Where We Left Off, and a Second Warning
A recap, a reminder of the boxes, and a note about the most argued-over stretch of Indian history after the one in Part 3.
Five parts have brought us here.
Part 1 built the land — the collision that raised the Himalayas, the rivers that ground them into soil kilometres deep, and the first people. Part 2 was Harappa: a thousand settlements, planned cities, standardised bricks and weights, and a script nobody can read. Part 3 was the Vedic age — the Rigveda composed on this land, the Battle of the Ten Kings on the Ravi, and the long argument about where those people came from. It ended with the centre of Indian civilisation moving east, away from Punjab.
Part 4 brought Punjab back into recorded history: Persian rule, Alexander and Porus on the Jhelum, the refusal at the Beas, then Chandragupta Maurya and Ashoka's inscriptions in stone.
Part 5 was seven centuries of arrivals — Greeks, Sakas, Parthians, Kushans, Hunas — every one of them absorbed within two or three generations. It produced the Gandharan Buddha, sent Buddhism up the Silk Road to China, and ended with the Gupta empire fragmenting around 550 CE and the Buddhist civilisation of the north-west in ruins.
This part covers what came next, and it is the longest stretch in the series: nine hundred years.
What the boxes mean
Same system throughout. Skip if you know it.
Blue explains a hard word, right where it appears.
Example — Sultan: an Arabic word meaning "authority" or "power". Used as a ruler's title across the Islamic world. A sultanate is the territory a sultan rules.
Green turns something too big or too strange to picture into something ordinary.
Brown shows the evidence itself, and is honest about what it can and cannot prove.
Orange means serious people disagree. Both sides get their strongest case, then a plain statement of where things stand.
Purple digs out something everyone takes for granted without proving.
Dark boxes close each chapter with the key points in the plainest words I can find.
The second warning
Part 3 came with a warning about the Aryan question. This part needs one too, for a different reason.
These nine centuries are when large numbers of people in Punjab became Muslim, and when Muslim dynasties came to rule northern India. That history is argued about today with as much heat as anything in Part 3, and often with far less care.
You will meet two ready-made stories.
One says: this was a period of conquest and forced conversion, temples destroyed and a population terrorised into changing religion.
The other says: this was a period of harmony and syncretism in which everyone got along and religion was barely an issue.
Neither is supported by the evidence. The first cannot explain why Islam ended up strongest in the places furthest from Muslim political power, and weakest in the places that power held longest and hardest. The second cannot explain the temples that really were destroyed, the taxes that really were levied on non-Muslims, or the slaughter that really did happen when Timur came through.
What actually happened is slower, more local and less satisfying than either story. Chapters 7 and 8 take it head on, and Chapter 8 in particular sets out the four competing explanations and tests each against the geography.
Underneath most modern argument about this period sits an assumption that would have baffled everyone involved:
That "Hindu" and "Muslim" were the two main teams, and that everybody knew which one they were on.
Consider what the period actually contains. Turkish sultans fighting other Turkish sultans. Hindu Rajput contingents in sultanate armies. Muslim rulers granting land to temples. Hindu officials running Muslim states' revenue. Mongols — neither Hindu nor Muslim for much of this period — attacking everybody. Sufi teachers with Hindu disciples and Hindu teachers with Muslim ones.
The word "Hindu" itself, as Part 4 noted, began as a Persian geographical term meaning people from beyond the Indus. It becomes a religious label slowly, and partly because of this period — you need a word for "not Muslim" before "Hindu" starts doing that job.
So the two-team picture is not just an oversimplification. It is, in part, a product of the very history it claims to describe.
The loyalties that actually organised people's lives were caste, clan, region, language, occupation, guild and the household of a particular ruler. Religion mattered — sometimes enormously. It was one line among many, and rarely the one that decided who fought whom.
1The Long Middle
Four and a half centuries in which almost nothing about Punjab was written down. This chapter is short — and the reason it is short is more interesting than most of what is in it.
1.1 — What we can say
After the Gupta empire fragmented around 550 CE, northern India broke into regional kingdoms that fought each other for the next several centuries.
A ruler called Harsha, based at Kanauj in the Ganges plain, put together a large kingdom in the early seventh century and held it until his death around 647 CE. It fell apart immediately afterwards, because he left no heir. Punjab was on its edge and not securely part of it.
To the north, Kashmir became a serious power. In the eighth century its ruler Lalitaditya campaigned widely and dominated much of the Punjab plain for a time.
Further east and south, three large kingdoms spent roughly two hundred years fighting over Kanauj — the Palas from Bengal, the Pratiharas from Rajasthan and the Ganges plain, and the Rashtrakutas from the Deccan. Historians call it the tripartite struggle. Punjab was not one of the three and was mostly outside it.
And in the north-west, from the ninth century, a dynasty appears that Chapter 2 is about: the Hindu Shahis.
That is close to all we can say with confidence about four and a half centuries.
1.2 — The traveller who saw the ruins
One outside observer gives us a glimpse, and it is a bleak one.
A Chinese Buddhist monk named Xuanzang travelled overland from China to India and back between roughly 630 and 645 CE, studying Buddhism and collecting texts. He wrote a detailed account of everywhere he went.
When he passed through Gandhara and the north-west, he described what Part 5 left standing: monasteries. Except most of them were no longer standing. He recorded monastery after monastery deserted, buildings collapsed, communities gone, and only a small number still functioning.
Part 5 ended with Gandharan Buddhism in trouble. Xuanzang is the confirmation, on the ground, roughly a century later.
He is not a hostile witness. He was a devout Buddhist who had walked across a continent specifically to see these places. He describes them the way a pilgrim describes a holy site he has found abandoned.
What he saw was the end of something that had lasted five hundred years — and he was one of the last people to see even the ruins while anyone still remembered what they had been.
1.3 — Why this chapter is short
Part 5 warned about the phrase "dark age". Here is the mechanism, laid out fully, because this chapter is the clearest example in the whole series.
The assumption is that how much we know about a period reflects how much happened in it.
It never does. What we know depends on who was writing things down and whether what they wrote survived. Both of those depend on institutions, not on events.
Consider what generates a written record in this period: a royal court employing poets and keeping archives; a religious institution copying and storing manuscripts; a state carving inscriptions to record grants and victories; a foreign visitor writing a travel account.
Now consider Punjab between 550 and 1000 CE. No large stable court. The Buddhist monasteries — which had been the main copying institutions of the north-west — were collapsing or gone. The script those monasteries wrote in, Kharosthi, was already extinct. No foreign pilgrim after Xuanzang left a comparable account for centuries.
So the record thins to almost nothing. And we call the result a dark age, as though the darkness were in the period rather than in our sight of it.
Meanwhile, during those same four and a half centuries: farmers farmed, villages traded, families were born and buried, a language was slowly changing from something like Prakrit into something like early Punjabi, and religious practice was shifting in ways that shaped everything after. All of it invisible.
This is Part 1's survivorship bias in its purest form — not what survived, but what was recorded at all.
1.4 — What was quietly happening
Two changes, both invisible in the record and both decisive.
Buddhism was declining across northern India. Not through a single event. The monasteries depended on royal and merchant patronage; when the political order fragmented and long-distance trade contracted, that money thinned. Meanwhile devotional Hindu traditions grew, and temples became the institution that attracted the grants monasteries had once received. By the end of this period, Buddhism in the north-west was essentially finished.
And the language was changing. The Prakrits of Part 5 were evolving into the stage linguists call Apabhramsha, and from that stage the modern languages of northern India — Punjabi, Hindi, Bengali, Gujarati, Marathi — would emerge. Nobody wrote this down as it happened. We can see it only by comparing what came before with what came after.
