By 1453 the Roman Empire consisted of the city of Constantinople, a few suburbs, some islands and a fragment of southern Greece. Its population had fallen from perhaps half a million at its height to something in the range of forty to fifty thousand.
Inside the walls were fields, orchards and abandoned quarters. The great buildings stood but many were unrepaired. The imperial treasury was empty and had been for a long time; emperors had spent the previous century travelling to western Europe in person to beg for help, offering the one thing they had, which was the reunion of the churches.
That offer was made and formally accepted at a council in Florence in 1439, and it destroyed what was left of the government's standing at home. The population regarded submission to Rome as a betrayal of everything the city had endured for, and a senior official is reported to have said he would rather see the Turkish turban in the city than the Latin mitre. The help, when it came, amounted to a few hundred men.
Against that, the Ottoman state was the most effective military power in the region, held the territory on both sides of the city, and had a twenty one year old sultan who had decided that taking it was the point of his reign.
The walls of chapter eight had held for a thousand years against everything, and the thing that finally answered them had existed in Europe for barely a century.
Mehmed the Second assembled a siege train including several very large bombards, the largest of them cast by a Hungarian engineer who, according to a Greek account, had offered his services to the emperor first and been told the city could not pay him. The great gun is said to have thrown a stone ball of several hundred kilograms, to have needed dozens of oxen to move, and to have been able to fire only a handful of times a day because the barrel had to cool.
Guns of that period were unreliable and some burst. What they could do, given weeks, was reduce a section of medieval curtain wall to rubble, and no amount of thickness helped, because a wall built to resist rams and mining is a static target for artillery.
The defenders repaired the breaches nightly with earth, timber and barrels, which absorbed shot better than masonry, and this is the reason the siege took seven weeks rather than seven days. But there were perhaps seven or eight thousand defenders for twenty kilometres of fortification, and the arithmetic could not be repaired.
There was one western response, and its smallness is the point. Three Genoese ships and an imperial transport fought their way through a far larger Ottoman fleet into the harbour in April, to great rejoicing, and that was substantially the aid Christendom sent. Venice debated a relief fleet, voted one, and it sailed too late to matter. Every western ruler had a nearer quarrel.
The sultan also solved the harbour. The Golden Horn was closed by a chain, so he had a road of greased timbers built across the hill behind Galata and dragged some seventy ships overland into the harbour behind the boom, which forced the defenders to thin their lines further to watch the sea walls.
The final assault came before dawn on the twenty ninth of May, in three waves: irregulars first to exhaust the defenders, then Anatolian troops, then the sultan's household infantry.
Two things went wrong at once for the defence. A small postern gate appears to have been found unlocked, letting a party of attackers onto the walls, and the Genoese commander of the land defence, a professional who had held the breaches for weeks, was badly wounded and carried out through the lines, which was visible to everyone and broke the defenders' cohesion.
The emperor, Constantine the Eleventh, was last seen fighting on foot at the breach after throwing off his insignia. His body was never certainly identified. It is an ending that would be suspect if a novelist wrote it, and it is attested by several independent accounts.
The city was given over to plunder for the customary period, and the accounts on both sides describe killing and enslavement on a large scale. The sultan entered later, rode to the church of Holy Wisdom, and had it converted to a mosque. He then began, immediately, the work of repopulating and rebuilding his new capital, bringing in settlers from across his territories, restoring the water supply, and appointing a Greek patriarch for the Orthodox population.
It is worth a section on the Ottomans, because European accounts of 1453 tend to present them as a force of nature rather than as a government.
They had begun around 1300 as one of many small Turkish principalities on the Byzantine frontier in western Anatolia, and what distinguished them was position and administration rather than ferocity. They crossed into Europe in the 1350s, took Adrianople, and by the time of the siege had ruled large parts of the Balkans for three generations. Constantinople was not a foreign conquest arriving from outside; it was an enclave inside a state that already surrounded it.
