The Sikh Empire

1Banda Singh Bahadur

In 1710 a man who had spent most of his life as a wandering ascetic did two things nobody expected. He captured the city that had killed the Guru's children. And then he abolished landlordism — two hundred and forty years before independent India got round to it.

From Madho Das to Banda Singh

At Nanded in 1708, Guru Gobind Singh met a Bairagi ascetic named Madho Das, who kept a hermitage there. What passed between them is described in various ways, but the outcome is clear.

Madho Das became Banda Singh Bahadur. The Guru gave him five arrows, a small band of Sikhs, a drum and a banner, and a hukamnama — a written order to the Sikhs of Punjab to support him. He sent him north.

Within weeks of that meeting, the Guru was dead. Banda arrived in Punjab as the representative of a man who no longer existed, carrying letters and arrows.

The army nobody expected

He raised his force from the countryside, and this is the detail that shapes everything that follows: he recruited overwhelmingly from the peasantry, including the lowest castes — people who under Mughal law and Hindu custom alike had never been permitted to bear arms at all.

Give a man who has been told his whole life that he may not carry a weapon a weapon, and tell him the land he works could be his, and you have created something the state has no idea how to fight.

Chappar Chiri and Sirhind, May 1710

On 12 May 1710, at Chappar Chiri near Sirhind, Banda's force met the army of Wazir Khan — the governor who six years earlier had ordered two children, aged nine and seven, bricked up for refusing to convert.

Wazir Khan was killed in the battle. Two days later Sirhind fell.

The state

What Banda did next is the part that qualifies this as a state rather than a rebellion. He established a capital at Lohgarh (Mukhlispur) in the Shivalik foothills, and he issued the three things that only a sovereign issues:

• Coins, struck not in his own name but in the names of Guru Nanak and Guru Gobind Singh.
• An official seal for state documents.
• A new calendar, dated from the victory.

In the language of the eighteenth century, that is a declaration of independence.

The land reform

And then he did the thing almost nobody knows about, which I think is the most remarkable act of his life.

He abolished the zamindari system in the territory he controlled, and gave ownership of the land to the people who actually farmed it.

Sit with the date. This is 1710. India would not legislate comparable land reform until after 1947. The French Revolution was seventy-nine years away. The Russian revolutionaries who made "land to the tiller" a slogan were two centuries in the future.

A peasant in Punjab in 1710, who had paid a share of his crop to a landlord his whole life as his father had, was told the field was his.

The Muslim question, handled properly

Banda has a reputation for anti-Muslim violence, particularly over what happened at Sirhind, and I am not going to skip it.

What the hostile sources say: Persian chronicles of the period describe severe destruction at Sirhind and considerable killing. Some later accounts describe the city as devastated wholesale.

What complicates it: Banda had Muslim followers in his own force — figures of around five thousand appear in some accounts — and is recorded as permitting them to pray and appointing Muslims to positions. Some modern Sikh historians have argued that the scale of destruction at Sirhind was exaggerated by chroniclers writing for a Mughal audience that wanted the rebels portrayed as monsters.

Where I land: Sirhind was sacked and there was real killing, including of civilians; that is not seriously deniable. Whether it reached the scale the Persian sources claim is contested, and those sources had obvious motives. What can be said with confidence is that this was a campaign of vengeance against a specific city and a specific governor, conducted by a force that included Muslims, rather than a war of religious extermination. The distinction is real, and it does not make the killing acceptable.

Gurdas Nangal and Delhi

The empire eventually concentrated against him. In 1715 Banda was besieged at Gurdas Nangal for roughly eight months. The defenders ate their horses, then bark and leaves. He was captured in December 1715.

He and around seven hundred prisoners were taken to Delhi and paraded through the city in June 1716, with severed heads carried on poles alongside them.

The prisoners were executed at a rate of roughly a hundred a day over several days. Contemporary European observers in Delhi recorded that not one accepted the offer of life in exchange for conversion.

Banda was kept until last. He was offered his life if he converted. He refused. His four-year-old son, Ajai Singh, was killed in front of him — the accounts say his heart was cut out and thrust into his father's mouth. Banda's eyes were then gouged out, his limbs severed one by one, and his flesh torn with red-hot pincers.

What the six years proved

Banda's state lasted about six years and then was extinguished completely.

But it had established two facts that could not be un-established. The first: a Sikh state was possible. It had coins, a seal, a capital and a policy. It had beaten a Mughal army in the field.

The second: what it would cost.

2The killing years

Between 1716 and 1762 the Sikh community came closer to being destroyed than at any point in its history — twice. What happened immediately afterwards is the part that requires explanation.

