1The Test
Part Nine ended with a claim: that somebody had finally built a movement which would not die with its leader. This part is where that claim is tested to destruction, twice, and holds.
- n October 1708 Guru Gobind Singh died at Nanded, two thousand kilometres from the Punjab, having ended personal Guruship for ever.
IBy every precedent in the previous nine parts of this book, that should have been the end of it. The Nandas,
Menander’s kingdom, Harsha’s empire, Yashodharman’s alliance, Jasrat’s Punjab, Hemu’s army — all of
them died with one man. This community had just lost its Guru, its town, its founder’s four sons, and the war.
Fourteen months later it took Sirhind.
What follows is the strangest forty years in this book. A follower sent north with a letter raises an army out of
nothing, destroys a Mughal provincial capital, sets up a government, strikes his own coinage, abolishes the
landlords — and is captured, paraded through Delhi and executed within six years.
And then the movement does not end again. It goes into the forests and the deserts, is hunted for thirty years with
a bounty on every head, loses thousands of people in a single campaign in 1746, and emerges in 1748 organised into
a standing confederate army that will rule the Punjab within twenty years.
Four things to hold on to
- This is a peasant movement as much as a religious one. Banda’s land settlement is the most radical act of government in this entire book, and it explains why he could raise armies faster than the empire could destroy them.
- The persecutors are provincial, and they are not all Muslim. The men who ran the hunt were the governors of Lahore — and the campaign of 1746, the worst of them, was driven by a Hindu Khatri official avenging his brother. This part will name people rather than communities.
- The Mughal empire collapses in the middle of it. Nadir Shah sacks Delhi in 1739 and takes the treasury with him. After that there is no imperial power in northern India, only provincial governors and whoever comes down the road from Kabul — and everything in Parts Eleven and Twelve follows.
- And for the first time this book can triangulate. We have hostile Persian chronicles, Sikh tradition, and an English eyewitness — the East India Company’s envoys in Delhi in 1716, who watched the procession and wrote home about it.
Same commitments as Part Nine, and they matter more here because the violence is worse on both sides.
The sources are read against their authors. Khafi Khan and the other Persian chroniclers of this period detest
Banda and record his campaigns as atrocity; Sikh tradition records them as justice. Both are describing a real war in
which real towns were sacked. This part does not launder either side.
The executions are stated, not described. The methods used at Lahore and Delhi in these decades are recorded in
detail in several traditions. The facts are given here; the details are not, for the same reason as in Part Nine.
And the labels are used honestly. CERTAIN for contemporary documents — including the coins, which
survive, and the Company letter. LIKELY for well-attested events reported later. STORY where a tradition is
doing the work.
2Samana, 1709 — The First Blow LIKELY
A man arrived in the Punjab with five companions and a letter, and within a year he had stormed a wealthy imperial town.
Banda Singh Bahadur — with an army that had not existed twelve months WON BY
earlier
- November 1709.
- Samana, in the country between the Ghaggar and the Sutlej, in the Punjab plain.
- Banda Singh Bahadur, sent north from Nanded, against the imperial administration of the Sirhind province.
- Persian chroniclers hostile to him, chiefly Khafi Khan and the Ibratnama; and Sikh tradition.
Who he was
He was born in the Jammu hills around 1670, became a wandering ascetic, and by 1708 was living at Nanded in the
Deccan as the head of a small establishment. He had no connection with the Punjab wars and no military record.
Guru Gobind Singh met him at Nanded in 1708. What passed between them is recorded in tradition rather than in
any document. What is not disputed is what he left with: a new name, a letter of authority — a hukamnama
— addressed to the Sikhs of the Punjab, a small escort of five experienced Sikhs, and instructions.
He went north. He had no army, no money, no fortress and no standing. What he had was a piece of paper
from the Guru and the fact that the community was, at that moment, extremely angry.
What happened
1. He raised men in the Malwa country south of the Sutlej, working out of villages in the Sonepat and
- Kharkhauda region. The hukamnama was circulated and the response was immediate.
- The first recruits were mostly peasants. The Persian sources are contemptuous about this — they describe the followers as cultivators and men of low occupation. The contempt is the most useful thing in the account, because it tells you exactly where the manpower came from. 3. He took some small places first to arm and pay the force.
- Then, in November 1709, he stormed Samana. It was a wealthy town, an imperial administrative centre, and it was taken and thoroughly sacked.
- Then Ghuram, Thaska, Shahabad, Kapuri and Sadhaura in quick succession, working north-west across the province.
- And by the spring of 1710 he was standing in front of Sirhind.
Samana was not chosen at random, and the Sikh tradition is explicit about why.
It was the home town of the men who had carried out the executions. The executioner of Guru Tegh Bahadur at
Delhi in 1675 came from Samana; so, by the tradition, did the men who put the two children to death at Sirhind in
December 1704. Sadhaura, taken shortly afterwards, was the town where the official responsible for the death of Pir
Buddhu Shah — the Muslim divine who had fought for Guru Gobind Singh at Bhangani and lost two sons — had his
seat.
The campaign was a reprisal and it was conducted as one. The Persian sources describe the sack of Samana in
terms of general massacre. Sikh tradition describes retribution against the guilty. Both accounts are describing a
town being stormed by an army raised out of grievance, and towns stormed in the eighteenth century did not
distinguish carefully.
This book is not going to pretend otherwise. The rage was earned. What it did to Samana was still what an army does to
a town.
- The authority was already established. The hukamnama worked because there was a community, an organisation and a habit of obedience to a written order from the Guru — all of it built in Parts Eight and Nine.
- The grievance was five years old and universal. Anandpur, the Sarsa, Chamkaur and Sirhind had happened to everybody in that country, not to a court.
- The Khalsa had already removed the barrier to recruitment. Caste did not disqualify anybody. His army came from the whole population, and no other army in this book could do that.
- And the empire was elsewhere. Aurangzeb had died in 1707; his sons were fighting over the throne; the field army was in the Deccan.
- Garrison troops, not a field army. Provincial faujdars had enough men to police a district and not enough to fight a campaign.
- The towns were rich and unwalled in any serious sense. Prosperity without defence is the oldest invitation in this book.
- The countryside was against them. A revenue administration is only as strong as the cultivators’ willingness to keep paying it.
- And nobody believed it was serious until Sirhind.
Think about the mechanism here, because it is genuinely without precedent in this book.
A man nobody in the Punjab had heard of arrived from the Deccan with five companions. He had no lineage
anybody cared about, no lands, no fortress, no treasury and no reputation as a soldier. He was not a Guru
and did not claim to be — the Guruship had been vested in the book at Nanded, and everybody knew it.
What he carried was a letter.
And on the strength of that letter, in a province under imperial administration with a functioning revenue
system and garrisons in every town, thousands of men left their fields and came to him.
That only works because of what was built before he arrived. Part Eight created a script, a book, a
network of local congregations and a habit of receiving written instructions from a central
authority. Part Nine created an order that anybody could join and that bound its members visibly
and permanently. Banda did not build any of it. He inherited a working institution and used it.
Compare it with every other rising in these ten parts. The Yaudheyas needed generations of tribal
organisation. Jasrat Khokhar needed to be a chief’s son in his own clan’s country. Every rebellion against
Delhi in Parts Six and Seven needed a noble with troops already on his payroll.
Here the qualification for command was a piece of paper, and the qualification for enlistment was
willingness.
The empire never understood this, which is why for the next forty years it kept killing individuals and
being surprised that it had not worked.
3Chappar Chiri and Sirhind, May 1710 LIKELY
Five and a half years after the winter of 1704, the town came down.
When: 12 May 1710 the battle; the town taken two days later.
Where: Chappar Chiri, a village on the plain a few miles from Sirhind.
Who: Banda’s army against Wazir Khan, faujdar of Sirhind — the official responsible for December 1704.
The evidence: Persian chronicles and Sikh tradition, which agree on the outcome and on very little else.
What happened
1. Wazir Khan came out to fight rather than stand a siege, with the provincial troops and artillery and a body
- of allied contingents.
- The armies met at Chappar Chiri. The Sikh force was larger in numbers and much worse equipped — the accounts describe men with whatever weapons they had brought from their villages.
- The Sikh left was broken by artillery and the line came close to collapsing.
- Then the centre held and pushed. The tradition describes a general assault delivered against the guns.
- Wazir Khan was killed on the field.
- The army broke and the town was entered two days later. Sirhind was taken and destroyed. The Persian accounts describe general slaughter and the burning of the town; the Sikh accounts describe the same events as retribution.
