Redrawing Punjab

1The language file

India reorganised itself on linguistic lines through the 1950s. Punjab was the one major linguistic region refused. It took fifteen years, tens of thousands of arrests, and a war to get what Andhra got in 1953.

What India did for everyone else

After independence, the demand for states drawn on language grew across India. The government resisted, then gave way:

• 1953 — Andhra State created for Telugu speakers, after Potti Sriramulu died on hunger strike.
• 1956 — the States Reorganisation Act redrew most of India on linguistic lines: Kerala, Karnataka, and a much-enlarged Andhra Pradesh.
• 1960 — Bombay State split into Maharashtra and Gujarat.

The principle was settled and applied nationally: a language community with a contiguous territory gets a state.

What India did not do for Punjab

The Akali Dal demanded a Punjabi Suba — a Punjabi-speaking state — from the early 1950s.

The States Reorganisation Commission of 1955 rejected it. Its reasoning had two parts:

1. That Punjabi and Hindi were not sufficiently distinct languages to justify a separation.
2. That the demand was, in substance, communal rather than linguistic.

The second reason was the real one, and everyone knew it. In Punjab, "Punjabi-speaking area" meant, in practice, "Sikh-majority area," because Punjabi in Gurmukhi script was identified with Sikhs. Creating a Punjabi Suba meant creating India's only state with a non-Hindu majority, less than a decade after a partition on religious lines. The government of India was not willing to do that.

And now the part that is hardest to write

In the 1951 and 1961 censuses, very large numbers of Punjabi Hindus — particularly in the towns — recorded their mother tongue as Hindi.

This was not random. Organised campaigns by Hindu organisations in Punjab — the Arya Samaj, the Jan Sangh, the Punjab Hindi Raksha Samiti — actively urged Punjabi Hindus to declare Hindi, explicitly in order to reduce the recorded number of Punjabi speakers and thereby defeat the Punjabi Suba demand.

It worked. The census produced a Punjab in which a large share of the population officially did not speak the language they actually spoke at home, and that data was then used to argue that no Punjabi-speaking state was justified.

Both sides of the census question

How it was finally won

The Akali campaign ran for fifteen years: fasts by Master Tara Singh, then leadership passing to Sant Fateh Singh, repeated morchas, and mass arrests. In the 1960–61 agitation alone, around 26,000 people courted arrest.

The concession came in 1966, and the timing tells you a great deal. It followed the 1965 war with Pakistan, in which Sikh soldiers had again served in numbers far beyond their share of the population and with conspicuous distinction. A parliamentary committee under Hukam Singh recommended reorganisation, and Indira Gandhi's government accepted it.

The Punjab Reorganisation Act, 1966 was passed on 18 September and came into effect on 1 November 1966.

Punjab got its state. Thirteen years after Andhra, and only after another war.

2The territory file

The 1966 Act gave Punjab a state and, in the same stroke, created almost every dispute of the next sixty years.

What the Act did

The old East Punjab was divided into four pieces:

The size arithmetic

Set the numbers out in sequence and the scale of the reduction becomes visible.

The boundary grievance

The Punjabi complaint about 1966 is that the linguistic principle was applied inconsistently, and that Punjabi-speaking areas were assigned to Haryana and Himachal.

The named areas most often cited include parts of the Kharar and Kandaghat regions, and villages in the Ambala and Sirsa belts.

The Punjabi case: the boundary was drawn using the contaminated census data described in Chapter 1. If a large share of Punjabi Hindus had recorded Hindi as their mother tongue, then any boundary drawn from that data would systematically under-count Punjabi-speaking territory and hand it to Haryana. The distortion was not incidental to the outcome; it produced the outcome.

The counter-case: census data is the only workable basis for a linguistic boundary, and no government can draw a line from what it suspects people would have said if they had answered differently. The Act also provided for later adjustment through commissions.

My assessment: the Punjabi objection here is logically sound and practically unanswerable — a boundary drawn from data that was deliberately corrupted will be wrong. But the corruption was not the Centre's doing, and there was no fair alternative method available in 1966. This is a real grievance whose author is largely local.

What Punjab lost in practical terms

• Its capital. Chandigarh became a Union Territory administered by the Centre — Chapter 3.
• Its High Court. Shared with Haryana, and still shared today.
• Control of Bhakra-Nangal. The great dam complex was placed under the Bhakra Beas Management Board, a central body — so Punjab's principal water infrastructure was no longer under Punjab's control.
• Control over the allocation of its own rivers, by way of Section 78 of the Act — Chapter 4.

3The capital file

Of the six files in this part, this is the only one where the verdict is not difficult. A capital was promised to Punjab in writing, with a deadline, by two Prime Ministers, fifteen years apart. Neither promise was kept.