Two institutions did most of the work. The first was a system of granting the revenue of land to cavalrymen in return for service, which supported a large army without a cash budget and which resembles, in its logic, the arrangements of chapter thirteen. The second was a standing infantry corps paid in cash, drilled, and by this period equipped with firearms, recruited by a levy of boys from Christian villages who were converted, trained and made the sultan's personal servants. That practice was resented and evaded by the families it fell on, and it also became a route to the highest offices of the empire, several grand viziers having entered it that way.
Ottoman rule over its Christian and Jewish subjects worked on the same principle described in chapter ten: legally subordinate protected communities, taxed at a higher rate, running their own religious and family law under their own leaders. When Spain expelled its Jews in 1492 a large number were received into Ottoman territory, and the sultan of the day is reported to have remarked on the wisdom of a king who impoverished his own kingdom to enrich another.
None of that makes the conquest gentle, and the sack was what sacks were. It does mean that what replaced the empire was a functioning multi religious state that governed south eastern Europe for four hundred and fifty years, which is a longer run than most of the alternatives on offer.
The date is one of the standard markers for the end of the Middle Ages, and most of what it is credited with is either wrong or was already happening.
It did not cause the Renaissance by driving Greek scholars west. Greek teachers had been coming to Italy for the best part of a century, invited and paid, and the great transfers of manuscripts and expertise happened before 1453 rather than after it. The refugees who did arrive reinforced an existing movement.
It did not close the trade routes to Asia and thereby cause the voyages of chapter twenty six. That story is durable and does not survive the evidence: Ottoman rulers wanted the customs revenue and kept the routes open, Venetian trade with the Ottoman empire continued and in some years grew, and Portuguese exploration down the African coast had been under way for four decades before the city fell, for reasons of gold and pepper and papal politics that had nothing to do with it.
What it did do was real enough. It removed the last political claim to continuity with Rome in the east, and the imperial title in the east passed, in the eyes of those who wanted it, to the ruler in Moscow, whose grandson would use the word tsar, which is Caesar. It gave the Ottomans a capital, a fleet base and the confidence for another century and a half of expansion, which took them to Belgrade, to Rhodes, to the gates of Vienna in 1529 and again in 1683, and made them a permanent participant in European politics rather than an external threat.
And it ended a state. The emperor who died in the breach was, by an unbroken if much interrupted line of succession, the successor of Augustus. Whatever one thinks of the institution, fifteen centuries is a long time for anything to last.
Almost immediately, the fall became a symbol, and it has been used ever since by people who needed it to mean something.
In western Europe it was read as a catastrophe for Christendom and as a punishment, and it generated a large literature of lament and a series of crusade proposals that came to nothing because no ruler would pay for them.
For Orthodox Christians it became the central fact of a long subjection, and the Russian claim to be the third Rome is built directly on it.
In the nineteenth century, when Greek and Balkan nationalists were constructing modern national histories, 1453 became the beginning of an occupation to be undone, and legends multiplied about the emperor turned to marble and waiting to return.
In Turkey the conquest is a foundational national event, commemorated annually, and the sultan who took the city is a central figure in political rhetoric to this day.
The historian's job here is not to arbitrate between those readings but to notice that all of them are doing present tense work, and that the actual event, a well planned siege carried by artillery against a state that had been dying for two centuries, is a smaller and more explicable thing than any of them.
Firm: the siege. We have multiple eyewitness accounts from different sides, a Venetian ship's surgeon inside the city, Greek observers, Ottoman chronicles, and Genoese correspondence. The chronology, the tactics and the outcome are secure.
Soft: the great gun. Its dimensions and performance come from narrative sources that had reason to be impressed, and the figures given for the weight of shot and the size of the barrel vary considerably.
Soft: the numbers of defenders and attackers, as always.
Soft: the unlocked postern. It appears in one main account and has been questioned; sieges frequently acquire a single small betrayal in the retelling, because it is more satisfying than the truth, which is usually attrition.
Contested: whether 1453 marks anything at all. Periodisation is a convenience, and the alternatives offered for the end of the Middle Ages include 1450 for printing, 1492 for the Atlantic, 1517 for the Reformation and 1500 for nothing in particular. This book uses the date because it genuinely closes a story that began in chapter four, not because anything else changed on that Tuesday.