- The treasury of the province went with it, and it was very large.
Sirhind was one of the wealthiest towns in northern India. It sat on the Grand Trunk Road at the point where the
Punjab meets the road to Delhi — the same crossroads that made it the place where Sher Shah sent his message to
Humayun in Part Eight, and where Bairam Khan refounded the Mughal empire in 1555.
It had gardens, sarais, mosques, tombs and a mint. It was the administrative capital of a rich province.
After May 1710 it was a ruin, and it was left as one. There was a deliberate refusal to rebuild, and the destruction of
Sirhind entered Punjabi memory as a completed act rather than an incident.
And it is worth saying clearly what that means, because this book has spent nine parts describing sacked towns. This
one was sacked by Punjabis, on Punjabi ground, against a Punjabi administration. It is not an invasion. It is the
first time in this entire history that the destruction on this soil is entirely home-grown — and the people doing it had
reasons that anybody reading Part Nine will understand and that do not make the burning of a town anything other
than what it was.
- Wazir Khan came out. Sirhind’s walls and guns were worth more than its field army, and he gave that away.
- Numbers and motivation. A badly armed force that will keep coming forward beats a well-armed one that will not.
- The commander was killed. For the sixth time in this book, a provincial army dissolved when the man holding it together went down — and this time the army with the institutional command structure was the peasant one.
- And the countryside supplied him. Food, intelligence, remounts and recruits, all locally, none of it available to the other side.
- There was no imperial army within reach. Bahadur Shah was in the Deccan dealing with the aftermath of his own succession war.
- Provincial forces are for policing. They are not designed to fight a popular rising and never have been.
- The revenue apparatus collapsed with the capital. Kill the treasury and the district officers stop being paid.
- And nobody local would fight for it. That is the whole of it, and the next war explains why.
Set the two Decembers against each other, because the distance between them is the point.
December 1704. A town in the hills has been starved out. A river comes up in the dark and cuts a family in
half. Two boys of seventeen and fourteen die outside a mud wall at Chamkaur. Two more, aged about nine
and six, are put to death at Sirhind on the orders of the faujdar, along with their grandmother. The man
who has lost all of them is moving through the Punjab in disguise, hidden by horse traders, with nothing.
May 1710. An army raised from the villages of that same country is standing on the field at Chappar Chiri.
The faujdar is dead on it. Two days later the town is burning.
Five and a half years.
Nothing in the previous nine parts of this book turns round that fast, and it is worth being precise about
why it could. It was not revenge alone; revenge is common in these pages and almost never
accomplishes anything. The Shahis wanted revenge and were extinct. Prithviraj’s kinsmen wanted
revenge and got nothing. Wanting it has never once been sufficient.
What made the difference is that this time there was an organisation still standing when the leader
was gone — congregations that could be written to, an order anybody could join, a shared identity that
made a man from one village recognise a man from another as bound to him.
The empire had spent a century removing individuals: a Guru in 1606, a Guru in 1675, four sons and a
mother in 1704. Every one of those removals worked perfectly and none of them worked at all.
4The State at Lohgarh, 1710–1711 CERTAIN
Not a battle. For about eighteen months there was a government in the Punjab that struck its own coins, dated its own era, and abolished the landlords — and its coins survive.
Punjab has an independent government for the first time since the WHAT HAPPENED
Shahis
- 1710–1711, and in fragments for several years after.
- The Sirhind province, from the Sutlej towards the Yamuna, governed from Lohgarh in the Siwaliks.
- Banda Singh Bahadur and the officers he appointed.
- The coins and the seal, which survive, plus hostile Persian description of the land settlement.
What was actually done
1. A capital. He fortified Lohgarh in the Siwalik foothills — the hills again, for the reasons Parts Eight and Nine
- set out — and governed from there.
- A coinage. Coins were struck bearing a Persian couplet crediting the sword, the cooking-pot, victory and ready patronage to Guru Nanak and Guru Gobind Singh. The name on the coin is not his own. The reverse names the mint in the formulae of imperial coinage.
- A seal. Official orders went out under a seal carrying the same couplet — again, not in Banda’s name.
- A calendar. A new era was dated from the taking of Sirhind.
- Officers. Governors and administrators were appointed to the districts, several of them men who had come with him from Nanded.
- And the land settlement. This is the one that matters most and is least discussed. The zamindari system was abolished in the territory he held, and the cultivators were made proprietors of the land they worked.
For five thousand years and nine parts, every state in this history has run on the same arrangement. The cultivator
works. An intermediary — zamindar, jagirdar, chaudhri, whatever the word is that century — collects from
him and passes a share upward. The state protects the intermediary because the intermediary delivers the
revenue. That is the machinery of Ashoka, of the Guptas, of the Sultanate, of Sher Shah and of the Mughals alike.
Banda dissolved it. In the districts he controlled, the men who held the land as intermediaries lost it, and the men who
ploughed it were declared its owners.
The Persian sources record this with genuine horror, and their horror is the best evidence that it happened. They
describe men of no standing being set over their betters and the natural order overturned. What they are describing, in
the language available to them, is a peasant revolution.
It also explains the military problem the empire could not solve. An army of cultivators who have just been given
their fields does not go home when it is beaten. It goes back to the fields, and comes out again next season, and
there is no aristocracy left in that country to raise a counter-force. This is why the fighting in the Sirhind province went
on for years after Banda himself was gone.
- It was popular in the most literal sense. The land settlement bought loyalty that no amount of money could have.
- The authority was borrowed and therefore unassailable. Ruling in the Gurus’ name, not his own, made the government hard to delegitimise from inside.
- The base was in the hills. Lohgarh could not be rushed.
- And the empire was busy. Bahadur Shah had a Deccan war, a Rajput problem and his own brothers.
- It made every landholder in northern India an enemy. Not only Muslim officials — Hindu zamindars and hill rajas too. The coalition that eventually formed against Banda was broader than religion.
- Territory means you can be found. A movement in the forests cannot be besieged. A government at Lohgarh can.
- It divided the community. Banda’s innovations — his own practices, his authority, the honours paid to him — were disputed by Sikhs who held that nothing could be added after 1708. That split hardened and lasted.
- And it was eighteen months old when the empire came.
Pick up one of Banda’s coins — they survive, in museums and collections, and they are the hardest
evidence in this entire part.
A coin is the loudest statement a government can make. This book has been reading them since Part Three:
the Yaudheya copper saying victory to the nation; the Indo-Greek silver with a king’s face wearing an
elephant’s scalp; Sher Shah’s rupiya; Kanishka’s gold with the Buddha on it. What a ruler puts on his
money is what he wants everybody to know.
Banda ruled a province, commanded an army, appointed governors and levied revenue. He had a mint.
And he did not put his own name on the coin.
The couplet credits everything — the sword, the pot that feeds people, victory, protection — to Guru Nanak
and Guru Gobind Singh. The seal on his official orders says the same. In a century when every ruler in Asia
announced himself in metal, the man who had taken Sirhind struck coins in the names of two men who
were dead.
You can read that as piety, and it was. You can also read it as the most sophisticated political act in
this part. It said that the government belonged to the community and not to him — which made it very
difficult for anyone to accuse him of setting himself up, and which meant that when he was captured and
killed six years later, the authority he had exercised did not die with him, because it had never been
his.
Part Nine described a Guru taking initiation from five men he had just created, so that the order would
outlive him. This is the same idea, in silver.
And it worked. Banda was executed in 1716. The Punjab was ruled by the Khalsa within fifty years.
5Lohgarh, December 1710 — The Empire Answers LIKELY
The emperor came north in person with the imperial army, issued an order for extermination, took the capital — and the man he came for walked out of it.
When: December 1710; the imperial order issued the same month.
Where: Lohgarh, in the Siwalik foothills near Sadhaura.
Who: The emperor Bahadur Shah I with the imperial field army, against Banda Singh Bahadur.
The evidence: Mughal records, including the firman itself, and the Persian chroniclers.
What happened
1. The emperor abandoned the Deccan and came north. That is the measure of how seriously it was taken: a
- Mughal emperor turning the imperial army around and marching it fifteen hundred kilometres.
- A firman was issued in December 1710 ordering the provincial officers to hunt down and kill the followers of Nanak wherever they were found. This is a document, not a tradition. It is the beginning of the thirty years described in War Seven.
- The imperial army invested Lohgarh. Sadhaura fell; the outworks were carried; the fort was surrounded in force.