How Chandigarh came to exist

Punjab lost Lahore in 1947. East Punjab needed a capital, and there was none — every significant city of the old province was on the other side of the line.

So one was built. Chandigarh was designed from scratch in the 1950s, with Le Corbusier as its principal planner, on land acquired from Punjabi-speaking villages in the Kharar area. It opened as the capital of East Punjab.

In 1966 the state that had built it was divided, and rather than assign the capital to one successor, the Act made Chandigarh a Union Territory and the joint capital of both — explicitly as a temporary arrangement.

It has now been temporary for sixty years.

The promises, in sequence

Why it kept failing

The mechanism of failure is worth understanding, because it is not simple bad faith.

Both settlements were structured as a swap: Punjab gets Chandigarh, Haryana gets compensating Hindi-speaking territory. The territory identified was the Abohar and Fazilka belt.

There were two problems with that.

First, Punjab would not give it up. Abohar and Fazilka are prime cotton-growing country, and no Punjab government could survive handing them over.

Second — and this is the decisive technical point — the swap was geographically impossible. The Mathew Commission found that Abohar and Fazilka tehsils are not contiguous with Haryana. There is Punjabi territory in between. You cannot transfer a district to a state it does not touch without also transferring a corridor, which nobody had agreed to.

The verdict, and Haryana's case

Haryana's argument, at full strength: Haryana has never had a capital of its own either. It has built sixty years of administration, courts and infrastructure in Chandigarh. It did not make or break the promise — the Centre did. And Haryana was offered compensating territory that Punjab then refused to hand over, so from Haryana's side the deal collapsed because Punjab would not perform its half.

That is a fair argument and it should be stated. But it does not change the finding.

4The water file

Part 1 predicted this from the geology. Here is the argument itself, explained as simply as I can manage — including the element of it that is genuinely indefensible and that almost nobody discusses.

The setup

From Part 7: the Indus Waters Treaty of 1960 gave India the Ravi, Beas and Sutlej. Those waters then had to be divided between Indian states.

1955: the allocation that started it

Before reorganisation, in 1955, the Centre allocated the available Ravi and Beas waters — around 15.85 million acre-feet (MAF) — as follows:

Read the first row again, because this is the part that gets lost in the noise about Haryana.

In 1955, Rajasthan — through which neither the Ravi nor the Beas flows — was allocated more of those rivers' water than Punjab, through which both flow.

This predates the 1966 reorganisation entirely. It is not about Haryana. And in my view it is the single strongest and least-discussed element of Punjab's entire water case.

1966 and 1976: Section 78

Section 78 of the Punjab Reorganisation Act gave the central government the power to allocate the waters between the successor states.

In 1976, acting under that section, the Centre allocated 3.5 MAF each to Punjab and Haryana, and 0.2 MAF to Delhi.

To deliver Haryana's share physically, a canal was needed to carry water from the Sutlej system across to the Yamuna. That is the Sutlej–Yamuna Link canal — 214 kilometres, of which the Haryana portion was built and the Punjab portion was not.

The two arguments, at full strength

The water dispute, both sides

5The economic file

The one bargain in this part that was kept. It made Punjab the richest state in India, it fed the country, and it planted every ecological problem that Part 13 will have to account for.

The deal

In the mid-1960s India was importing wheat under American food aid and facing serious famine risk. The response was the Green Revolution: high-yielding seed varieties, chemical fertiliser, assured irrigation, and — critically — a guaranteed purchase price.

Punjab was selected as the lead region for a specific reason set out in Part 6: it already had the irrigation infrastructure. Eighty years of canal building, plus the Bhakra-Nangal complex, meant Punjab could deliver assured water, which is what the new seeds required.

The bargain, stated plainly:

It worked, spectacularly

Punjab — around 1.5% of India's land area — came to supply an enormous share of the central grain pool, at times in the region of 60% of the wheat and 40% of the rice procured nationally. India moved from food imports to self-sufficiency. By the 1970s and 1980s Punjab had the highest per-capita income of any Indian state.

This is not a small achievement and it should not be grudged. A great deal of what India became after 1970 rests on what Punjabi farmers did in this period.

The invoice, which arrives in Part 13

Every element of the success carried a cost that was not visible at the time:

• Monoculture. Punjab's cropping collapsed into a wheat–paddy cycle, because those were the two crops with guaranteed procurement. Diversity disappeared.
• Paddy. Rice is not a natural Punjab crop. It is a wetland crop being grown in a semi-arid plain, and it consumes water on a scale the region does not have. Punjab grows rice it does not eat, using water it does not have, because that is what the procurement system pays for.
• Groundwater. Free or heavily subsidised electricity for tubewells, from the 1970s onward, made it rational for every farmer to pump. The water table began falling and has not stopped.
• Soil and chemicals. Fertiliser and pesticide dependence, declining soil organic matter, and the health consequences that Part 13 examines.
• Debt. Mechanisation required capital. Input costs rose faster than support prices. Farm debt became structural rather than occasional.
• Fragmenting holdings. Punjabi inheritance divides land among sons. Prosperous farms became several small ones, then several unviable ones.