A price on your head

After Banda, the Mughal administration in Punjab — particularly under Zakariya Khan, governor of Lahore from 1726 to 1745 — set out to eliminate the Sikhs as a community rather than merely defeat them.

A formal scale of rewards was established: payment for information about a Sikh's whereabouts, a larger payment for a Sikh brought in alive, a larger one still for a Sikh's head.

The Sikhs left the settled districts. They went into the Lakhi Jungle in the south-west, into the Shivalik foothills, and across the Sutlej into the deserts of Rajasthan. Small mobile bands, living off the country, striking and vanishing.

Two names from this period are recited in the Ardas to this day:

• Bhai Mani Singh (1738), the custodian of the Harmandir Sahib, executed by being dismembered joint by joint for failing to pay a tax imposed on a Sikh gathering. Tradition records that when the executioner began at the wrist, he pointed out that the order specified joint by joint, and that the fingers came first.
• Bhai Taru Singh (1745), a young farmer arrested for feeding Sikhs. Offered his life if he cut his hair. His scalp was removed instead.

Chhota Ghallughara, 1746 — the Lesser Massacre

In 1746, a force under Yahya Khan, governor of Lahore, and Diwan Lakhpat Rai trapped a large body of Sikhs in the marshes near Kahnuwan, in Gurdaspur district.

Estimates: around 7,000 Sikhs killed, with a further 3,000 captured and subsequently executed at Lahore.

The Punjabi word used for it is Ghallughara — massacre, or holocaust. It was called the Chhota, the lesser, only because of what happened sixteen years later.

Vadda Ghallughara, 5–6 February 1762 — the Greater Massacre

Ahmad Shah Abdali — also called Ahmad Shah Durrani — invaded India nine times between 1748 and 1767. In February 1762 he came specifically for the Sikhs.

He caught the main Sikh body near Kup–Rohira, in today's Sangrur district. Critically, this was not an army on campaign. It was a moving community: fighters travelling together with their families — women, children, the elderly — trying to reach safety in the south.

The Sikh fighters formed a defensive cordon around the non-combatants and moved while fighting, trying to shield the column. The killing ran from Kup–Rohira through to Gehal, across a single day and into the next.

Abdali then went to Amritsar. He blew up the Harmandir Sahib with gunpowder and had the sacred pool filled with the carcasses of slaughtered cattle.

And then the part that needs explaining

Here is what happened next, and it is the reason this chapter exists.

By October 1762 — eight months later — the Sikhs fought Abdali again at Amritsar on Diwali.

By 1764 they had sacked Sirhind again and killed its governor.

By 1765 they had taken Lahore, the capital of Punjab, and struck their own coin in the name of the Gurus.

Five reasons it worked

1. There was nothing to capture. The Sikhs at this point had no fixed capital, no treasury in one place, no standing army that could be destroyed in a single set-piece battle. They were small mobile cavalry bands — jathas — that dispersed under pressure and reassembled afterwards. You cannot behead an organisation with no head.

2. They funded themselves off the invader. Abdali's great weakness was that he had to go home, and going home meant moving enormous baggage trains of loot and captives back through Punjab. Sikh cavalry specialised in hitting those columns. Horses, weapons and treasure flowed from the invasion into the resistance.

3. Rakhi. This is the institutional innovation and it is underrated. The Sikhs offered villages rakhi — protection — in exchange for a share of the harvest, typically around one-fifth. In return, the village was defended against raiders, including Afghan ones.

Understand what that actually is. A body that provides security over a defined territory and collects a regular proportional levy from it is not a bandit gang. It is a government. Rakhi converted the Sikhs from fugitives into a de facto administration with tax revenue, years before they held a single city.

4. The theology of Part 4, operating exactly as designed. A bounty system works by making people afraid. It does not work well on a community that recites a daily prayer listing its own dismembered and scalped, and asks for the same courage. Chapter 8 of Part 4 was not an abstraction; this is the field test.

5. Abdali kept leaving. He raided nine times and never once stayed to govern. The Sikhs were permanently there. In a contest between a visitor and a resident, the resident eventually wins.

3The misls

Between 1765 and 1799, Punjab governed itself through twelve armed republics and a biannual general assembly. It worked — and then it stopped working, in a way that tells you a great deal about Punjab.

The structure

At a Sarbat Khalsa — a general assembly of the whole Panth — held at Amritsar in 1748, a Gurmata was passed reorganising the scattered jathas into twelve misls, together forming the Dal Khalsa, under the leadership first of Nawab Kapur Singh and then of Jassa Singh Ahluwalia.

The word misl comes from Persian and Arabic usage meaning "equal," or a file or record. Both senses are apt.