- And Banda escaped. The accounts describe a night breakout through the encircling army. One tradition has a follower dressed in his clothes remaining behind to be taken in his place; the Persian sources simply record, with irritation, that he was not among the captured.
- He went into the hills — into the country north of the Sutlej, in the Jammu direction — and the imperial army could not follow him into it in winter.
- Bahadur Shah died in February 1712, and the empire fell into another succession war. Within a year Banda was back in the plains, retaking towns.
The December 1710 order is worth pausing on because it changes the character of everything after it.
Up to this point the conflict had a shape this book has seen many times: a state putting down a rebellion in a province.
The firman turned it into something else. It made membership of a religious community, by itself, a capital matter
— and because the Khalsa’s articles of faith are worn and visible, it made that membership impossible to conceal.
Part Nine noted this as the cost of the 1699 settlement: uncut hair and a steel bracelet are an asset in a battle
and a death sentence in a persecution. Here is the persecution.
The practical consequence over the following decades is described in War Eight. A great many people cut their hair and
stopped being identifiable. A great many did not, and went into the forests. And the ones who did not are the ones
who are still there in 1748 with an army.
It is also worth noting what the order could not do. It required provincial officers to find people, and provincial
officers in the Punjab were about to lose the ability to find anybody at all, because the empire that paid them
was three decades from collapse.
- It brought the field army. Nothing in the Punjab could stand against a full Mughal army with siege guns, and nothing did.
- The emperor came himself. Which meant the provincial governors and the hill rajas all cooperated, for once.
- A fixed capital can be attacked. The oldest trade in this book: territory buys legitimacy and costs you the ability to disappear.
- They took a fort and not a movement. The army was dispersed, not destroyed, and the districts stayed hostile.
- The imperial army could not stay. A field army in the Siwaliks in winter is a wasting asset, and the Deccan war was still running.
- The emperor died fourteen months later, and his sons fought over the succession for the next two years.
- And the land settlement had already happened. Retaking a province does not un-give the fields.
Here is the fact that puts this whole part in proportion, and it comes from the empire’s own behaviour
rather than from anything anybody claimed.
In 1710 the Mughal empire was fighting a war in the Deccan that had already consumed twenty-five years,
the last part of Aurangzeb’s life and an incalculable amount of money. It was the single strategic obsession
of the dynasty.
And Bahadur Shah left it, and marched the imperial army fifteen hundred kilometres north,
because of what was happening in one province of the Punjab.
Consider who he was going to fight. A wandering ascetic from the Jammu hills, in the country for two years,
commanding an army of cultivators with village weapons, holding a hill fort in the Siwaliks. No dynasty. No
foreign backer. No cavalry tradition. No siege train. On paper, a provincial disturbance.
The emperor of India came in person.
Governments are honest in their logistics if nowhere else. You can discount every claim in every
chronicle and still read the seriousness of a threat off the map, by looking at what the state moved and how
far. In this case it moved everything, across the width of the subcontinent, in the middle of a different war.
And having done that, and having taken the fort, and having issued an order for the extermination of a
religious community — the empire had achieved, by the spring of 1712, precisely nothing. Banda was
in the plains again. The province was in arms again. And Bahadur Shah was dead.
6Gurdas Nangal, 1715 — Eight Months in a Farmhouse LIKELY
The empire finally caught him by doing the only thing that has ever worked in this book: it stopped attacking and sat down.
When: Besieged from about April 1715; surrendered December 1715.
Where: Gurdas Nangal, near Gurdaspur — the fortified house of a landholder named Duni Chand.
Who: Abdus Samad Khan, governor of Lahore, against Banda Singh Bahadur and a few hundred followers.
The evidence: Persian chroniclers, who describe the conditions inside in detail because they were told by survivors.
What happened
1. The new emperor, Farrukhsiyar, made it a priority. A competent and determined governor, Abdus Samad
- Khan, was appointed to Lahore with orders to finish it, and given the resources to do it.
- Banda was pushed out of the plains and, in the spring of 1715, was cornered at Gurdas Nangal — an ordinary fortified farmhouse, not a fortress, with a mud wall and a ditch.
- Abdus Samad Khan did not assault it. He built a line of circumvallation around it — a wall of his own — and sat down.
- The siege lasted about eight months. The Persian accounts of conditions inside are among the grimmest passages in this book: the food gone, then the animals, then the bark of trees and grass, and men dying of hunger where they stood.
- Some got out. Small parties broke through at various points and escaped into the country.
- In December 1715 the survivors were taken. Banda was captured alive with several hundred others.
Look at how sieges resolve across these ten parts, because Gurdas Nangal is the clearest possible statement of the rule.
Assaults on determined defenders usually fail. Mahmud failed twice at Lohkot in Part Five. The hill rajas failed for
four years at Anandpur in Part Nine. Timur’s grandson lost his cavalry outside Multan in Part Seven.
Blockades almost always succeed. Multan surrendered when its water was cut in 712. Akbar starved Sikandar Shah
Suri out of Mankot in six months in Part Eight. Anandpur fell to a blockade after the assaults had failed. And here, an
ordinary farmhouse held out for eight months against the governor of Lahore and then surrendered because there was
nothing left to eat.
The reason is arithmetic, not courage. A besieger eats from a province. A besieged man eats from a store. The store
always runs out first, and the only variable is how long the defenders are prepared to make it take.
Which is why the note in Part Five is worth repeating here in its harshest form: a fortress without a field army
behind it is not a defence. It is a place to die slowly. Nobody was coming to Gurdas Nangal, and everybody inside
knew it by about the third month.
- He was patient. Eight months of doing nothing, in a career where impatience had destroyed his predecessors.
- He built a wall round the wall. Circumvallation prevents both escape and relief, and it converts a siege into an accounting exercise.
- He had continuity of command. One governor, one objective, from 1713 to 1715, at a moment when Delhi changed emperors three times.
- And Banda was cornered outside his own country. Gurdaspur is not the Malwa; the local population was not the one that had risen in 1709.
- He had become findable. Six years of holding territory, striking coins and appointing governors made him an address rather than a rumour.
- The community was divided. The dispute over his innovations and his authority had hardened, and not everybody came when he called.
- He took shelter instead of dispersing. The one mistake. The method that had worked in 1710 was to break up and vanish; in 1715 he went into a building.
- And there was no relief because there was no second army. Everything was in that farmhouse.
The Persian accounts of Gurdas Nangal are written by men who wanted Banda destroyed, and they record
the eight months in detail because the detail was the point: this was what came of rebellion.
They describe the grain gone, the horses and oxen eaten, then the leaves and the bark stripped, then men
eating what could not be eaten and dying of it. They describe the survivors at the surrender as barely able
to stand.
And they record, without meaning to, the thing that mattered most. Over those eight months, in
small parties, at night, men got out through the lines and away into the country.
They were not deserting in any useful sense — there was nowhere to desert to, with a price on their heads
and the Punjab under a standing order for their extermination. They went out into a province where being
recognised meant death, and they survived in it.
Some of those men were still alive in the 1720s and 1730s. Some of them were still alive in 1748.
The empire counted the prisoners at Gurdas Nangal, executed them, and considered the account
closed. Khafi Khan’s tone at the end of it is one of a problem finally solved.
What it had actually done was take a movement that held territory — findable, besiegeable, defeatable —
and convert it back into a movement that held nothing at all. Which turned out to be very much harder
to kill, and is the whole of the next thirty years.
7Delhi, 1716 — The Procession CERTAIN
The empire staged the most elaborate public demonstration it could devise — and the two Englishmen watching wrote down the one thing it did not want recorded.
When: The procession in late February 1716; the executions through the spring; Banda executed in June 1716.
Where: Delhi — the road into the city, and then the place of execution near Mehrauli.
Who: Banda Singh Bahadur and roughly seven hundred prisoners.
The evidence: Persian chronicles, and a letter written from Delhi in March 1716 by John Surman and Edward Stephenson of the East India Company.
What was staged
1. The prisoners were brought from the Punjab to Delhi and paraded into the city in a procession designed
- to be seen by the whole population.
- Banda was carried in an iron cage on an elephant, dressed in an embroidered robe and a mock headdress, with a guard behind him.
- Ahead of the cage went cartloads of heads on poles — the accounts give a figure in the thousands — and behind it the surviving prisoners, roped, on camels.
- The executions began at once and continued daily for about a week at the kotwali in the centre of the city, in public.
- Banda was kept alive for three months and then put to death in June 1716, near Mehrauli. His young son was killed in front of him first. This book will not go further than that sentence.
