A body of the Prime Minister and the Chief Ministers which approved the Five Year Plans. The 52nd meeting was on 9 December 2006. |
| Neech | Hindi: low, base, vile. Also carries a caste register in the phrase neech jaati, low caste. |
| Operative sentence | The one sentence of a speech that does the damage and gets repeated. Chapter One. |
| Presumptive loss | The gap between what a government received and what an auditor calculates it could have received under a method it did not use. Part Sixteen, Chapter Three. |
| Public Accounts Committee | The parliamentary committee that examines the Comptroller and Auditor General’s reports. Chaired by convention by a member of the opposition. |
| Rajya Sabha | The upper house of the Indian Parliament, elected by state legislatures. |
| Thodi | Hindi: a little, slightly. The word dropped from the most common English version of the 1984 sentence. Chapter Four. |
| Verbatim record | A word-for-word transcript. Parliament keeps one; rallies and press conferences do not. |
A Word on Sources
Everything in this part was checked against a source rather than recalled, and the rule was stricter here than anywhere else in the series: date, venue and operative wording, or the statement does not appear.
The strongest sources here are the parliamentary record, for 5 December 1961 and 11 August 2005; the official release from the Prime Minister’s Office of 10 December 2006, which reproduces the surrounding paragraph of the National Development Council speech; and video footage, for 19 November 1984 and 9 May 2019. The correspondence with Zhou Enlai exists as published diplomatic documents.
Where a statement was made in Hindi, the operative words are given in Hindi and the competing English versions are set out rather than resolved. This applies to 19 November 1984, to 7 December 2017 and to 9 May 2019. This part does not have an English “exact wording” for the 1984 sentence and does not pretend to.
Two things are contested and flagged in the text rather than smoothed away: which House heard the Aksai Chin remark, and whether the word chaiwala was used in January 2014. On the second this book accepts the speaker’s own later account.
Speeches are not reproduced. What appears here is the operative sentence in each case, with the surrounding argument summarised in this book’s own words. A political speech is somebody’s work and quoting it whole is neither necessary to the argument nor right, and the sentence in dispute is in every case short.
A note from the author
I did not want to write this part and I said so before starting it, at the end of Part Sixteen, where I recorded that I was looking forward to material whose only question was whether the words were spoken.
That turned out to be exactly wrong. This has been the hardest part of the seventeen to keep honest, and the difficulty was not verification. It was that the material is so much more enjoyable to write than anything else in this series.
A sentence needs no context, no grading and no counterfactual. It can be dropped on a page and it does the work by itself. After sixteen parts of qualifying every claim, the pull of that was much stronger than I expected, and Chapter Eight exists because I noticed myself enjoying the writing and went looking for why.
What I found is in that chapter and I think it is the most useful thing here: words and acts do not correlate. The government that produced the worst sentence in this part appointed a commission of inquiry. The government that produced the best one did nothing for four decades. If I had written this part without checking that, I would have produced something far more satisfying and completely misleading.
The place where I have most obviously been generous is Chapter Two. Every instinct of a prosecutor says to keep the 2006 sentence, which is devastating in its popular form and which I could have quoted accurately while saying nothing about the paragraph. I have instead spent the opening chapter of substance dismantling the best material I had. Part One committed me to that and I would not have done it otherwise.
And a declaration, because Part Ten made one and the rule is that it goes where the writing could pull. Chapter Four is about a sentence spoken after the killing of Sikhs, and I am Punjabi and have written on that material before. The chapter concedes the timing defence in full — the speech did not and could not have caused those deaths — and it says so twice, in the argument box and in the closing box. That concession is the part I checked most carefully, because it is the part I least wanted to make.
Lovepreet Singh
Part Seventeen of The Congress Record
The Living Archive · misterlove.in
Part Eighteen is The Comparison, 2014–2026: measured on the same ruler — where this government did the thing Congress could have done and did not, where it continued earlier policy under a new name, and where it has not moved. The second series auditing that government on this scale is still promised.