The twelve were: Ahluwalia, Bhangi, Kanhaiya, Ramgarhia, Sukerchakia, Nakai, Dallewalia, Nishanwalia, Karorsinghia, Shaheedan (the Nihangs), Singhpuria, and the Phulkian group — the last of which produced the states of Patiala, Nabha and Jind, and which for that reason survives in the political geography of Indian Punjab today.

1783: the Sikhs in Delhi

In 1783 a Sikh force under Baghel Singh of the Karorsinghia misl entered Delhi and occupied parts of the city, including the Red Fort area.

The settlement negotiated with the Mughal emperor gave Baghel Singh permission to identify and build gurdwaras at the sites in Delhi associated with the Gurus. That is how Sis Ganj — where Guru Tegh Bahadur was beheaded — Rakab Ganj and Bangla Sahib came to be built.

The honest assessment

What the misl period got right: Punjab was governed by Punjabis for the first time in seven centuries. The Afghan invasions stopped — Abdali's successors could not hold anything. Agriculture recovered. And there was a genuine, functioning mechanism for collective decision-making that had real authority over powerful armed men.

What went wrong: the misls fought each other. Constantly. The Kanhaiya against the Sukerchakia. The Bhangi against almost everyone. Territory changed hands between Sikh chiefs as often as it was taken from outsiders.

And as individual misldars grew rich and their positions became hereditary, the authority of the Sarbat Khalsa declined. By the 1790s the assembly had largely stopped functioning as a binding institution, and Punjab was a patchwork of feuding principalities.

4Ranjit Singh

A one-eyed, illiterate teenager from Gujranwala took Lahore in 1799 and built the last independent state in India. He also, in the process, quietly dismantled the thing that had made Punjab governable.

The man

Born 1780 at Gujranwala, into the Sukerchakia misl — one of the smaller ones. Smallpox in childhood cost him an eye and scarred his face. He never learned to read or write, and never pretended otherwise.

In 1799, aged nineteen, he took Lahore. And the manner of it is worth noting: he was effectively invited. The city's leading residents, including Muslim notables, had had enough of the Bhangi misldars who held it and appealed to him to come. He entered without a serious fight.

In 1801, on Vaisakhi, he was proclaimed Maharaja.

The coins, and what they say

Here is a detail I find genuinely striking. Ranjit Singh did not put his own name or face on his currency.

The coins of his empire — the Nanakshahi rupee — were struck in the names of Guru Nanak and Guru Gobind Singh. He called his government the Sarkar Khalsaji — the government of the Khalsa — and described himself as its servant.

Whether that was sincere humility, shrewd politics, or both, it is a remarkable thing for an absolute monarch to do, and it tells you where legitimacy was understood to reside.

The empire

One line in that table deserves its own paragraph.

Ranjit Singh is the last Indian ruler to conquer and hold Afghan territory. For eight hundred years — Part 2 and Part 3 of this work — the traffic through the Khyber Pass ran one way. Ghazni, Ghor, the Mongols, Timur, Babur, Nadir Shah, Abdali. All of them came down through those passes into Punjab.

Ranjit Singh reversed it. Peshawar under Punjabi administration, and a Punjabi governor at the mouth of the Khyber. Whatever else is argued about this period, that fact stands alone in the subcontinent's history.

The army

He built it deliberately and expensively. The Fauj-i-Khas was drilled to European standards by officers he recruited from Napoleon's disbanded armies:

• Jean-Baptiste Ventura and Jean-François Allard, both former Napoleonic officers, who arrived in 1822 and trained the infantry and cavalry.
• Claude Auguste Court, who reorganised the artillery.
• Paolo Avitabile, an Italian, who governed Peshawar with a severity that is still remembered there.

The result was widely regarded as the most formidable indigenous army in India. The British, who fought everyone else in the subcontinent during those decades, did not fight Ranjit Singh. For forty years they signed treaties with him instead. That is the clearest testimony available.

Who actually ran the Sikh Empire

This section contains what I regard as the single most useful set of facts in Part 5, and I want to lay it out carefully because it cuts in more than one direction.

The population of the Sikh Empire was overwhelmingly Muslim. Once Multan, Kashmir, Peshawar and western Punjab were absorbed, Muslims formed a large majority — most estimates put them in the region of 70 to 80 per cent. Sikhs were a minority in their own empire, probably around a tenth of the population. Hindus were a substantial share.