The East India Company had an embassy in Delhi at exactly this moment, negotiating trading rights with Farrukhsiyar.
Two of its members, John Surman and Edward Stephenson, watched the procession and reported it in a letter home
in March 1716.
Their account is dry, commercial and completely uninterested in the theology of any of it. They describe the cage, the
elephant, the heads, the numbers. They are foreigners with no stake in the quarrel, writing to a company that wanted
news of the court’s mood, not a moral verdict.
And in the middle of the report they record something that neither the Persian chroniclers nor the Sikh tradition could
have made them say. They note how the prisoners went to their deaths — with a composure that struck the
writers as remarkable — and record that not one of them, to the last, renounced his religion to save himself.
Think about what kind of evidence that is. Sikh tradition says the same thing, and Sikh tradition would say it
either way. Persian chroniclers hostile to Banda have every reason to describe rebels dying badly. The only genuinely
neutral witnesses in Delhi in the spring of 1716 were two English merchants keeping their principals informed — and
they wrote it down because they found it worth mentioning.
That is the whole reason this book keeps insisting on asking who is writing. Nine parts of triangulating between
partisan sources, and here, at last, is a witness with nothing to gain.
- Make it public. The whole event was theatre, staged in the busiest streets of the capital, for an audience.
- Make it total. Seven hundred prisoners and cartloads of heads is not a punishment; it is an attempt to demonstrate that a thing has ceased to exist.
- Make it humiliating. The cage, the robe, the mock crown — the point was to show a man who had struck coins in the Gurus’ names being displayed as a curiosity.
- And it was a new emperor’s demonstration. Farrukhsiyar had taken the throne in a civil war and needed a victory the capital could see.
- Public death makes witnesses. Delhi watched seven hundred people refuse to save themselves, and Delhi talked about it.
- It confirmed the story the community told about itself. Everything the Khalsa had been founded on in 1699 was demonstrated in the street, at the empire’s expense and on the empire’s initiative.
- It killed the leadership and not the institution. For the third time in this book, and by now it is not an excuse: 1606, 1675, 1716.
- And the empire had three decades left. Farrukhsiyar was strangled by his own ministers in 1719. Nadir Shah was in Delhi in twenty-three years.
There is a particular kind of historical evidence that this book has been hunting since Part One, and it has
almost never been available.
Everything in these ten parts has come from someone with a reason. Greek writers glorifying Alexander.
Court poets cataloguing their patron’s conquests. Persian secretaries counting the dead for the man who
paid them. Devotional literature written to edify. Victory inscriptions. Chronicles commissioned by the
winner.
Ask who is writing, and the answer has always been: somebody who needed you to believe something.
In March 1716, two employees of a trading company sat in Delhi writing a routine business letter. They were
there to secure customs concessions. They had no view about the Khalsa, no knowledge of Anandpur or
Sirhind, no stake in Mughal legitimacy and no idea that anybody would ever read their letter for this. They
mentioned the procession because it was the news of the week.
And what struck them enough to put in a commercial dispatch was not the cage or the heads or the
emperor’s power. It was the behaviour of the prisoners.
That is why the letter matters more than a hundred pages of chronicle. It is not a better source because
Englishmen wrote it — it is a better source because nothing about their situation gave them a reason
to say it.
The empire spent a fortune staging a spectacle in the streets of its capital to prove that a movement had
been extinguished.
The only genuinely disinterested people who saw it came away describing the people in the ropes.
8The Price on the Head, 1716–1745 LIKELY
For a generation, being recognisable in the Punjab was worth money to whoever turned you in.
When: 1716–1745, with the policy hardest under Zakariya Khan after 1726.
Where: The whole Punjab, administered from Lahore.
Who: Abdus Samad Khan (governor 1713–1726) and his son Zakariya Khan (1726–1745).
The evidence: Persian administrative record, Sikh tradition, and the physical sites in Lahore.
What the policy was
1. The 1710 firman remained in force. Provincial officers were required to find and kill members of the
- community wherever they were.
- A bounty was placed on heads. Payment for a captured Sikh, more for a killed one — a standing rate, administered through the district officers, which turned the hunt into an economy.
- Executions were public and centralised. A platform at the horse market in Lahore — the Nakhas — became the standing place of execution. The site is now known as Shahidganj, the place of martyrs.
- Access to Amritsar was cut. The Harmandir Sahib was occupied and closed to the community at intervals, and gatherings there were treated as an offence in themselves.
- And the identity itself was the offence. Because the Khalsa’s articles of faith are worn, a man could be identified on the road. The policy did not require evidence of anything; it required a look.
The executions of these decades are recorded in detail in the Sikh tradition, and the details are commemorated
annually. This book states the facts and stops there, for the same reason it stopped in Part Nine.
Bhai Mani Singh was a scholar, a custodian of the Harmandir Sahib and one of the most learned men in the
community — he had worked on the scripture. He obtained permission to hold a gathering at Amritsar on payment of a
fee; troops were then sent to the gathering; the fee could not be paid. He was executed at Lahore, by
dismemberment, in about 1738. The date is given variously in the sources.
Bhai Taru Singh was a young farmer from a village near Amritsar who had been feeding fugitives. He was arrested,
offered his life if he would cut his hair, and refused. He was executed at Lahore in about 1745.
Massa Rangar was the local official put in charge of the occupied Harmandir Sahib, who used it for entertainments. In
1740 two Sikhs, Mehtab Singh and Sukha Singh, rode into Amritsar disguised as revenue collectors, killed him inside the
building, and rode out.
These three names carry more weight in Punjabi memory than most of the battles in this book, and the
reason is worth naming. A battle is a contest. What these decades produced instead was a long, administered, ordinary
business of hunting people — and the memory of it is not about heroism in the field but about individuals who were
given a way out and did not take it.
- It was administrative, not military. No campaign to lose. District officers, a rate of payment, and a network of informers.
- The identity was visible. The single most effective feature of the policy, and it came free.
- It paid. Bounties recruit collaborators from the general population without any loyalty being involved.
- And Zakariya Khan was competent. Nineteen years in the same post, with a consistent policy — more continuity than the empire in Delhi managed in the same period.
- It could not reach the forests and the deserts. Which is War Eight.
- It gave nobody a way back. A policy that offers only death produces people with nothing to lose, and there is no worse enemy.
- Public executions of people who refuse to recant are recruitment, as Delhi had already demonstrated in 1716 and learned nothing from.
- And the state ran out of time. Zakariya Khan died in 1745. Nadir Shah had already gutted the empire in 1739. Within three years there was a Sikh army with a formal command structure.
In about 1739, according to the tradition, two men decided that the official position — that the Sikhs no
longer existed in the Punjab — needed correcting.
Bota Singh and Garja Singh came out of hiding, walked to a sarai on the Grand Trunk Road near Tarn
Taran, and set up a toll post. Two men, with sticks, on Sher Shah Suri’s highway, collecting a fixed charge
from carts and pack animals in the name of the Khalsa.
Travellers paid, because two armed men on a road are two armed men on a road, and went on their way.
Nothing happened. Nobody reported it. And that was the problem — because the entire point was to be
noticed.
So Bota Singh wrote a letter to Zakariya Khan, the governor of Lahore, announcing what he was doing. The
verse attributed to him is doggerel and is famous precisely because it is: a note stating the rate for a cart
and the rate for a donkey, and telling the governor’s household, by name, that this is Bota Singh
speaking.
A detachment was sent. The two of them were killed on the road.
The story is STORY — it comes from the Sikh tradition and no Persian source records it, which is what
you would expect of an incident the administration would rather not have on file.
But the logic of it is the most revealing thing in this whole part. Two men, hunted, with a price on
their heads, in a province where being seen was fatal, deliberately made themselves visible on the busiest
road in India and then wrote to the government to make sure it knew.
They were not trying to hold a road or raise money. They were trying to be on the record. The official
line was that the community had been exterminated in 1716, and two men in a ditch on the Grand Trunk
Road decided that the line needed a public contradiction, and paid for it with their lives.
Ten parts of this book have been about who gets written down. This is the only entry in it where somebody
died specifically to be.
9The Forest Years — How They Survived LIKELY
Not a battle. The military method that came out of these decades is the reason the Punjab is Sikh-ruled by 1765, and it was invented by people with nothing.
- Roughly 1716–1745.
- The Lakhi Jungle of the Malwa; the desert country towards Bikaner; the Siwaliks; and across the Sutlej.