Apabhramsha: the late stage of the Prakrit languages, roughly 600–1200 CE, out of which the modern North Indian languages developed. The word literally means something like "falling away" — a term coined by Sanskrit grammarians who regarded ordinary speech as a corruption of the proper language.
Note what that name reveals about who was doing the naming. The everyday language of millions of people was labelled by scholars as a degraded version of the language those scholars themselves wrote in.
After the Guptas fragmented around 550 CE, northern India broke into competing regional kingdoms. Punjab was on the edge of all of them.
Xuanzang, travelling from China around 630–645 CE, found the Buddhist monasteries of the north-west already largely in ruins.
We know almost nothing about Punjab in these four and a half centuries — because the institutions that write things down had collapsed, not because nothing happened.
Two invisible changes were decisive: Buddhism faded out of northern India, and the local language changed from Prakrit into something on the way to Punjabi.
2The Hindu Shahis: The Last Defence
For about a hundred and fifty years one dynasty held the passes into India. They lost, retreating from Kabul to the Indus to the Salt Range and finally out of history. The interesting question is not why they lost. It is why they were alone.
2.1 — Who they were
In the ninth century, a dynasty took power in the region around Kabul and the north-west frontier. Historians call them the Hindu Shahis, to distinguish them from an earlier Turkish dynasty in the same place.
Their territory covered eastern Afghanistan, Gandhara, and stretched into the Punjab plain. Their job, whether they thought of it that way or not, was to hold the passes.
Their capitals moved as they were pushed back, and the sequence tells the whole story:
- Kabul — the original seat, west of the mountains.
- Udabhandapura (Hund), on the Indus — after Kabul was lost.
- Nandana, in the Salt Range of Punjab — after the Indus line broke.
Each move is a defeat. By the last one they are inside Punjab, defending a mountain fort rather than a frontier.
2.2 — The kings
Three names matter.
Jayapala ruled from around 964. He fought the rising Ghaznavid power to his west, lost badly at Peshawar in 1001, and — according to the sources — killed himself afterwards rather than live with the defeat.
Anandapala, his son, assembled the largest coalition anyone managed in this period and was defeated around 1008.
Trilochanapala, his son, was pushed out of Punjab by 1021. The dynasty ended within a few years after that.
An unusual amount of what we know about the Hindu Shahis comes from a man who arrived with the army that destroyed them.
Al-Biruni was a Central Asian scholar — a mathematician, astronomer and geographer — who came to India in the entourage of Mahmud of Ghazni. He stayed for years, learned Sanskrit, read Indian scientific and religious texts in the original, and wrote a long, careful book about India that is unlike almost anything else from the period.
On the Hindu Shahis he is notably respectful. He describes the dynasty as generous and noble, and writes about their fall with something close to regret — which is a remarkable thing to find in a book written under the patronage of the ruler who ended them.
Why this matters as evidence: a hostile source praising its subject is unusually reliable on that point. Al-Biruni had no reason to flatter a defeated Indian dynasty and every reason not to.
And the limitation: he is one man, writing decades after some of the events, in a language and a framework of his own. Most of what he says about the Shahis cannot be checked against anything else, because there is nothing else.
2.3 — Why did nobody help?
This is the question people ask most about this period, usually in a tone of accusation. It deserves a serious answer rather than a moral one.
Anandapala did assemble a coalition. Sources describe contingents from several northern Indian kingdoms joining him around 1008. So the idea that nobody came is wrong. What is true is that the coalition lost — and that after it lost, nobody tried again on that scale. One failed alliance is a different thing from indifference.
There was no "India" to defend. There were a dozen kingdoms that had spent two centuries fighting each other over Kanauj. Each had rivals closer to home than Ghazni, and a ruler who sent his army north-west left his own borders open to the neighbour he actually feared.
Communication and supply made it worse: moving an army from Bengal or the Deccan to the Indus took months and cost enormously, to defend land that was not yours, against a threat that had not yet reached you.
Mahmud's early campaigns looked like raids, not conquest — he came, took plunder, and went back to Ghazni. Raids had happened before. It would have been reasonable, and wrong, to conclude this was weather rather than climate.
By the time it was clear that Punjab was being annexed rather than robbed, the frontier dynasty that might have anchored a defence was gone.
This is a well-understood problem and not unique to India. It has appeared in every era where several rivals face a common danger: everyone benefits if someone stops it, everyone would prefer that someone else pay for stopping it, and so it is not stopped.
What is worth resisting is the retrospective moral version — "they should have united". They had no framework in which uniting made sense, no shared identity that would have made it obvious, and no way of knowing what we know now about how it turned out.
The question "why didn't India unite?" contains its own answer, and it is the wrong one.
The assumption is that "India" existed as a thing that could unite.
In 1000 CE there was no Indian state, no Indian identity, no Indian army and no concept under which a king in Bengal owed anything to a king near Kabul. There was a subcontinent containing many kingdoms, many languages and many peoples, who had been at war with each other continuously for centuries.
Asking why they did not unite against an outsider is like asking why the kingdoms of Europe did not unite against the Ottomans. Sometimes they partly did, usually they did not, and nobody today reads that as a moral failing of "Europe" — because we do not imagine Europe as a single thing that ought to have acted as one.
The assumption comes from reading a modern nation backwards into a period that had none. It is the same error as Part 3's, in another form: taking a category that exists now and treating it as though it must always have existed.
The Hindu Shahis held the north-west from the ninth century. Their capitals moved Kabul → Indus → the Salt Range in Punjab — each move a defeat.
Jayapala lost at Peshawar in 1001. Anandapala assembled a coalition and lost around 1008. Trilochanapala was pushed out of Punjab by 1021.
Much of what we know comes from al-Biruni, a scholar who arrived with the army that destroyed them — and who wrote about them with respect.
Help did come once, and failed. After that, each kingdom had rivals closer than Ghazni and no reason to bear the cost alone.
And "why didn't India unite?" assumes an India that did not exist. There were many kingdoms, not one country.
3Mahmud of Ghazni
Seventeen campaigns into India in thirty years. He is remembered on one side of the modern border as a destroyer and on the other as a hero, and both versions leave out most of what he was actually doing.
3.1 — Where he came from
Ghazni is a town in eastern Afghanistan. In the tenth century it was held by a Turkish military family who had begun as slave-soldiers in the service of a Persian dynasty and had made themselves independent.
Slave-soldier: across the medieval Islamic world, rulers bought or captured boys — often Turks from Central Asia — and trained them as elite professional soldiers. Because they had no local family or faction, they were considered more loyal than free nobles.
The system worked so well that these soldiers repeatedly took power for themselves. The Ghaznavids began this way. So, as Chapter 5 shows, did the first dynasty of the Delhi Sultanate. A slave in this sense could become an emperor, and several did.
Mahmud ruled from 998 to 1030. His empire ran across Afghanistan and into Iran and Central Asia. India was not where he lived. It was where he went for money.
3.2 — What he did
Between 1000 and 1027 he led something like seventeen campaigns into India.
The early ones broke the Hindu Shahis. Later ones reached far into the Ganges plain and across to Gujarat. Targets included Multan, Thanesar, Mathura, Kanauj, and — most famously — the great temple at Somnath on the coast of Gujarat, sacked in 1025–26.
The temples were the point. In this period a major temple was not only a religious building. It was a treasury, holding accumulated donations in gold, silver and jewels, and often the wealth of the surrounding region.
And in 1021, Punjab was annexed. After Trilochanapala's defeat, Lahore was placed under a Ghaznavid governor. It stopped being a target and became a province — the first permanent Muslim-ruled territory in India, and the subject of Chapter 4.
3.3 — Gold or God?
He described his campaigns in religious terms, and the court writers who recorded them presented him as a destroyer of idols acting for the faith. He sought and received recognition from the caliph in Baghdad. The targets were temples specifically. He is remembered in a great deal of later Islamic writing as a model of pious kingship.