Now look at the government:

And the policy:

• No state religion. The Sikh Empire never declared one.
• Jizya abolished. The tax on non-Muslims was gone; there was no equivalent tax imposed on non-Sikhs.
• Endowments to all. He funded gurdwaras, and also Hindu temples and Muslim shrines. The Badshahi Mosque in Lahore, which had been used as a military magazine, was eventually restored to religious use.
• Cow slaughter was banned — a genuine restriction on the Muslim majority, and it belongs in an honest account.
• He never imposed the death penalty. This is widely reported across sources and, if accurate, is extraordinary for any nineteenth-century sovereign anywhere in the world.

He also had the Harmandir Sahib rebuilt in marble and covered in gold leaf, which is why the world calls it the Golden Temple. That building's appearance is his doing.

The criticisms, at full strength

Now the other side, because Rule 3 applies to people I admire as much as to anyone else.

He killed Punjab's own political institutions. This is the most serious charge and I think it is correct. The Sarbat Khalsa and the Gurmata — the assembly and the binding collective resolution described in Chapter 3 — did not survive him as functioning bodies. He replaced a confederal republic with a personal autocracy. It worked brilliantly, because he was brilliant. It left nothing behind that could function without him. Chapter 5 is the invoice for that decision.

Kashmir. Kashmiri accounts of Sikh rule are not warm. Taxation was heavy, the shawl industry was squeezed, and a severe famine in the 1830s was badly handled. The empire's periphery was governed as conquered revenue territory, and it felt like it.

Avitabile's Peshawar. The Italian governor's methods — public hangings in numbers, bodies left displayed — are remembered locally as a byword for terror.

The Sutlej line. In the Treaty of Amritsar, 1809, he accepted the Sutlej as his eastern boundary, effectively conceding the cis-Sutlej states — Patiala, Nabha, Jind — to British protection. He never fought the British and never tested them.

5Ten years

Ranjit Singh died on 27 June 1839 in possession of the strongest state in India. By March 1849 it did not exist. Understanding how is more useful than admiring how it was built.

The succession, in four years

Four rulers in four years, and all but one violently removed.

Why it collapsed — five reasons

1. There was no institution. This is the central one. Ranjit Singh had concentrated every function of the state in himself: he was the final authority on appointments, revenue, war, justice and diplomacy. There was no council with independent standing, no law of succession, and no functioning Sarbat Khalsa to arbitrate. When he died, the state had no skeleton — only muscle.

2. An enormous army with no master. The Khalsa Army was around a hundred thousand strong, professional, well paid and European-trained. Suddenly it had no commander whose authority it recognised. It organised itself into regimental committees — panchayats — and became a political actor in its own right, effectively deciding who sat on the throne and demanding pay increases from whoever did. An army that chooses the government is a phase most collapsing states pass through.

3. Factions. The Dogra brothers against the Sikh sardars, both against the army panchayats, everyone against Rani Jindan. The court became a place where ministers were murdered as a routine political method.

4. A child on the throne.

5. The British were next door and had finished everything else.

The First Anglo-Sikh War, 1845–46

The Khalsa crossed the Sutlej on 11 December 1845.

Four major battles followed: Mudki, Ferozeshah, Aliwal and Sobraon.

The fighting was extremely hard. At Ferozeshah in December 1845 the British came close to catastrophe; the Governor-General, Hardinge, was present and reportedly made arrangements to destroy state papers in case of defeat. British accounts of these battles are consistently, almost uncomfortably, admiring of the Sikh infantry and artillery.

And the reason they lost

The Wazir, Lal Singh, and the commander-in-chief, Tej Singh, were in secret communication with the British throughout the war.

This is not Punjabi folklore. It is documented in the British records. Both men were in contact with Captain Peter Nicholson, a British officer at Ferozepur, sending information and receiving instructions.

The consequences showed on the field. Reserves were withheld at moments when they would have been decisive. Withdrawals were ordered from positions that were winning. At Sobraon, the final battle, Tej Singh left the field early, crossed the bridge of boats over the Sutlej, and the bridge was destroyed behind him — trapping his own army on the wrong bank with the river at its back.

I want to be careful here, because "we were betrayed" is a story defeated peoples tell themselves everywhere and it is usually a comfort rather than an analysis. In this case the documentation is unusually good, and it is the victors' documentation, which is the kind that counts. The soldiers fought. The generals did not.

The treaties, and the sale of Kashmir

The Treaty of Lahore, March 1846: a heavy indemnity, the cession of the Jalandhar Doab, and a drastic reduction of the Khalsa Army.

The Treaty of Amritsar, March 1846: the Lahore Darbar could not pay the indemnity. So Kashmir was sold — transferred to Gulab Singh Dogra for seventy-five lakh rupees.