- Bands — jathas — of a few dozen to a few hundred, under their own leaders.
- Sikh tradition, and the frustration visible in the Persian administrative record.
The method
1. No fixed ground. Banda had held territory and been besieged in it twice. The lesson was learned completely:
- after 1716 the community held nothing that could be attacked.
- Everything on horseback. A band that can move eighty kilometres in a night cannot be caught by a district officer with foot police. Horses became the central economic fact of the movement — bought, taken, bred, and valued above everything else.
- Refuge terrain, chosen deliberately. The Lakhi Jungle — a great belt of scrub and forest in the Malwa — the desert margins, the hill country, and the far side of the Sutlej, where Lahore’s writ ran thin. Part Eight’s rule, applied as policy: the hills and the waste preserve you.
- Living off the country, and then formalising it. Bands took what they needed, and over time this hardened into a system — rakhi, protection: a village paid a fixed share of its harvest to a band, and in return that band kept other bands and the revenue officials off it. This is a government forming out of a protection racket, and it is how most governments in history have started.
- Assembling twice a year. The scattered bands gathered at Amritsar at Vaisakhi and Diwali where they could — the Sarbat Khalsa, the whole community in assembly, which took binding decisions by resolution.
A guerrilla movement lives or dies on one question, and it is not tactical. Will the countryside feed it and lie to the
government about it?
In the Punjab between 1716 and 1745, in large parts of the Majha and the Malwa, the answer was yes — and it is worth
being clear about why, because piety is only part of it.
These were the same villages. The men in the Lakhi Jungle were the sons and brothers of the men still ploughing.
The community was not a foreign body hiding in the countryside; it was the countryside, with a portion of it living
rough.
The alternative was worse. Revenue officials, faujdars and bounty hunters were not popular, and a village that paid
rakhi to a band got a real service in return.
And Banda’s land settlement had happened. Its formal reversal did not undo the memory of it. In a province where
a movement had once given the fields to the men who worked them, that movement had a standing claim on those men
that no governor could match.
The Persian sources notice the effect without ever quite naming the cause. They complain, repeatedly, that the
rebels vanish; that information is unobtainable; that detachments arrive to find nothing. That is not a
description of clever fugitives. It is a description of an entire province declining to cooperate.
- Nothing to besiege. The single most important adaptation, learned at Gurdas Nangal.
- Mobility over everything. Cavalry bands with no baggage, no artillery and no families in the column.
- Distributed leadership. Dozens of jathas under their own leaders. Kill one and you have removed one.
- And an assembly that could still make decisions. The Sarbat Khalsa meant the movement could act as a whole without ever concentrating as a whole.
- Everything except survival. No towns, no revenue, no schools, no building — thirty years of a community’s ordinary life simply not happening.
- The visible identity was a death sentence. Many cut their hair and went back to their villages. The split between those who did and those who did not is a real and painful part of this period.
- Discipline was hard to hold. Bands living off the country are bands living off the country, and not all of them behaved well.
- And when they were caught in the open, in 1746, it was a massacre. That is War Eleven.
Stop and notice what this method is, because you have read it before in this book. Repeatedly.
Part Five, 1008: the Khokhars of the Salt Range charge Mahmud’s line at Waihind and are never subdued.
Part Six, 1206: Muhammad of Ghor crushes them on the Jhelum and is dead in his tent within a fortnight.
Part Seven, 1420–1442: Jasrat Khokhar holds the northern Punjab for twenty years by withdrawing into
the Salt Range every time a Delhi army appears and coming back when it leaves. Part Eight, 1541: Sher
Shah Suri builds the largest fortress of his century to hold down the same people, and it does not work. Part
Nine, 1586 and 1672: the frontier tribes destroy two Mughal armies in defiles and are eventually paid
rather than beaten.
Every one of them wins by the same means: no capital, no field army, terrain the enemy cannot use, a
population that will not inform, and no need to win — only the need to still be there next season.
What is different after 1716 is that this method is being used, for the first time, by a movement with a
written scripture, a formal initiation, a standing assembly and a doctrine.
The Khokhars had none of those and so their victories were always local and always temporary — four
hundred years of being unremovable and never once turning it into a state.
Here, the same military method is attached to an institution. The bands in the Lakhi Jungle could
assemble at Amritsar and pass a binding resolution. They could distribute territory among themselves and
be held to it. They could hand a title to a man by common consent, which is War Nine.
Guerrilla survival plus an institution equals a government. That combination appears in the Punjab in
these decades for the first time in five thousand years, and it takes Lahore within twenty of them.
10The Nawabship, 1733 — A Title Given From Below LIKELY
The governor of Lahore offered a peace, a jagir and a title. The way the community decided who should take it is the most interesting thing in this part.
When: The offer in 1733; the organisation formalised soon afterwards; the peace broken within a couple of years.
Where: Amritsar, in assembly.
Who: Zakariya Khan making the offer; the Sarbat Khalsa deciding what to do with it.
The evidence: Sikh tradition, and the fact that the organisation it produced is documented from 1748 onwards.
What happened
1. The policy of extermination had failed and Zakariya Khan knew it. Seventeen years of bounties and executions had produced a province he could not govern outside the towns.
2. So he changed approach. He offered a grant of land — a jagir around Dipalpur, Kanganpur and Jhabal — and
- the title of Nawab, to whoever the community named as its head.
- The Sarbat Khalsa met at Amritsar to consider it. The offer was contentious: accepting a title from the government that had been hunting them was not an obvious step.
- They took it, and chose the recipient. The tradition records that the title was conferred on Kapur Singh, a man of no particular rank, who was at that moment performing the ordinary service of fanning the assembly.
- And then they reorganised. The scattered bands were formalised into two divisions — the Buddha Dal, the army of the elders, and the Taruna Dal, the army of the young, itself subdivided into five bodies each with its own leader, banner and drum.
- The peace broke down within about two years and the hunt resumed. The organisation did not break down.
Compare this with every other transfer of authority in these ten parts.
In Part Five a sultan’s slave takes the throne because he holds the army. In Part Six the Punjab frontier command
marches on Delhi. In Part Seven a governor left in charge by Timur makes himself the government. In Part Eight a
thirteen-year-old is crowned on a brick platform because he is his father’s son. Every single one is either
inheritance or force.
Here, a title offered by a hostile government to “the head of the Sikhs” was received by an assembly, debated, and
conferred by that assembly on a man it selected. The tradition’s insistence that he was fanning the gathering at the
time is doing exactly the work you would expect — it says that the office was given, not taken, and that the qualification
was service.
Whether the detail is literal is STORY. What the story is asserting is a constitutional principle, and the
principle is visible in what happened next.
Fifteen years later the same assembly would create the Dal Khalsa and appoint its commander. Fifty years later the
misls would be dividing conquered territory among themselves by resolution at Amritsar. That is a political system,
and this is where it is first visible in operation.
- The policy had failed and was expensive. Seventeen years of bounties with the province still ungovernable outside the walls.
- A jagir converts an enemy into a stakeholder — the standard Mughal method, used successfully on Rajputs and Afghans for two centuries.
- He needed the countryside quiet. Revenue was not coming in, and revenue was what a governor was judged on.
- It conceded that the community existed. Seventeen years of official policy had been that it did not. An offer of terms is a recognition.
- It gave them a base and a breathing space, which they used to reorganise into a formal military structure.
- It legitimised the assembly. By addressing an offer to “the Sikhs” he made the Sarbat Khalsa the body that answered.
- And when the peace collapsed, they were better organised than before it started.
The image at the centre of this story has been repeated for nearly three hundred years, and it is worth
taking seriously as a piece of political thinking rather than as a piece of piety.
An assembly is sitting at Amritsar. On the table is an offer from the government of Lahore: land, revenue
and the title of Nawab, for whoever will be recognised as the head of the community. It is the first formal
recognition the Khalsa has ever been offered, and it comes from the man who has spent seventeen years
killing them.
They accept, and they have to decide who.
Every other assembly in this book would have chosen the most successful commander, or the man with the
most horsemen, or somebody’s son.
The tradition says they gave it to the man who was fanning them while they talked.
Kapur Singh went on to be one of the most effective leaders the community produced — he held the Dal
together through the worst of the persecution, and he was the man who handed the command to Jassa
Singh Ahluwalia in 1748. This was not a joke appointment, and the assembly clearly knew what it was
doing.
But the point of the story is the framing, and the framing is deliberate. In a culture where the langar
had established a century earlier that everybody eats in the same row, the highest office available
was handed to a man performing a menial service, in front of everyone, as an argument about
what authority is for.