He was building an empire in Central Asia and Iran, and empires need cash. Professional armies must be paid continuously, whether or not they are fighting. India was the richest accessible source of portable wealth in his world, and temples were where it was concentrated.
Several things fit the money explanation better than the religious one. He did not stay — he took plunder and went home, which is not what someone campaigning to spread a religion does. He attacked Muslim rulers too, in Multan and in Iran, with the same energy. He employed Hindu soldiers in his own army, and had Hindu commanders. And the destruction of a temple served a practical purpose beyond ideology: it demonstrated that the local ruler could not protect what mattered most, which is how you break a region's will to resist.
Asking which motive was "the real one" probably misunderstands how people work. Mahmud could believe he was serving God and also need the gold, without experiencing any contradiction. Most people who do large violent things believe they are justified.
What is not in dispute: the campaigns happened, the temples were destroyed, the plunder was enormous, and Punjab was annexed.
3.4 — The memory war
This needs saying plainly, because you will meet it constantly.
Mahmud is remembered very differently depending on where you are standing.
In much of Indian public memory he is the archetypal destroyer — Somnath in particular has become a symbol, invoked in politics for well over a century. In much of Pakistani public memory he is a founding hero, and military hardware has been named after him.
Both versions are doing present-day work. Neither is a description of an eleventh-century Turkish ruler whose main strategic concern was a rival dynasty in Central Asia.
Here is the assumption doing the damage, and it runs through the whole of this part:
That people in the past belonged to the same teams we belong to now, and were fighting the fights we are fighting.
Under this assumption, Mahmud attacking Somnath becomes an episode in a continuous war between two communities that runs from then to now.
But look at what he actually did. He attacked Muslim-ruled Multan. He fought Muslim rivals in Iran. He employed Hindu troops under Hindu officers. His most bitter rivalries were with other Muslim dynasties over Central Asian territory.
He was a Turkish warlord in a world of competing Turkish, Persian and Indian warlords, using the ideological tools available to him. Attaching him to a modern communal narrative requires ignoring most of his actual career.
The test that follows from this, and it cuts both ways: when someone tells you what a medieval ruler "represents", ask whether the claim survives contact with who that ruler actually fought. Very often it does not — and very often the person making the claim has not checked.
To be clear about what this is not saying. It is not saying the destruction did not happen or did not matter. It happened, it was violent, and for the people it happened to it was a catastrophe. What it says is that the violence belonged to eleventh-century politics, and reading it as an early chapter of a twentieth-century quarrel gets both centuries wrong.
3.5 — The scholar who came with the army
One more thing came out of Mahmud's campaigns, and it is remarkable.
Al-Biruni, whom Chapter 2 introduced, spent years in India as part of the Ghaznavid presence. He learned Sanskrit — properly, not superficially. He read Indian works on astronomy, mathematics and philosophy in the original. And around 1030 he produced a long book describing Indian religion, science, customs and thought.
What makes it extraordinary is the method. He states at the outset that he intends to report what Indians actually believe, in their own terms, rather than argue against it. He quotes his sources. He notes where he does not understand something. He corrects earlier foreign writers who got things wrong.
This is one of the strangest pairings in this series.
The same expedition produced the sack of Somnath and the first serious attempt by an outsider to understand Indian civilisation on its own terms — a book still used by scholars today as a source on eleventh-century Indian science and society.
Both men were on the same payroll.
History does not sort itself into good things and bad things happening to different people at different times. It is usually the same events, producing both.
Mahmud of Ghazni ruled from Afghanistan, 998–1030, and led around seventeen campaigns into India.
He targeted temples, because temples were treasuries. Somnath in 1025–26 is the most famous.
The religious framing was real and also useful. He also attacked Muslim rulers, employed Hindu troops, and did not stay — which is not what a religious conquest looks like.
In 1021 Punjab was annexed and Lahore given a governor. That is the permanent change, not the raids.
And the same expedition produced al-Biruni's book — the first serious outside study of India, written by a man who learned Sanskrit to do it properly.
4Lahore Becomes a Capital
In 1021 Lahore was a provincial town. Within a century and a half it was the seat of a dynasty, a centre of Persian poetry, and the home of the first book ever written on Sufism in Persian. This is the moment Punjab changes character.
4.1 — From target to province
The raids of Chapter 3 were spectacular but temporary. What actually changed Punjab was quieter.
After 1021, Punjab was governed. A Ghaznavid governor sat in Lahore. Taxes were assessed and collected. Soldiers were garrisoned and paid. Officials with Persian titles wrote records in Persian. Cases were heard. Land was granted.
Raids come and go. Administration stays, and it changes everything it touches — because to deal with a government you must eventually deal in its language, its categories and its calendar.
4.2 — And then a capital
In the middle of the twelfth century the Ghaznavids ran into a rising power in Afghanistan — the Ghurids, from the mountains of central Afghanistan. Around 1150 the Ghurids sacked Ghazni itself.
The Ghaznavids, driven out of their homeland, moved their capital to the one substantial territory they had left.
Lahore became the capital of a dynasty. Not a provincial headquarters — the seat of the court itself, from around 1163 until the Ghurids finished them off in 1186.
What actually changes when a town becomes a capital?
Money arrives. Tax revenue from the whole territory flows to where the ruler is and is spent there.
People arrive. Officials, soldiers, scribes, judges, and everyone who sells things to them — builders, weavers, cooks, grooms, jewellers.
Talent arrives. Poets, scholars and religious teachers go where the patronage is. A court is a job market for the educated.
And a language arrives with them. Because the court works in Persian, ambitious people learn Persian.
Part 3 described Punjab going from the centre of the Vedic world to the edge of the Gangetic one. This is Punjab acquiring a centre of its own for the first time in over a thousand years — and the last time before it happens again under Ranjit Singh in Part 10, in the same city.
4.3 — Persian
This is the most consequential thing in the chapter and it is easy to skip past.
Persian became the language of government in Punjab, and stayed the language of government for eight hundred years.
Ghaznavids, Ghurids, the Delhi Sultanate, the Mughals, and even the Sikh empire of Ranjit Singh in Part 10 all ran their administration in Persian. It was finally replaced as the official language of the region by the British, in the 1830s.
Persian (Farsi): the language of Iran. In the medieval period it became the prestige language of government, poetry and refinement across a vast area — Iran, Central Asia, Afghanistan, and northern India — rather as Latin functioned in medieval Europe.
Learning Persian was how an ambitious young man from any background got a government job. That is why so many Persian words are in Punjabi, Urdu and Hindi today: not through conquest so much as through paperwork.
Look at ordinary Punjabi vocabulary and the deposit is everywhere: words for government, law, land, revenue, courts, letters, and a great deal of everyday life. As Part 1 pointed out, even the name Punjab is Persian.
Lahore also became a place where Persian was written well. The best-known poet of the Ghaznavid court there, Masud Sa'd Salman, spent long years imprisoned and wrote a body of prison poetry that is still read — a Punjabi-born Persian poet, writing in a Punjabi city, in the eleventh and early twelfth centuries.
4.4 — Data Ganj Bakhsh
And in the same century, Lahore acquired the figure who still defines it religiously for millions of people.
Ali Hujwiri came to Lahore from Ghazni and died there around 1072. He is universally known as Data Ganj Bakhsh — roughly, "the giver who bestows treasure".
He wrote a book called the Kashf al-Mahjub, "the unveiling of the hidden". It is the earliest treatise on Sufism written in Persian, and it became one of the standard works of the tradition, studied for centuries afterwards across the Persian-reading world.
His shrine in Lahore is one of the most visited religious sites in South Asia. Chapter 7 explains what a shrine like that actually does.