The political consequence, which is counterintuitive

Here is a point I think is underrated, and it matters for Part 9.

The Green Revolution created a prosperous, confident, politically assertive rural class — Jat Sikh farmers with cash, tractors, education for their sons, and time. And that is precisely the class that provided the membership, the money and the organisational muscle of the Akali movement through the 1970s.

Anyone who explains the 1980s as a product of poverty or backwardness has the economics backwards. Punjab in 1980 was not a deprived region. It was a rich region with a list of grievances and the means to press them.

6The constitutional file

The most-cited Sikh grievance against the Indian Constitution. Its legal effect is much narrower than nationalist writing claims — and the political handling of it was still needlessly stupid.

What Article 25 actually says

Article 25 of the Constitution guarantees freedom of conscience and the right to profess, practise and propagate religion. Attached to it is Explanation II, which provides that in the relevant sub-clause, the reference to Hindus "shall be construed as including a reference to persons professing the Sikh, Jaina or Buddhist religion."

What it does and does not do

Let me be precise, because both sides overstate this.

What it does: the sub-clause it attaches to concerns throwing open Hindu religious institutions of a public character to all classes and sections. Its purpose was to ensure that reform legislation — particularly temple-entry laws abolishing caste exclusion — applied to Sikh, Jain and Buddhist institutions too. Read narrowly, it is an anti-caste provision.

What it does not do: it does not declare Sikhs to be Hindus for all purposes. Sikhs have distinct legal recognition elsewhere: the Anand Marriage Act 1909 (amended in 2012 to provide for separate registration of Sikh marriages), and recognition as a minority under the National Commission for Minorities Act 1992. Article 25 itself explicitly protects the wearing and carrying of kirpans as part of the Sikh religion.

So why does it matter so much?

Because constitutions are symbolic documents as well as legal ones, and being defined inside another religion in your country's founding text is not a small thing for a community whose entire modern history — Part 4, and Kahn Singh Nabha's Ham Hindu Nahin in 1898 — has been an assertion of distinctness.

And there is a fact from 1949 that is rarely mentioned and should be.

My assessment

On the law: the practical effect of Explanation II is limited, and Sikh nationalist writing routinely overstates it — it is not a declaration that Sikhism is a branch of Hinduism, and it has not been used as one.

On the politics: this clause has been cited as a grievance in every significant Akali document since 1950. Amending or clarifying it would have cost the Indian state essentially nothing — no revenue, no territory, no water, no precedent that threatens anyone. It is the cheapest concession on the entire list.

It has not been made in seventy-five years.

That is not a failure of law. It is a failure of political imagination, and of the most ordinary kind: a grievance that could have been closed for free was left open, and it kept being available to whoever wanted to use it.

7The Anandpur Sahib Resolution

The most quoted and least read document in this story. Before you can judge what anyone did with it, you have to know what it says — and the first surprising fact is that there is no single agreed text.

The context

By 1973 the Akali Dal was out of power in Punjab, internally factional, and needed a platform. Its working committee adopted a set of resolutions at Anandpur Sahib on 16–17 October 1973.

What it demanded

Grouping the demands honestly, they fall under five heads:

Notice how much of that list is simply the earlier chapters of this part. Chandigarh is Chapter 3. Water is Chapter 4. Article 25 is Chapter 6. The boundary is Chapter 2. The Anandpur Sahib Resolution is, in substantial part, a bill for unpaid promises.

The versions problem, which is the crux

Here is the thing that makes every argument about this document so slippery, and it is rarely stated plainly.

There is no single canonical text.

• The 1973 working-committee resolutions focused substantially on a special constitutional status for Sikhs and on Punjab's specific demands.
• The 1978 Ludhiana version, endorsed at the 18th All-India Akali Conference on 28–29 October 1978, widened the autonomy demand to all states, recasting Centre–State relations generally rather than seeking a special arrangement for Punjab alone.
• Multiple translations circulated, differing in emphasis and in some cases in substance, and the Akali Dal itself issued different formulations at different times.

This is not a technicality. It means that both readings — "federalism charter" and "secessionist charter" — can be argued from real text, and that people quoting it at each other are frequently quoting different documents.