Fifteen years later that same assembly divided the Punjab into commands and appointed a commander-in-
chief. Within twenty it was a government.
Nine parts of this book have described power passing by birth, by murder, by inheritance or by
conquest. This is the first time in five thousand years that it is handed to somebody by a show of
hands.
11Nadir Shah, 1739 — The Road From Kabul CERTAIN
In four months a Persian army came down the Punjab road, destroyed the Mughal empire as a functioning power, and went home with the treasury.
When: Lahore in January 1739; Karnal on 24 February; Delhi in March; back across the Indus by summer.
Where: Down the Grand Trunk Road through the whole Punjab, and back up it.
Who: Nadir Shah of Persia against the emperor Muhammad Shah.
The evidence: Abundant — Persian, Mughal and European accounts, and the objects, which are in Iran.
What happened
1. He came through the passes in the winter of 1738–39, having already taken Kandahar and Kabul, and
- moved down the Punjab road.
- Lahore did not fight. Zakariya Khan — the same governor who had spent seventeen years hunting the Sikhs — submitted, paid, and was left in place.
- Karnal, 24 February 1739. The Mughal army was destroyed in a few hours on the plain north of Delhi. It was very much larger and it was beaten by better artillery, better cavalry tactics and a commander who knew what he was doing.
- Delhi was occupied in March. A rumour that Nadir was dead led to attacks on his troops in the streets; the response was a general massacre lasting several hours, ordered in retaliation and stopped when the emperor pleaded for it to end. The death toll is given in tens of thousands.
- The city was stripped. The treasury, the jewels, the Peacock Throne, the Koh-i-Noor, and a figure in the hundreds of millions of rupees, along with craftsmen, elephants and horses.
- And then he went home, back up the same road, in the summer of 1739.
- On the way back the Sikhs raided him. Bands came out of the Lakhi Jungle and the river country and took what they could off the baggage columns and, by the tradition, freed prisoners being carried north.
The Mughal empire did not end in 1739 — it had an emperor in Delhi until 1857. What ended in 1739 was the idea
that it could protect anything.
The consequences run through the rest of this book and they are worth listing plainly.
The treasury was gone. An empire that cannot pay cannot command, and the provinces stopped pretending within a
decade.
The frontier was gone. Nadir took the territory west of the Indus formally, and after his assassination in 1747 it
passed to Ahmad Shah Abdali, who founded the Afghan kingdom out of it. Whoever holds Kabul holds the door
into the Punjab — this book’s most reliable rule since Part Five — and from 1747 that door belongs to a new,
aggressive power that will come through it nine times.
And the road had been demonstrated again. Everybody in Asia now knew that an army from Kabul could reach
Delhi in six weeks and that nothing in between would stop it.
For the Punjab specifically, the arithmetic was simple and terrible. It sat between a capital that could no longer
defend it and a new military power that had every reason to keep coming through it. Everything in Parts
Eleven and Twelve is that sentence being paid out.
- A professional army against a ceremonial one. The Mughal army of 1739 was enormous, magnificent and had not fought a serious war in a generation.
- Better artillery and mounted firepower. His men used the zamburak — light swivel guns on camels — and his cavalry could shoot.
- Command. One commander who had risen by war, against a court where the emperor’s generals were rivals.
- And Karnal was fought in a few hours. There was no second Mughal army.
- Lahore submitted immediately. The governor had troops and used them to hunt his own population rather than to hold the road.
- Seventeen years of internal war had emptied the province of anybody willing to fight for the government.
- The passes were long gone. Kandahar and Kabul had fallen before he ever reached India.
- And nobody in Delhi believed it until Karnal.
There is a story told about Nadir Shah’s return march through the Punjab in the summer of 1739, and like
all the best stories in this book it may not have happened and is exactly right anyway.
His columns, loaded with the wealth of Delhi, were being raided. Small mounted parties came out of the
scrub at night, took what they could off the baggage, and were gone before anything could be organised
against them. It was not war; it was attrition by people who could not be caught.
The tradition has him asking Zakariya Khan who these people were.
The governor’s reply, in the version that has come down, is that they are a set of religious mendicants who
live in the saddle, whose homes are their horses, and who sleep in the open.
And Nadir Shah is said to have looked at that and observed that a people who live like that will one day
take this country.
Treat it as STORY. It is recorded in later Persian and Sikh sources and it has the shape of a prophecy
assembled after the event, which is how prophecies usually get their accuracy.
But hold the actual scene, because that part is not in doubt. In 1739 the man who had just destroyed
the Mughal empire in a single afternoon was riding home through the Punjab, and the only people in that
province giving him any trouble at all were fugitives with a price on their heads, operating out of a jungle,
from a movement the government had officially declared extinct twenty-three years earlier.
Zakariya Khan’s own army did not fire a shot at Nadir Shah. The people he had spent his career
hunting did.
12The Chhota Ghallughara, 1746 LIKELY
The one time in these decades they were caught in the open, and it cost about seven thousand lives in a month.
When: May and June 1746.
Where: The Kahnuwan marshes near Gurdaspur, and the hills beyond towards Basohli.
Who: Lakhpat Rai, diwan of Lahore, with provincial troops, against the Sikh bands caught in the open.
The evidence: Persian record and Sikh tradition, which agree on the shape and roughly on the scale.
What happened
1. It began with a private quarrel. Jaspat Rai, the faujdar of Eminabad, was killed in a skirmish with a Sikh
- band. His brother Lakhpat Rai was the diwan — the chief revenue officer — at Lahore, and he took it as a personal matter.
- He mounted a systematic campaign, with the backing of the governor, and pursued it with a thoroughness the province had not seen since 1716. Sikhs in Lahore itself were rounded up and killed.
- The bands were driven north-east out of their usual country, towards the Ravi and the hills.
- They were caught at Kahnuwan, in marsh and scrub country near Gurdaspur, with the pursuit behind them and the hill states in front refusing passage.
- The fighting was a running massacre. They tried to break through into the hills; the hill rajas would not let them in; they were driven back into the marshes and cut up.
- The figures usually given are about seven thousand killed and three thousand taken. The prisoners were brought to Lahore and executed.
- It is remembered as the Chhota Ghallughara — the lesser holocaust — a name that only makes sense because of what happens in 1762, in Part Eleven.
This book said in its introduction that it would name individuals, and 1746 is where that matters most.
Lakhpat Rai was a Hindu Khatri. So was his brother Jaspat Rai, whose death started it. They were career officials of
the Mughal provincial administration, and the campaign of 1746 was driven by a revenue officer avenging his brother,
with the resources of the state behind him.
The hill rajas who closed the passes and would not let the fugitives through were Hindu rulers — the same
hill states that had besieged Anandpur in Part Nine, doing the same thing for the same reasons.
And a few years later, another Hindu Khatri official at Lahore, Diwan Kaura Mal, took the opposite view entirely and
became known for his sympathy towards the community, to the point of being called by an honorific in Sikh sources.
This is not an argument that religion did not matter in the eighteenth-century Punjab. It obviously did,
enormously. It is an argument that if you try to read these decades as one community against another, you cannot
explain a single one of the four people just named.
Officials acted as officials, with careers, family obligations and grudges. Hill rajas acted as hill rajas, protecting their
borders. That is what the record shows.
- Personal motivation at the top. Lakhpat Rai pursued it with an energy that a salaried officer executing a policy never brings.
- They were caught in the open. Away from the Lakhi Jungle, away from the Malwa, in country that was neither refuge nor home.
- The hills were closed to them. The one escape route that had always worked was shut by the rajas.
- And the marsh trapped them. Cavalry cannot move in flooded scrub, which removed the movement’s single greatest advantage.
- Not everybody was there. Bands in the Malwa and across the Sutlej were untouched.
- The organisation survived. The Dal structure of 1733 and the assembly at Amritsar were institutions, not people, and institutions are not in marshes.
- The state was two years from collapse. Zakariya Khan had died in 1745, his sons fought over the succession, and Abdali crossed the Indus in 1748.
- And within two years the survivors formalised into a standing army. Which is War Thirteen.
The events of 1746 are called the Chhota Ghallughara. Ghallughara is a Punjabi word meaning something
close to holocaust or catastrophe — a killing on a scale that requires its own noun.
And the word chhota means lesser.
Think about what has to be true for a language to name an event that way. Seven thousand people killed in
a month, three thousand more executed afterwards, out of a community that numbered — nobody knows
exactly, but not in the millions.