Set two dates side by side.
1025–26: Mahmud sacks Somnath.
c. 1072: a Sufi teacher dies in Lahore, having written the first Persian book on the inner life of the spirit, and becomes the object of devotion for people of every background in the city for the next thousand years.
Fifty years apart. The same religion. The same political sphere.
Any account of this period that gives you only one of these is not describing it.
Something happens to how we describe cultural change depending on who brought it, and it is worth catching.
Part 5 described Greek artistic technique arriving in Gandhara and producing the Buddha image. That is usually called influence, or synthesis, or a flowering.
This chapter describes Persian arriving in Punjab and reshaping its administration, vocabulary, poetry and religious literature. That is very often called imposition.
The processes are structurally the same. A prestige language and its associated culture arrive with a ruling group, are adopted by locals who want to get on, and produce something new that is neither the source nor the original.
So why do they get different words? Usually because of what happened afterwards, and how the descendants of each are positioned today.
I am not arguing the two are morally identical. Conquest is conquest, and Chapter 6 describes real brutality. The point is narrower and worth holding: notice when the same process gets a warm word in one chapter and a cold one in another, and ask what is actually doing the choosing.
Raids change little. Administration changes everything. From 1021 Punjab was governed from Lahore, with taxes, records and courts.
After the Ghurids sacked Ghazni, Lahore became the Ghaznavid capital from around 1163 to 1186 — Punjab's first real centre in over a thousand years.
Persian became the language of government here for eight hundred years, until the British replaced it in the 1830s. That is why so much of Punjabi's official and everyday vocabulary is Persian.
Ali Hujwiri — Data Ganj Bakhsh — died in Lahore around 1072 and wrote the first Persian book on Sufism. His shrine is still one of South Asia's most visited.
5Tarain, and the Conquest
Two battles on the same ground, one year apart, between the same two commanders. The first was an Indian victory. The second changed northern India for six hundred years. What happened in between is the most useful military lesson in this series.
5.1 — The Ghurids
The Ghurids came from the mountains of central Afghanistan. They destroyed Ghazni around 1150, finished the Ghaznavids at Lahore in 1186, and then looked east.
Muhammad Ghori — properly Muizz al-Din Muhammad — is the commander who matters. Unlike Mahmud, he was not raiding. He was conquering, and he intended to hold what he took.
5.2 — Tarain, 1191
The most powerful Indian ruler in his path was Prithviraj Chauhan, whose kingdom covered much of what is now Rajasthan and Haryana, with Delhi and Ajmer as its centres.
They met at Tarain, north of Delhi, in 1191.
Prithviraj won. The Indian centre held, the Ghurid army broke, and Muhammad Ghori was wounded and carried from the field.
Then Prithviraj let him go. The Ghurid army withdrew to Afghanistan, and there was no pursuit.
5.3 — Tarain, 1192
Ghori came back the following year, to the same place, against the same opponent.
This time he won completely. Prithviraj's army was destroyed. Prithviraj himself was captured and killed. Within a few years Ghurid forces held Delhi, Ajmer and much of the Ganges plain.
Same ground, same commanders, one year apart, opposite results. This is close to a controlled experiment, and historians have proposed several explanations.
The Ghurid army relied on light horse archers: fast, mobile, shooting from the saddle and refusing to close. Against them, the Indian army's strength — heavy infantry, war elephants and armoured cavalry in a solid line — becomes a weakness. You cannot catch what will not stand and fight, and you take casualties all day.
The classic technique is the feigned retreat: pretend to break, draw the enemy into a disordered pursuit, then turn and destroy them. Sources describe Ghori using exactly this in 1192. It works only if the enemy has not seen it before — which is precisely the difference a year of preparation makes.
Ghori's army was a paid, permanent, professional force under one command, and it fought the same way every time.
Prithviraj's was a coalition of allied kings, each bringing his own contingent under his own commander, each with his own interests. Such an army fights well when things go well and comes apart when they do not, because there is no single chain of command to hold it together under pressure.
The most consequential decision may have been made after the first battle, not during the second.
Prithviraj won and did not pursue. Ghori was allowed to withdraw, reach safety, rebuild and return with a reformed army and a plan.
Why? Probably some mixture of conventions of warfare that treated a defeated enemy's withdrawal as the end of the matter, the practical difficulty of campaigning far from your base, and the more urgent problem of rivals at home. Whatever the reason, it converted a decisive victory into a year's delay.
A caution about the sources. Almost everything we have was written by the winning side's chroniclers or by later Indian bardic poetry composed centuries afterwards, which turned Prithviraj into a legendary figure and is not reliable as history. The details of both battles — numbers, movements, speeches — should be treated as approximate at best.
What is certain: the first battle was lost by the Ghurids, the second was won by them, and northern India had a new ruling order within a decade.
5.4 — What followed
Muhammad Ghori was assassinated in 1206 — at Damyak, on the Jhelum, in Punjab. He had no son.
His most capable commander in India was a man called Qutb al-Din Aibak, who had begun life as a slave-soldier. With Ghori dead, Aibak took power over the Indian territories and had himself proclaimed ruler at Lahore.
That is the beginning of the Delhi Sultanate, which would rule northern India in one form or another for over three hundred years — and it began with a former slave being crowned in a Punjab city.
The standard framing of this period is a question: "Why did India fall?"
Two assumptions hide inside it.
First, that there was an India to fall. Chapter 2 dealt with this. In 1192 there were many kingdoms; Prithviraj's defeat was the defeat of one of them, and several others carried on independently for a long time afterwards.
Second, and more interesting: that something needs explaining at all. The question assumes India ought to have won, so its losing requires a special cause — decadence, disunity, moral failure, some flaw in the civilisation.
But armies with better tactics and unified command beat armies without them, everywhere, always. The mounted archer beat heavy formations across Eurasia for centuries — in Iran, in Russia, in China, in Hungary, in the Islamic world itself. India was not singled out by history. It was subject to the same military logic as everyone else.
What makes the "why did we fall" framing so persistent is that it is not really a historical question. It is an attempt to find a moral lesson in a defeat, usually one that supports whatever the person asking already wants to argue about the present.
The historical answer is duller and better: a professional army with a superior tactical system, under one commander, defeated a feudal coalition — twice, once losing and once winning, which is exactly what you would expect.
The Ghurids replaced the Ghaznavids and came to conquer rather than to raid.
Tarain 1191: Prithviraj Chauhan won, and let the defeated army withdraw. Tarain 1192: Muhammad Ghori won completely.
What changed: mounted archers and the feigned retreat against heavy formations; a professional army under one command against a coalition of allied kings; and a year in which the loser of 1191 was allowed to rebuild.
Ghori was assassinated in 1206 on the Jhelum. His slave-soldier commander Qutb al-Din Aibak was crowned at Lahore — the beginning of the Delhi Sultanate.
And "why did India fall?" is the wrong question twice over: there was no single India, and better armies beating worse ones needs no special explanation.
6Punjab as the Frontier
For three hundred years the Delhi Sultanate ruled northern India, and for most of that time Punjab was its shield. Being a shield is not a compliment. This chapter is about what it costs.
6.1 — The Sultanate in one page
Five dynasties ruled from Delhi between 1206 and 1526. You do not need to memorise them, but the shape matters.
| Dynasty | When | What it is known for |
|---|---|---|
| Mamluk (Slave) | 1206–1290 | Founded by Aibak. Consolidation, and the first Mongol crisis under Balban. |
| Khalji | 1290–1320 | Alauddin Khalji: expansion into the south, price controls, and the defeat of the largest Mongol invasions. |
| Tughlaq | 1320–1414 | The greatest extent, then decline. Ends with Timur. |
| Sayyid | 1414–1451 | Weak, small, picking up the pieces after Timur. |
| Lodi | 1451–1526 | Afghan dynasty. Partial recovery. Ends when Babur arrives — Part 7. |
Iqta: the core administrative device of the Sultanate. Instead of paying an officer a salary in cash, the sultan assigned him the right to collect the revenue of a district, out of which he maintained troops.