The Second Anglo-Sikh War, 1848–49

It began in April 1848 with the revolt of Mulraj, the governor of Multan, and the killing of two British officers. It spread until much of the Sikh army had joined what became a general rising.

Chillianwala, January 1849, was one of the bloodiest actions the British fought in India. British casualties were around 2,300, several regiments lost their colours, and the result was at best a draw. In Britain it was received as a disaster and there were calls for the commander's removal.

Gujrat, February 1849, was decisive — won largely by overwhelming British artillery.

The Khalsa surrendered at Rawalpindi. Annexation was announced on 29 March 1849.

Duleep Singh, aged ten, was deposed and pensioned. The Koh-i-Noor was taken and sent to Queen Victoria. Duleep Singh was converted to Christianity, taken to England, became a favourite at court, later attempted to return to India and reclaim his throne, was stopped at Aden, and died in a Paris hotel in 1893.

Of the surrender itself, one line survives. As the Khalsa soldiers laid down their weapons, a veteran is recorded as saying:

6The audit

Punjab has been independent. We have the records. So instead of arguing about whether it could be again, let us establish what it actually took the first time, and check the list.

The method

I am going to treat the Sikh Empire the way I would treat a company that once worked and is now proposed to be restarted. What were the inputs that made it function? Are those inputs available now?

This is not a rhetorical exercise and I am not stacking it. Where Indian Punjab today has something, I will say so.

The finding that ought to trouble everybody

Look at the last row again, and then at Chapter 5.

Even the successful version of an independent Punjab had a fatal structural defect. It was held together by one man's ability, because he had dismantled the collective institutions that might have outlasted him. It survived forty years while he lived and disintegrated in ten once he did not.

That is not an argument that Punjabis cannot govern — the misl period in Chapter 3 shows they can. It is an argument that the specific model everyone reaches for as the precedent was, on its own record, structurally fragile in exactly the way Punjab has historically been fragile: brilliant under a unifying figure or an external threat, and prone to fragmentation without one.

The counter-argument, at full strength

I promised in the Author's Note that I would state the other side properly, so here it is, made as well as I can make it.

My answer, and where I am deferring it

That is a serious argument and it deserves a serious response rather than a dismissal. It is right that I should not measure a hypothetical modern state against a nineteenth-century conquest empire, and Chapter 6 alone does not settle the question.

But I will note one thing here and hold the rest for Part 12.

Every example in that list has a specific compensating asset, and the asset is usually the whole explanation. Singapore has the world's busiest transshipment port sitting on the Strait of Malacca — its entire economy is that location. Switzerland is landlocked but surrounded by friendly, wealthy, integrated European states, with five centuries of accumulated institutions and capital and a banking system that everyone needs. Israel has the largest sustained military and economic aid relationship in modern history. Slovakia and the Baltics joined the European Union and NATO, which means they exchanged one union for a larger and richer one rather than becoming genuinely independent actors.

So the honest form of the question is not "can small states succeed?" — obviously they can. It is: what is Punjab's compensating asset?

That is the question Part 12 will answer, with numbers rather than adjectives: water rights, procurement dependence, fiscal transfers, trade access, and what happens to a landlocked territory whose only two possible transit neighbours are the two states it would have just separated from.

I am not going to pre-judge it here. I have put the counter-argument in its own words, and I have told you where I will meet it.

7Scorecard: what Part 5 proves

Timeline of Part 5

What happens in Part 6

Part 6, Loyal and Rebel at Once, covers 1849 to 1947 — and it is where the pattern I have been tracking since 326 BCE reaches its most extreme form.

Within eight years of annexation, the same Punjabi soldiers who had fought the British at Chillianwala were fighting for them in the Revolt of 1857. The British then built a theory around it — the "martial races" doctrine — and turned Punjab into the recruiting ground of the Indian Army. By the First World War, a province with a small share of India's population was supplying an enormous share of its soldiers. Punjabi troops fought in France, Mesopotamia, East Africa and Gallipoli.

And simultaneously, the same province produced the Ghadar Party, the Komagata Maru, Jallianwala Bagh, Bhagat Singh, the Babbar Akalis, and the Gurdwara Reform movement — which was the first mass non-violent civil resistance campaign in India and is largely forgotten outside Punjab.

Part 6 also covers the two British decisions that shaped everything afterwards: the canal colonies, which made Punjab the granary of India and created the water politics of Part 8; and the Land Alienation Act and communal electorates, which hardened Punjab's communities into political blocs and made 1947 possible.

The question Part 6 answers: how was Punjab simultaneously the most loyal and the most rebellious province in British India? And the answer is not hypocrisy. It is something much more interesting, and it explains a great deal about what happens in 1984.