It is called the lesser one because sixteen years later there was a bigger one. In February 1762, at Kup
near Malerkotla, Ahmad Shah Abdali caught the Dal Khalsa moving its families and killed a number that
the sources put much higher. That is the Vadda Ghallughara, the greater, and it is War Six of Part Eleven.
A community that has to distinguish between its catastrophes by size is telling you something about the
century it was having.
And here is the thing that makes the eighteenth-century Punjab genuinely unlike the rest of this
book. After 1746 they did not scatter for a generation and rebuild slowly. Within two years they held a
general assembly at Amritsar and organised themselves into eleven standing commands with a
commander-in-chief. Within nineteen years of Kahnuwan they held Lahore.
Ten parts of this history have described defeats that ended things — the Shahis, the Chauhans, the Lodis,
Hemu, Banda. This is the first community in these pages that takes losses on this scale and comes
out of it better organised than it went in.
13Manupur, March 1748 — Abdali Comes LIKELY
The Mughals won this one, which makes it the last time. What came through the passes in 1748 came back eight more times.
When: 11 March 1748.
Where: Manupur, near Sirhind — the same crossroads as ever.
Who: Ahmad Shah Abdali against a Mughal army under the wazir Qamaruddin and his son Muin-ul-Mulk.
The evidence: Mughal and Afghan accounts; well documented on both sides.
Who Abdali was
He had been a commander in Nadir Shah’s army. When Nadir was assassinated by his own officers in 1747, the
Afghan contingent elected Ahmad Khan of the Abdali tribe as their leader, and he took the territories east of Persia
— Kandahar, Kabul, and the country up to the Indus — and made them a kingdom.
He was, from 1747, the ruler of everything on the far side of the passes. Part Five established the rule and
every part since has confirmed it: whoever holds Kabul holds the door into the Punjab. He held it, and he had an
army that needed paying, and the richest country within reach was on the other side of the mountains.
What happened
1. He crossed the Indus in January 1748 and took Lahore without much difficulty. The governor’s family was
- in the middle of a succession dispute and the province was not defended.
- He moved down the road towards Delhi.
- A Mughal army came out to meet him under the wazir Qamaruddin, with his son Muin-ul-Mulk and the imperial prince.
- Qamaruddin was killed before the battle — by a stray shot, at prayer, in his tent — and the command passed to his son. 5. At Manupur on 11 March the Afghan army was beaten. The accounts describe hard fighting and an explosion in the Afghan camp that decided it.
- Abdali withdrew to Kabul, and Muin-ul-Mulk — known ever after as Mir Mannu — was made governor of Lahore as the reward.
- And he came back the following year, and the year after that. Nine invasions in twenty years.
Manupur is the last battle the Mughal empire wins in the Punjab, and it is worth understanding why the victory was
worthless.
The empire had no money. Nadir Shah had taken the treasury nine years earlier. A province could be defended once
by an army that turned up; it could not be garrisoned permanently.
The winner became the problem. Mir Mannu, rewarded with Lahore, found himself governing a province with an
Afghan army on one side and the Dal Khalsa on the other, and within two years he was paying Abdali tribute out of the
revenues of the Punjab. The province was ceded formally soon after.
And the invader had come to stay in the sense that mattered. Abdali did not need to hold the Punjab; he needed to
be able to reach it. He came back in 1749, 1752, 1756, and repeatedly after that.
The pattern that Part Five described at Peshawar in 1001 is now running again, with a different dynasty and identical
mechanics. A military power based at Kabul, a rich and undefended plain three hundred miles away, and a
central government that cannot reach either. Everything in Part Eleven follows.
- Abdali’s first invasion was small. A new king consolidating a new kingdom, testing the road.
- They met him near their own base. Sirhind is close to Delhi and far from Kabul.
- Muin-ul-Mulk fought well after taking over on the death of his father, which made his reputation.
- And they had luck. The explosion in the Afghan camp is in every account.
- No treasury. The single fact underneath everything after 1739.
- No frontier. The passes and the Indus country belonged to the invader before he set out.
- No province. The Punjab had been fought over internally for thirty years and would not fight for Delhi.
- And a third party was now in the field. Three weeks after Manupur, the Khalsa organised itself into a standing army — and neither Delhi nor Kabul had factored it in.
Manupur is a few miles from Sirhind, and it is worth stopping to count what has happened at that
crossroads in this book.
1540: Sher Shah Suri reaches Sirhind and sends the message that drives Humayun out of India. 1555:
Bairam Khan wins the battle of Sirhind and the Mughal empire is refounded. 1704: two children are put to
death at Sirhind by the faujdar. 1710: the town is stormed and destroyed by an army raised in the villages
around it. 1748: the last Mughal victory in the Punjab is won a few miles away.
And it is not finished. 1761: Ahmad Shah Abdali and the Marathas will decide the fate of northern India at
Panipat, sixty miles down the same road. 1764: the Sikhs will take Sirhind and it will change hands for the
last time.
The reason is the reason it has always been. Sirhind sits where the Punjab road meets the road to
Delhi. Everything coming from the passes and everything going to the capital passes through that gap, and
has done since before anybody wrote it down.
Part Six made the same observation about Tarain, twenty miles away, where three battles were fought in
twenty-five years. Part Seven made it about Panipat, forty miles further on, where three more were fought
over two hundred and thirty-five.
There is a stretch of the Grand Trunk Road, perhaps a hundred and twenty kilometres long, on
which the government of northern India has been decided at least ten times in a thousand years.
It is ordinary wheat country with a motorway through it now, and there is nothing to look at.
14Vaisakhi 1748 — The Dal Khalsa LIKELY
Eighteen days after Manupur, the assembly at Amritsar turned sixty-five fighting bands into one army with a commander — and this book’s oldest failure was finally solved.
- Vaisakhi, 29 March 1748.
- Amritsar, at the Akal Takht.
- The Sarbat Khalsa, with Nawab Kapur Singh presiding and Jassa Singh Ahluwalia appointed to command.
- Sikh sources; the structure it created is documented in operation from this point onwards by everybody, including the British.
What was decided
1. The scattered bands were consolidated. The many jathas that had survived the persecution were
- reorganised into eleven divisions — later twelve — each under a named leader, each with its own banner and drum.
- These are the misls. The word means, roughly, an equal or a file — and the equality is the point. Each misl was autonomous, kept its own territory and revenue, and none was subordinate to another.
- A commander-in-chief was appointed for the whole force in the field: Jassa Singh Ahluwalia, who had been raised in Kapur Singh’s household.
- The whole was called the Dal Khalsa — the army of the Khalsa.
- A fort was built at Amritsar — Ram Rauni — giving the community, for the first time since Lohgarh, a defensible place of its own.
- And the assembly retained authority over all of it. The Sarbat Khalsa met at Amritsar at Vaisakhi and Diwali, passed binding resolutions — gurmatas — and allocated territory and tasks among the misls.
Count the coalitions that have failed in these ten parts, because the list is now very long and completely consistent.
Ten kings on the Ravi in Part One. The Malavas and Kshudrakas in Part Two, who combined and could not agree on
a commander. Nine republics submitting separately in Part Four. Yashodharman’s alliance, which dissolved on the day
it won. Anandapala’s confederacy in Part Five, destroyed in ten minutes when one elephant bolted. Prithviraj’s
coalition at Tarain in Part Six. The Khokhars, Langahs and Afghans of Part Seven, who never once made common cause.
The hill rajas of Part Nine, changing sides every three years.
Every single one failed for the same reason: several leaders of equal standing, no agreed command, and no
mechanism for making a decision that everybody would obey.
The Dal Khalsa is that problem solved, and the solution has three parts. The misls are formally equal, so nobody is
being subordinated to a rival. The command in the field is appointed by the assembly, not seized or inherited, so it
is legitimate. And the assembly itself is a standing body with a fixed calendar and a religious sanction, so its
resolutions bind.
It is not a perfect system and this book will not pretend it is — the misls fought each other repeatedly, and by the 1790s
the confederacy was tearing itself apart, which is how Ranjit Singh becomes possible in Part Twelve.
But between 1748 and 1767 it does what no Indian coalition in five thousand years has managed: it holds
together long enough to beat a superior invader, repeatedly, and then takes the country.
- It formalised what already existed. The bands, the leaders and the assembly were all thirty years old. This gave them a constitution rather than inventing one.
- Equality removed the reason to defect. No misl had to accept another’s overlordship.
- The command was temporary and functional. Jassa Singh commanded in the field; he did not rule anybody.