It solves a real problem — how to pay an army when you have land but little coin. It also creates one: the holder starts behaving as though the land is his. Nearly every medieval state that used a system like this had the same trouble, and much of the Sultanate's internal politics is sultans trying to stop iqta-holders becoming hereditary lords.
6.2 — Eighty years of Mongols
From the 1220s the Mongols were the central fact of Punjab's existence.
Genghis Khan's forces reached the Indus in 1221, pursuing a fleeing enemy. They did not stay. But over the following century, Mongol armies came through the passes again and again — raiding, occupying, sometimes reaching the walls of Delhi itself.
Punjab was in the way of every one of them.
The Sultanate's response hardened over time. One sultan, Balban, is described as deliberately devastating the frontier — destroying forts, clearing forest, wrecking anything that could shelter or supply an invading force. His own son was killed fighting Mongols near Multan.
Later, Alauddin Khalji met the largest invasions head-on and defeated them, and afterwards rebuilt the frontier fortifications properly.
Think about what a frontier policy looks like from a village.
When a Mongol army comes, your grain is taken, your animals are driven off, your people are killed or enslaved, and your village may be burned.
When your own government's army comes, your grain is requisitioned to feed it, your carts are taken for transport, and your young men may be conscripted.
And when your government decides on a scorched frontier policy, your own side destroys your village so the enemy cannot use it.
Multan and Lahore were sacked repeatedly across this period. Between the sackings, they were garrison towns paying for garrisons.
"Punjab was the frontier province" is a sentence in a textbook. On the ground it meant three centuries of being the place where other people's wars happened.
6.3 — Timur, 1398
The worst single event of the period came at the end of it.
Timur — known in Europe as Tamerlane — ruled a vast empire from Samarkand. In 1398 he invaded India.
His route ran straight through Punjab. He crossed the Indus, took Multan, moved through Dipalpur and the eastern Punjab towns, and reached Delhi in December 1398.
What he did at Delhi is one of the worst massacres in the recorded history of South Asia. The city was sacked, an enormous number of people were killed, and tens of thousands were carried off as slaves. Timur's own account, dictated afterwards, describes the killing without apology.
Then he left. He was not interested in ruling India; he had an empire elsewhere.
The Sultanate never fully recovered. The Tughlaq dynasty ended within a few years. What followed was a much smaller, weaker state, and northern India fragmented again for a century.
Our main sources for Timur's invasion are his own court histories — accounts commissioned by Timur, written to glorify him.
That produces an unusual evidential situation. Normally you discount a chronicle's claims of slaughter as boasting. Here the boasting is the slaughter: the texts describe massacres approvingly, as achievements.
So the direction of bias runs towards exaggeration, not concealment. The numbers given are almost certainly inflated, as medieval numbers nearly always are — but the events themselves are being reported by people with every incentive to report them, which makes the fact of the massacre about as well established as anything in this period.
What we cannot get from these sources: anything about the victims. Not their names, not their number, not what happened to the survivors or the region afterwards. The chroniclers were not interested, and nobody on the other side was in a position to write.
6.4 — What a frontier province becomes
Three centuries of this left marks that outlast the events.
A militarised society. Where raids are normal, every community needs armed men. Local strongmen, fortified villages and warrior clans become permanent features rather than emergencies. Punjab's later reputation as a recruiting ground — which Part 11 shows the British exploiting on an industrial scale — has roots here.
Weak central control. A region that is repeatedly overrun cannot be closely governed. Real power sat with local chiefs who could defend their own patch and who paid the distant capital when it was strong and ignored it when it was not.
Movement of people. Warfare, devastation and slave-taking moved populations around — and so, in the other direction, did the arrival of soldiers, administrators, traders and religious teachers from Central Asia, Afghanistan and Iran, many of whom settled.
And a religious opening. This last one matters most for Chapters 7 and 8. Where the old order had been broken — old temples plundered, old patronage networks destroyed, old elites displaced — there was room for new institutions. The Sufi hospices of the next chapter grew in exactly that space.
Five dynasties ruled from Delhi between 1206 and 1526. Punjab was the frontier province for almost all of it.
The Mongols came through the passes repeatedly for eighty years. Punjab was in the way every time — and one sultan's answer was to deliberately devastate his own frontier so invaders could not live off it.
Timur invaded in 1398, marched through Punjab, and sacked Delhi with enormous slaughter. He left immediately. The Sultanate never recovered.
Three centuries as a frontier left Punjab militarised, loosely governed, with a shifting population — and with space for new religious institutions to grow.
7The Sufis
While sultans fought over Delhi, a different kind of institution was spreading through Punjab's countryside — with an open door, a free kitchen, and a teacher who wrote verses in the local language. One of those verses is the first Punjabi anybody can still read.
7.1 — What Sufism actually is
Sufism is the inward, mystical strand of Islam. Not a separate religion, not a sect — a way of practising the same faith with the emphasis in a different place.
The core idea, in the plainest terms: religion is not only about doing the right things. It is about changing what you are. Rules and rituals matter, but the goal is direct experience of God, and that requires working on yourself — stripping away ego, desire and self-importance until nothing is left between you and the divine.
The methods are practical: repeated recitation of God's names, meditation, fasting, poetry and music, and above all submitting yourself to the guidance of a teacher who has travelled the road before you.
Pir / Shaikh / Murshid: a Sufi master or spiritual guide.
Murid: a disciple — someone who has placed themselves under a pir's instruction.
Silsila: literally "chain". A Sufi order — a lineage of teachers running back through the generations. The main ones in South Asia are the Chishti, Suhrawardi, Qadiri and Naqshbandi.
Khanqah: a Sufi lodge or hospice. Where the pir lived, taught, fed people and received visitors.
Dargah: the shrine built at a Sufi saint's tomb, which becomes a place of pilgrimage after his death.
7.2 — The khanqah
The institution matters more than the theology for understanding what happened in Punjab.
A khanqah was not a mosque. It was a compound where a teacher lived with his disciples, and it did several things at once:
- It fed people. Most khanqahs ran a free kitchen, open to anyone who came, without questions about who they were. Travellers, the poor, the sick, anybody.
- It sheltered travellers. On roads with no inns and real danger, this was not a small thing.
- It taught. Disciples lived there for years.
- It arbitrated. Disputes were brought to the pir, whose judgement carried weight because his authority did not come from the state.
- It was open. The Chishti order in particular made a point of accepting visitors regardless of religion or caste, and of refusing money and land from rulers.
Picture a Punjab village in 1250. The nearest sultanate official is far away and interested in you only at harvest. The local strongman is a fact of life. There is no hospital, no school, no court you can afford, and no help if the crop fails.
Now a khanqah appears a day's walk away. Anyone can go. There is food. The man who runs it takes nothing from the government, listens to anybody, and settles disputes without a bribe.
You do not need a theological argument to explain why people went.
And note what this means for Chapter 8. Whatever else the Sufis were, they were the only institution in much of the countryside that gave something to ordinary people and asked nothing about their birth.
7.3 — Punjab's two great figures
Bahauddin Zakariya settled at Multan and died there around 1262. He belonged to the Suhrawardi order, which took a different line from the Chishtis: it accepted royal patronage, held property, and engaged with political power rather than avoiding it. Multan under him became a major centre of learning, and remains a city of shrines.
And then there is the figure who matters most for everything that follows in this series.