The two readings, at full strength

What the Anandpur Sahib Resolution meant

My assessment

On the substance: read as a whole, and particularly in the 1978 version, this is a federalism document. Its central constitutional demand is a maximalist version of an argument that was mainstream in Indian politics at the time and is not, in itself, secessionist. It is not more extreme than what the Rajamannar Committee had already proposed in Tamil Nadu.

On the framing: and yet. The religious language, the invocation of the Khalsa, the claim to special status for a community rather than a state — these have no equivalent in the DMK's or the CPI(M)'s federalism arguments. That framing is what made the document usable in two opposite directions: by Congress as proof of secessionism, and later by militants as proof of an unfulfilled national demand. The Akali Dal wrote a document that could be read against itself, and then spent a decade being surprised that it was.

8The broken-bargain audit

Six files, one table. This is the analytical purpose of Part 8, and the answer is more mixed than either side of this argument wants it to be.

What the table shows

Three things, and none of them is the story either camp tells.

1. The grievances are real, but they are not of one kind. One file is an unambiguous broken promise. Two are genuinely contested with a serious opposing case. One is a symbolic injury rather than a breach. One was delivered late. One was kept. Anyone who describes Punjab 1947–1980 as a uniform record of betrayal is not describing this table. Anyone who describes it as a list of unreasonable demands is not describing it either.

2. Not every wound came from Delhi. The census episode of Chapter 1 is the clearest case of Punjabis damaging Punjabis, and it corrupted the boundary that produced the territorial grievance. Attributing that to the central government is factually wrong and, I think, politically convenient for everyone involved.

3. Punjab was rich while this was happening. The most assertive phase of Punjabi politics coincides exactly with Punjab being the wealthiest state in India. Deprivation is not the explanation.

And now the Zafarnama question

Part 4 gave me a test drawn from the tradition's own foundational sources rather than imported from outside. Guru Gobind Singh, in the Zafarnama: the sword is righteous when all other means have failed. Not when they are slow. Not when they are frustrating. When they have failed.

So: by 1980, had they?

Had all other means failed by 1980?

9The Emergency, and how Part 9 begins

Two facts about the second half of the 1970s that almost never appear in national accounts, and which change how the next part reads.

The Akali Dal and the Emergency

Indira Gandhi declared a state of Emergency on 25 June 1975. Civil liberties were suspended, the press was censored, and opposition politicians were jailed in large numbers.

The Shiromani Akali Dal launched a sustained campaign against it — the Save Democracy morcha — using the morcha method described in Part 6: successive batches of volunteers courting arrest, continuously, for as long as it took.

Estimates of the numbers arrested run to around 40,000 Akali volunteers over the period of the Emergency. It is generally reckoned the largest and most sustained organised resistance to the Emergency mounted by any single political party in India.

I want that stated because it is verifiable, it is rarely mentioned, and it materially affects how one reads the accusation that follows.

1977 to 1980

The Emergency ended in 1977. Congress was defeated nationally. In Punjab, an Akali Dal–Janata coalition formed a government under Parkash Singh Badal.

In 1980, Indira Gandhi returned to power at the Centre, and the Congress government dismissed the Akali government in Punjab, as it did several other opposition state governments.

And the accusation that opens Part 9

The Akali Dal held the SGPC, the gurdwaras and the rural Sikh vote. For Congress, breaking that hold was the central problem of Punjab politics.

The allegation — which is one of the most consequential claims in modern Indian political history — is that senior Congress figures, with Giani Zail Singh and Sanjay Gandhi named most often, deliberately cultivated a young, uncompromising, orthodox religious preacher who could challenge the Akali Dal's religious authority and split its vote.

That preacher was Jarnail Singh Bhindranwale.

Part 9 examines that allegation properly — what evidence exists, who has said what and when, what is documented and what is inference, and what it does and does not explain about everything that followed.

Timeline of Part 8

What happens in Part 9

Part 9 is Bhindranwale — Full Autopsy, and it is the part this whole work has been building toward.

I will establish who he actually was and where he came from; examine the evidence for the claim that Congress cultivated him, and separate what is documented from what is inference; go through his actual recorded statements rather than the versions that circulate on both sides, including the ones that are fabricated; cover the Nirankari clash of 1978, the murder of Lala Jagat Narain in 1981, the Dharam Yudh Morcha, the fortification of the Akal Takht, and Operation Blue Star in June 1984 — with casualty figures given as ranges and sources named.

And then I will apply the two tests I have built for this purpose, both drawn from Sikh tradition rather than imposed from outside: the Zafarnama's last-resort condition, established in Part 4 before I had met the man it would judge, and the broken-bargain test from Part 6, whose evidence base is the table in Chapter 8 of this part.

I do not intend to deliver a verdict of saint or monster. Both are available cheaply and neither survives the record.