- And the timing was perfect. The empire had just spent its last victory, the Afghans had just been introduced, and the province was open.
- Equality also means nobody can compel. A misl that decided not to come to a muster could not easily be made to.
- Territory divided is territory to argue about, and by the 1780s they were arguing with cavalry.
- The assembly worked under pressure and decayed without it. Once the Afghans stopped coming, the gurmatas stopped binding.
- And it took a war to hold it together. Part Eleven is that war, and it is the worst one yet.
Put the two dates side by side, because the distance between them is the whole argument of this part.
June 1716. Delhi. Seven hundred prisoners are executed in public over several days at the kotwali. Cartloads
of heads have been paraded through the streets. The man who took Sirhind is put to death near Mehrauli
after three months in captivity. The imperial order for the extermination of the community has been in
force for five years and is being enforced. Two English merchants write home describing the spectacle, and
note that not one of the condemned took the offer of life. The state’s position is that the matter is
closed.
March 1748. Amritsar. An assembly of the same community meets at the Akal Takht, consolidates sixty-
five surviving bands into eleven standing divisions, appoints a commander-in-chief by common consent,
builds a fort, and allocates the districts of the Punjab among its commanders.
Thirty-two years. One generation.
In between: a firman, a bounty on every head, a standing place of execution at Lahore, the loss of Amritsar,
the executions of Mani Singh and Taru Singh, thirty years in the Lakhi Jungle, and seven thousand dead in
the marshes at Kahnuwan.
And they came out with a constitution.
Nine parts of this book described a province that things were done to — invaded, garrisoned, taxed, spent,
written about by other people in other languages. Part Nine gave it a book and an order. Part Ten is where
those two things are tested by a state that set out deliberately to end them, and do not end.
Within seventeen years of this assembly, the Dal Khalsa took Lahore and struck coins there.
Which means the Punjab was governed by Punjabis for the first time since the Hindu Shahis lost
Kabul in 870 — a gap of eight hundred and ninety-five years.
15The Whole Chapter on One Page
Five patterns to carry into Part Eleven
1. Killing the leader is now demonstrably useless. 1606, 1675, 1704, 1716 — four decapitations by three
- governments, every one of them successful and none of them effective. Part Nine claimed that an institution had been built which would outlive its leaders; Part Ten is the proof, delivered by the state’s own repeated attempts to disprove it.
- The coalition problem is solved. Ten parts of failed alliances — the Ravi, the Malavas, Anandapala’s confederacy, Tarain, the hill rajas — every one destroyed by having several leaders and no agreed command. The Dal Khalsa of 1748 fixes it with formal equality between misls, a commander appointed by assembly, and a standing body that can pass binding resolutions. Watch it hold for twenty years and then decay in Part Twelve.
- Territory is a liability until you can defend it. Banda held ground and was besieged out of it twice. After 1716 the movement held nothing and could not be caught. The moment it takes Lahore in Part Eleven, it becomes findable again — and that changes everything about how it has to fight.
- Whoever holds Kabul holds the door. Part Five’s rule, now running for the eighth time. Nadir Shah took the frontier in 1739; Abdali inherited it in 1747; and from 1748 there is an aggressive military power on the far side of the passes with a standing reason to come through them. Nine invasions in twenty years is what that sentence costs.
- And the empire is gone. After 1739 there is no power in Delhi capable of defending or governing the Punjab. What fills that vacuum — Afghans from the west, Marathas from the south, and the misls from inside — is the entire subject of the next two parts.
The Afghans and the Ghallugharas, 1748–1767. Nine invasions in twenty years. Mir Mannu’s governorship and the
persecution of the early 1750s; the Punjab ceded to Kabul; the Marathas take Lahore and reach Attock in 1758 —
the furthest north an Indian power has reached since the Mughals — and are annihilated at Panipat in January 1761,
the largest battle fought in India in a century. Then the worst day in Sikh history: the Vadda Ghallughara at Kup in
February 1762, and the destruction of the Harmandir Sahib. And then the turn — Sialkot, Sirhind in 1764, and in 1765
the Dal Khalsa takes Lahore and strikes coin there, in the names of the Gurus, on the same legend Banda used
fifty-five years earlier.
16Words Used in This Part
A written order or command letter issued under the authority of the Guru. The document Banda Singh Bahadur carried north from Nanded in 1708.
A Mughal district military commander. Wazir Khan of Sirhind held the post; so did Jaspat Rai at Eminabad.
The system of intermediary landholders who collected revenue from cultivators and passed a share to the state. Abolished by Banda in the territory he held.
An imperial order. The firman of December 1710 required provincial officers to hunt and kill the followers of Nanak wherever found.
A band — the basic unit of the movement during the persecution, from a few dozen to a few hundred, under its own leader.
The great belt of scrub and forest in the Malwa country south of the Sutlej, the principal refuge of the 1720s and 1730s.
“Protection.” A fixed share of the harvest paid by a village to a band, in return for being kept clear of other bands and of revenue officers. The beginning of Sikh territorial administration.
The general assembly of the whole community, meeting at Amritsar at Vaisakhi and Diwali, whose resolutions bound everybody.
A binding resolution passed by the Sarbat Khalsa.
The army of the elders and the army of the young — the two divisions into which the bands were organised in the 1730s.
The army of the Khalsa, formed at Amritsar on Vaisakhi 1748 out of the surviving bands.
One of the eleven — later twelve — autonomous divisions of the Dal Khalsa, each with its own leader, banner and territory, and none subordinate to another.
A Punjabi word for a catastrophe or holocaust. The Chhota — lesser — is 1746; the Vadda — greater — is 1762, in Part Eleven.
“The place of martyrs” — the site at Lahore, at the old horse market, used as the standing place of execution during the persecution.
The horse market at Lahore where the execution platform stood.
A light swivel gun mounted on a camel. Part of why Nadir Shah’s army destroyed a much larger Mughal one at Karnal in a few hours.
Ahmad Shah, elected leader of the Afghan contingent after Nadir Shah’s assassination in 1747, founder of the Afghan kingdom, and the central figure of Part Eleven.
Sources used for this part
- Khafi Khan, Muntakhab-ul-Lubab, and the Ibratnama of Mirza Muhammad Harisi — contemporary Persian accounts, both strongly hostile to Banda
- The letter of John Surman and Edward Stephenson, East India Company envoys at Delhi, March 1716 — the disinterested
- eyewitness account of the procession and executions
- Mughal administrative records, including the firman of December 1710
- The coins and seal of Banda Singh Bahadur, which survive
- Sikh sources of the eighteenth and nineteenth centuries — Prachin Panth Prakash and related works — for the persecution, the martyrdoms and the internal organisation, treated as tradition and labelled as such
- Persian and European accounts of Nadir Shah’s invasion of 1739
- Afghan and Mughal accounts of Ahmad Shah Abdali’s first invasion and the battle of Manupur
- Ganda Singh’s documentary work on Banda Singh Bahadur and on the eighteenth-century Punjab, and Hari Ram Gupta’s history
- of the Sikhs
- The sites — Chappar Chiri, Gurdas Nangal, Shahidganj at Lahore, and the Akal Takht at Amritsar
The open questions here include: what actually passed between Guru Gobind Singh and Banda at Nanded, and the exact
nature of the authority conferred; the scale of the killing at Samana and Sirhind, where hostile Persian sources and
Sikh tradition give irreconcilable accounts of the same events; the dates of the executions of Bhai Mani Singh, which the
sources place variously across several years; the casualty figure for the Chhota Ghallughara; and the extent to which the
1733 nawabship story reflects what was actually decided.
Three things about method.
First, this part does not launder Banda’s campaigns. Samana and Sirhind were stormed and sacked, with the
killing that implies, by an army raised out of a real and terrible grievance. The Persian sources exaggerate and the Sikh
sources justify, and underneath both there is a burned town. That has been said on the page rather than in a footnote.
Second, the persecution is described by naming officials rather than communities, because that is what the
record supports. Abdus Samad Khan, Zakariya Khan, Yahiya Khan and Lakhpat Rai were individuals with careers,
orders and grudges — and Lakhpat Rai, who ran the worst campaign of the period, was a Hindu Khatri avenging his
brother, while another Hindu Khatri official at Lahore a few years later was remembered by the community with
respect.
Third, this part states the executions and does not describe them, for the reasons given in Part Nine. The methods
used at Lahore and Delhi in these decades are recorded in the tradition and are commemorated annually. Nothing is
added here by rehearsing them.