7.4 — Baba Farid
Fariduddin Masud, known as Ganjshakar — "treasury of sugar" — lived roughly from the 1170s to 1266. He was a Chishti, trained in Delhi, and he settled at a small town then called Ajodhan, on the Sutlej. The town is now called Pakpattan and is named after him.
He is universally known as Baba Farid.
He did three things that changed Punjab.
One: he stayed in the countryside. He turned down the great centres and settled in a small town among ordinary farming people, and made his khanqah the centre of the region.
Two: he composed in Punjabi. Not Persian, not Arabic — the language people around him actually spoke. His verses are short, plain and hard: about hunger, old age, death, patience, and the futility of pride.
Three: his words entered the Sikh scripture. Over three hundred years later, Guru Arjan compiled the Guru Granth Sahib and included Farid's compositions in it — four hymns and a large body of couplets.
Sit with what that means.
The oldest Punjabi anybody can still read was composed by a Muslim Sufi in a small town on the Sutlej in the thirteenth century.
And it survives because Sikhs preserved it in their scripture, where it is recited daily, today, alongside the words of the Gurus.
Part 5 ended with Punjab having lost two entire writing traditions — the Harappan script and Kharosthi. Here is where the third one starts, and it starts with a poet who wrote for farmers in their own speech.
Whenever anybody tells you Punjab's religious traditions are separate streams, this is the fact to put in front of them. Not as an argument that everything was harmonious — Chapters 3 and 6 show it plainly was not — but because it is true and it is in the scripture.
An honest complication, and a good example of how careful scholarship works.
Sikh tradition attributes the verses to Baba Farid, and they are placed under his name in the scripture. The content, imagery and religious outlook fit a thirteenth-century Chishti of his kind. The connection between the Gurus and the Pakpattan lineage is well attested — Guru Nanak is recorded as visiting.
The language of the verses looks closer to sixteenth-century Punjabi than to thirteenth-century speech. Farid's successors at Pakpattan inherited his name and title, and Guru Nanak is recorded meeting one of them — a man who held the seat generations later and was also called Farid. Verses composed by a successor could easily have been transmitted under the founder's name, which is entirely normal in devotional traditions.
What is not in dispute, and matters far more: the verses are Chishti Sufi compositions in Punjabi from the Pakpattan lineage, and they are in the Guru Granth Sahib. Whether the man who first spoke them died in 1266 or in 1554 does not change what that fact says about how Punjab's traditions grew into each other.
And notice the method. The doubt comes from linguistic dating — the same technique Part 3 used on the Rigveda. Language changes at a rate, and that rate can be read.
Sufism is the inward strand of Islam — the idea that religion is about changing what you are, not only doing the right things.
The khanqah was the institution that mattered: a lodge with a free kitchen and an open door, feeding and sheltering anyone, settling disputes, and taking nothing from the state.
Bahauddin Zakariya made Multan a centre. Baba Farid (d. 1266) settled at Pakpattan on the Sutlej.
Farid composed in Punjabi — the oldest Punjabi anyone can still read — and his verses are in the Guru Granth Sahib.
Whether every verse is his own or a successor's is genuinely argued. That it is Chishti Sufi Punjabi poetry inside the Sikh scripture is not.
8How Punjab Became Partly Muslim
This is the question the whole part has been building toward. There are four standard answers. One of them can be tested against a map — and when you do, it fails badly.
8.1 — What has to be explained
By the twentieth century, Punjab as a whole had a Muslim majority. The distribution mattered enormously: the western districts were overwhelmingly Muslim, the eastern districts had Hindu and Sikh majorities, and that line is essentially where Punjab was cut in two in 1947.
How did that happen?
Note first what it is not. It is not a single event. It is a change spread over roughly eight hundred years, happening at different speeds in different districts, to people who mostly left no record of why they did it.
8.2 — The four explanations
Four explanations are in circulation. They are not mutually exclusive, and the useful thing is to test each against evidence rather than to argue from principle.
The claim: people converted because the alternative was death, or because the pressure of living under Muslim rule made it impossible not to.
For: there was real violence in this period. Chapters 3 and 6 describe it. There were periods of pressure on non-Muslims, discriminatory taxation, and destruction of temples. Coercion existed and should not be minimised.
Against — and this is decisive: the theory makes a prediction, and the map contradicts it. See 8.3.
The claim: the Muslim population is largely descended from soldiers, administrators, traders and religious figures who came from Central Asia, Afghanistan and Iran and settled.
For: such migration certainly happened, and some Punjabi Muslim families do trace descent from it.
Against: the numbers do not work. Immigration on the scale required to produce tens of millions of Muslims is far beyond what the sources suggest. Immigrants were concentrated in towns; the Muslim majority was overwhelmingly rural.
The claim: people converted to get jobs, land, tax relief and access to the ruling class.
For: plainly true for some. This is how elites usually behave under a new order, and it happens everywhere.
Against: it explains the wrong people. Patronage reaches those near power — courtiers, officials, urban notables. It does not reach a farmer in a village a hundred kilometres from anywhere, and that farmer is who needs explaining.
The claim: people at the bottom of the caste order converted to a religion that declared all believers equal.
For: intuitively powerful, and it is a real theme in Sufi and later reformist preaching.
Against: the evidence is weak where it should be strongest. Conversion was not concentrated among the lowest castes; whole regions changed, cutting across caste. And converts frequently kept their caste identity afterwards — Muslim communities in Punjab retained clan and caste distinctions, and marriage patterns largely followed them. If escape from caste were the motive, it worked poorly.
8.3 — The test that breaks the sword theory
This is the most useful piece of reasoning in the chapter, and it takes only a map.
If conversion happened by force, then Islam should be strongest where Muslim political power was strongest, longest and closest. That is what force means: the more of it applied, and the longer, the greater the effect.
Now look at where Muslim political power actually was. The heartland of the Delhi Sultanate and later the Mughal empire was the upper Ganges plain — Delhi, Agra, the Doab. That is where the capital sat, where the armies were, where the administration was thickest, and where the rule lasted longest and most continuously.
So where should the Muslim majorities be?
They are not there.
| Region | Result by the twentieth century |
|---|---|
| Upper Ganges plain — the centre of Muslim political power for six centuries | Large Hindu majority. Muslims a substantial but clear minority. |
| Western Punjab — frontier | Overwhelmingly Muslim. |
| Sindh — frontier | Overwhelmingly Muslim. |
| Eastern Bengal — the far end of the empire, roughly as distant from Delhi as it is possible to be | Overwhelmingly Muslim. |
| Kashmir — mountainous, peripheral | Overwhelmingly Muslim. |
Read that table again, because it is close to an experiment.
The Muslim majorities formed at the edges. The Hindu majorities held in the centre.
If the sword were the mechanism, this is exactly backwards. Six hundred years of Muslim rule from Delhi, with the full weight of the state, produced a Hindu-majority region. The far end of Bengal, which armies barely reached, went overwhelmingly Muslim.
A theory that predicts the opposite of what happened is not a weak theory. It is a wrong one.
This does not mean no one was ever forced. Individual coercion certainly occurred. It means force cannot be the main mechanism, because the pattern it would have produced is not the pattern we observe.
8.4 — What the pattern suggests instead
If conversion clusters at the frontiers rather than the centre, the explanation has to be something that frontiers have and centres lack.
The account most historians now find persuasive works like this.
Frontier regions were where new land was being brought under the plough. Forest cleared, marsh drained, pasture converted to fields. This was happening on Punjab's western reaches and, most dramatically, in eastern Bengal.
That work needed organisers. Someone had to gather labour, get authority over the land, provide seed and tools, and settle disputes. In practice, the people who did this were often religious figures — a Sufi granted land by a ruler, who built a khanqah, attracted followers, and turned wilderness into cultivated fields and a village.
So the shrine became the centre of the new community — not as a conquest, but as the institution the village was built around. It fed people, arbitrated, and marked the calendar.
And religious identity followed slowly. Not a moment of conversion but a drift across generations. First the local holy man is a Sufi, and you go to him as you would go to any holy man. Then his festivals become your festivals. Then his name is in your family's prayers. Then your grandchildren describe themselves as Muslim, while still keeping local customs, local shrines and local caste identity.
Syncretism: the blending of elements from different religious traditions into one practice. A useful word, but treat it carefully — it implies two clean systems mixing, when what usually happens is people simply doing what their neighbours do without classifying it.
The whole question is usually framed as: "why did people convert?" That framing carries an assumption which does not fit what happened.
The assumption is that conversion is an event — a moment when a person stops being one thing and becomes another.
That is how conversion works in some traditions and some periods. It is not how this worked.
What the evidence suggests is a process spread across generations, in which no single person ever "converted" at all. A family goes on doing most of what it did, adds some practices, drops others, and after four or five generations answers a different word when a census-taker asks. Nobody in that chain experienced a moment of change.
And this is why the census matters so much to how we see it. The British counted people by religion, forced everyone into one box, and published the totals. A sharp line appeared on paper where the reality had been gradual and overlapping. Part 12 shows what happened when that paper line became a political one.
So a better question than "why did people convert?" is: "how did a landscape of local practices, holy men and overlapping customs turn into two counted populations?" The answer to that involves census-takers and politicians as much as it involves Sufis.
By the twentieth century western Punjab was overwhelmingly Muslim and eastern Punjab was not — and that line is roughly where Punjab was cut in 1947.
Four explanations are offered: force, immigration, patronage, escape from caste. Each fits some people; none fits the pattern.
The sword theory fails a simple test. Islam ended up strongest at the frontiers — west Punjab, Sindh, east Bengal, Kashmir — and weakest in the Gangetic heartland where Muslim power was strongest and longest. Force predicts the opposite of what happened.
The better explanation is agrarian: frontier land being cleared and settled, often organised around Sufi shrines, with religious identity following slowly over generations.
And "conversion" is the wrong word. Almost nobody converted. Families drifted, over centuries — and a sharp line only appeared when the British started counting.
9The World Nanak Was Born Into
By the late fifteenth century Punjab had a language beginning to be written, a Persian-speaking government, Sufi shrines across the countryside, yogis in the hills, and poets from the lowest castes attacking both priesthoods at once. Then, in 1469, a child was born near Lahore.
9.1 — The languages of Punjab in 1469
A person in fifteenth-century Punjab lived among several languages, each with its own job.
- Punjabi — what people actually spoke, descended from the Prakrit and Apabhramsha of Chapter 1. Beginning to be written, but with no settled script and no official status.
- Persian — government, law, revenue records, courts, high poetry. The language of getting on.
- Arabic — Islamic scripture and religious scholarship.
- Sanskrit — Hindu learning, ritual and scholarship, understood by a small educated group.
- Braj and other regional speech — used by devotional poets and travelling singers, understood across a wide area.
Note the gap in that list. The language ordinary people spoke had no script of its own, no official use, and almost no literature. What existed was verse composed to be sung and remembered — Farid's couplets, and the songs of wandering singers.
Gurmukhi, the script Punjabi is written in today, did not yet exist. It comes in Part 7.
9.2 — The yogis
Alongside the Sufi shrines, another spiritual tradition was strong in Punjab and the hills.
The Nath yogis traced themselves to a teacher called Gorakhnath. They practised severe physical and breathing disciplines aimed at mastering the body and achieving liberation. They wore distinctive earrings, travelled constantly, lived apart from ordinary society, and had a reputation for supernatural power.
They were not a fringe. They were an established, widely respected and somewhat feared presence, and they had their own critique of ritual religion — that priests, temples and scriptures were beside the point, and that what mattered was direct internal work.
Guru Nanak would later hold a famous recorded dialogue with Nath yogis, which tells you how significant they were: they were the group whose position had to be answered.
9.3 — The Bhakti poets
And a third current was moving through northern India, sharper than either.
Bhakti: devotion. A broad religious movement across India, over many centuries, holding that a direct loving relationship with God is what matters — and that priests, ritual, scripture-learning and caste status are not required and may get in the way.
Its northern poets were, strikingly often, not from priestly or elite backgrounds:
- Namdev, from Maharashtra, a calico-printer by community.
- Kabir (roughly 1440–1518), from Varanasi, a weaver, raised in a Muslim weaving community — who attacked Hindu and Muslim religious authority with equal ferocity, mocking the pandit and the mullah in the same verse.
- Ravidas, a leatherworker — from a community treated as untouchable.
They composed in ordinary spoken language, not Sanskrit or Persian. Their verses travelled by being sung.
And the compositions of all three are in the Guru Granth Sahib. So are Farid's. When Guru Arjan compiled the scripture in Part 7, he included the work of a Muslim Sufi, a Maharashtrian calico-printer, a Varanasi weaver and a leatherworker, alongside the Gurus.
Look at the room Nanak was born into.
A Persian-speaking government that most people could not talk to. Sufi shrines in the countryside feeding anyone who came. Nath yogis saying ritual is worthless and only inner work counts. Bhakti poets — a weaver, a leatherworker — saying God does not care about your caste or your scripture, and saying it in language a farmer could understand. And a spoken language with no script and almost no written literature.
Every single one of those elements turns up in what Nanak did. Composition in Punjabi. Rejection of ritual and caste. A free kitchen open to all. Engagement with both Hindu and Muslim authority and submission to neither. A scripture that includes voices from outside his own tradition.
This does not diminish him. It locates him. Original people do not appear from nowhere; they appear where the materials are, and do something with them that nobody else did.
9.4 — 1469
In 1469, in a village called Talwandi, not far from Lahore, a child was born to a Hindu family of the accountant caste, in a district administered by a Muslim government, in a countryside dotted with Sufi shrines, in a province that had been a frontier for three centuries.
He was named Nanak.
Part 7 begins here.
In 1469 Punjab used Punjabi to speak, Persian to govern, Arabic and Sanskrit for religion — and Punjabi had no script and almost no literature.
Nath yogis taught that ritual is worthless and only inner discipline counts. Nanak would later have to answer them directly.
Bhakti poets — Namdev, Kabir, Ravidas, from weaving and leatherworking communities — attacked Hindu and Muslim religious authority alike, in ordinary language. All three are in the Guru Granth Sahib, along with Baba Farid.
Nanak was born in 1469 near Lahore, into exactly this room. Part 7 begins there.
10An Honest List of What We Do Not Know
Nine centuries, and for large stretches of them the record is almost silent. Here is where that leaves us.
Genuinely unknown
- Almost everything about Punjab between 550 and 900 CE. Four and a half centuries with no substantial local records.
- The size of the population at any point in this part, and how much it fell after Timur.
- How many people became Muslim, when, and where — in any detail. We have twentieth-century census figures and almost nothing before.
- Why any individual family changed religion. Not one ordinary person left an account of it.
- Whether the Farid verses in the Guru Granth Sahib are Baba Farid's own or a successor's.
- What actually happened at either battle of Tarain in any tactical detail. The sources are the winners' chronicles and later legend.
- How much Timur actually killed. The figures come from texts written to glorify him.
- What ordinary religious practice looked like in a Punjab village at any point in these nine centuries.
- How Punjabi actually developed between Apabhramsha and Farid's verses. The intervening stages are unrecorded.
- Whether the destroyed temples were rebuilt, and how communities recovered afterwards.
Solid
- The Gupta order collapsed by around 550 CE and northern India fragmented.
- Xuanzang found the Buddhist monasteries of the north-west largely ruined by around 630–645 CE.
- The Hindu Shahis held the north-west from the ninth century, retreating from Kabul to the Indus to the Salt Range before ending in the 1020s.
- Mahmud of Ghazni campaigned into India roughly seventeen times between 1000 and 1027, sacked Somnath in 1025–26, and annexed Punjab in 1021.
- Al-Biruni came with the Ghaznavids, learned Sanskrit, and wrote the first serious outside study of Indian civilisation.
- Lahore was a Ghaznavid provincial seat from 1021 and the dynasty's capital from around 1163 to 1186.
- Persian became the language of government in Punjab and remained so until the British replaced it in the 1830s.
- Ali Hujwiri died at Lahore around 1072 and wrote the first Persian treatise on Sufism.
- Prithviraj Chauhan defeated Muhammad Ghori at Tarain in 1191; Ghori won there in 1192.
- Qutb al-Din Aibak was proclaimed ruler at Lahore in 1206, beginning the Delhi Sultanate.
- Mongol incursions struck Punjab repeatedly through the thirteenth and early fourteenth centuries.
- Timur invaded through Punjab in 1398 and sacked Delhi with enormous slaughter.
- Bahauddin Zakariya established a Suhrawardi centre at Multan; Baba Farid a Chishti one at Pakpattan.
- Verses attributed to Farid are in the Guru Granth Sahib, alongside Namdev, Kabir and Ravidas.
- By the twentieth century, Muslim majorities were concentrated at the geographical peripheries of Muslim political power, not at its centre.
- Guru Nanak was born in 1469.
Notice something about the two lists. Nearly everything in the solid column is about rulers, battles, dynasties and named religious figures. Nearly everything in the unknown column is about ordinary people.
That is not a coincidence and it is not laziness. It is what the evidence is. Court chronicles, dynastic inscriptions, the writings of religious scholars, the accounts of foreign travellers — every surviving source for these nine centuries was produced by someone with power, money or an institution.
So we can say with confidence who ruled Lahore in 1170, and we cannot say what a farmer twenty kilometres away believed, ate, feared or hoped.
The single largest change described in this part — millions of people gradually coming to describe themselves as Muslim — happened almost entirely below the level at which anything was written down. Chapter 8's explanation is the best available reconstruction. It is a reconstruction.
When someone tells you with certainty why the people of medieval Punjab did what they did, remember: those people left no record at all, and everyone claiming to speak for them is inferring.
Sources & further reading — Part 6
Timeline of Part 6
Dates before about 1000 CE are approximate. Where a date is disputed, the dispute is in the chapter.
| When | What happened |
|---|---|
| c. 550 CE | The Gupta empire fragments. Where Part 5 ended. |
| 606 – 647 CE | Harsha rules from Kanauj. Punjab on the edge. His kingdom dissolves at his death. |
| c. 630 – 645 CE | Xuanzang travels from China through India and finds the Buddhist monasteries of the north-west largely in ruins. |
| 8th century | Lalitaditya of Kashmir dominates much of the Punjab plain. |
| 8th – 10th centuries | The tripartite struggle for Kanauj between Palas, Pratiharas and Rashtrakutas. Punjab outside it. |
| c. 850s CE | The Hindu Shahis take power in the Kabul region and the north-west. |
| 962 – 997 CE | Ghazni becomes an independent power under Alptigin, then Sabuktigin. |
| 998 – 1030 CE | Mahmud of Ghazni rules. Around seventeen campaigns into India. |
| 1001 CE | Jayapala defeated at Peshawar. The Shahi retreat begins. |
| c. 1008 CE | Anandapala's coalition of northern Indian kingdoms is defeated. |
| 1021 CE | Punjab annexed. Lahore placed under a Ghaznavid governor. The permanent change. |
| 1025 – 26 CE | Somnath sacked. |
| c. 1030 CE | Al-Biruni completes his study of India, having learned Sanskrit to write it. |
| c. 1072 CE | Ali Hujwiri (Data Ganj Bakhsh) dies at Lahore. His Kashf al-Mahjub is the first Persian treatise on Sufism. |
| c. 1150 CE | The Ghurids sack Ghazni. The Ghaznavids lose their homeland. |
| c. 1163 – 1186 CE | Lahore is the Ghaznavid capital — Punjab's first real centre in over a thousand years. |
| c. 1173 – 1266 CE | Baba Farid, Chishti Sufi of Pakpattan. Composes in Punjabi. |
| 1186 CE | Muhammad Ghori ends the Ghaznavids at Lahore. |
| 1191 CE | First Battle of Tarain. Prithviraj Chauhan defeats Ghori and lets him withdraw. |
| 1192 CE | Second Battle of Tarain. Ghori wins completely. |
| 1206 CE | Ghori assassinated on the Jhelum. Qutb al-Din Aibak proclaimed at Lahore — the Delhi Sultanate begins. |
| 1221 CE | Genghis Khan's forces reach the Indus. Eighty years of Mongol pressure on Punjab begin. |
| c. 1262 CE | Bahauddin Zakariya dies at Multan, having made it a Suhrawardi centre. |
| 1290 – 1320 CE | Khalji dynasty. Alauddin Khalji defeats the largest Mongol invasions and rebuilds the frontier. |
| 1398 CE | Timur invades through Punjab and sacks Delhi. The Sultanate never recovers. |
| 1414 – 1451 CE | Sayyid dynasty — a much smaller state. |
| 1451 – 1526 CE | Lodi dynasty. Partial recovery. |
| 1469 CE | Guru Nanak born at Talwandi, near Lahore. Part 7 starts here. |
Glossary
Every hard word used in Part 6, in plain English.
| Word | What it means |
|---|---|
| Apabhramsha | The late Prakrit stage, roughly 600–1200 CE, out of which Punjabi, Hindi, Bengali and the other modern North Indian languages grew. |
| Bhakti | Devotion. A movement holding that a direct loving relationship with God matters more than ritual, scripture-learning or caste. |
| Chishti | The Sufi order of Baba Farid. Known for refusing state patronage, keeping an open door and running free kitchens. |
| Dargah | The shrine at a Sufi saint's tomb, which becomes a place of pilgrimage. |
| Ghaznavids | The Turkish dynasty of Ghazni in Afghanistan. Mahmud's family. Annexed Punjab in 1021 and later ruled from Lahore. |
| Ghurids | The dynasty from central Afghanistan that destroyed the Ghaznavids and conquered northern India under Muhammad Ghori. |
| Iqta | The right to collect a district's revenue, granted to an officer instead of a salary, out of which he maintained troops. |
| Khanqah | A Sufi lodge. Where the teacher lived, taught, fed anyone who came, and settled disputes. |
| Mamluk | A slave-soldier. Also the name of the first Delhi Sultanate dynasty, founded by one. |
| Murid | A disciple of a Sufi teacher. |
| Nath | An order of yogis tracing to Gorakhnath, practising severe physical and breathing disciplines. Widely influential in Punjab. |
| Persian (Farsi) | The language of Iran, and the prestige language of government and poetry across medieval Central and South Asia. Punjab's official language for eight hundred years. |
| Pir / Shaikh / Murshid | A Sufi master or spiritual guide. |
| Silsila | Literally "chain". A Sufi order — a lineage of teachers. |
| Slave-soldier | A boy bought or captured and trained as an elite professional soldier. Considered loyal because he had no local faction. Several became rulers. |
| Sufism | The inward, mystical strand of Islam. Emphasises direct experience of God and working on the self, under a teacher's guidance. |
| Suhrawardi | The Sufi order of Bahauddin Zakariya at Multan. Unlike the Chishtis, it accepted royal patronage and property. |
| Sultan | Arabic for "authority". A ruler's title across the Islamic world. |
| Syncretism | The blending of elements from different religious traditions. Use carefully — it implies two clean systems mixing, which is rarely how it feels from inside. |