Ayyubid Sultanate

Saladin

Drafted from Hillenbrand, The Crusades: Islamic Perspectives (1999)formation

Salah al-Din Yusuf ibn Ayyub was a Kurdish officer in the service of Nur al-Din of Damascus, sent to Egypt on his uncle's staff as part of a Syrian intervention in a Fatimid vizierial quarrel. He inherited the vizierate when his uncle died, ended the Fatimid caliphate in 1171, and on Nur al-Din's death in 1174 took Damascus from his late master's heirs.

The order of operations matters and is usually reversed in the telling. Saladin spent roughly twelve years fighting other Muslim rulers — Zangids in Aleppo and Mosul, Fatimid loyalists in Egypt — before he fought the Crusader states seriously. The unification of Egypt and Syria under one ruler is what made Hattin possible in 1187, and Hattin is what delivered Jerusalem after eighty-eight years of Frankish rule.

Hillenbrand's caution is worth carrying: the Saladin who reaches modern readers is substantially a construction of nineteenth-century European romanticism and twentieth-century Arab nationalism, both of which needed him for their own purposes. The contemporary Muslim sources are more mixed, and several of them are written by people who thought a Kurdish officer had usurped his employer's family.

What Saladin did to Egypt

Drafted from Todd, Saladin (2011)formation

Ending the Fatimid caliphate took one Friday. In September 1171 the khutba in Cairo was read in the name of the Abbasid caliph in Baghdad instead of the Fatimid imam-caliph al-Adid, who was dying in his palace a short walk away and is generally supposed not to have been told. There was no rising. Two centuries of Isma'ili rule ended without a battle, which is itself the most interesting fact about it.

What followed was slower and mattered more. The Fatimid palace libraries were dispersed and sold. The Isma'ili teaching establishment — the dar al-ilm, the missionary apparatus that had made Cairo the intellectual capital of a transnational Shi'i movement — was shut down, and in its place Saladin and his officers founded Sunni madrasas and khanqahs, endowed with agricultural revenue in perpetuity. Eddé's point is that this is the durable conquest: within two generations Egypt was a Sunni country with a Sunni legal establishment, and the Fatimid two hundred years survived mainly as architecture.

The fiscal reorganisation ran alongside it. The Fatimid crown lands and the enormous palace establishment were broken up and reassigned as iqta' — revenue assignments to officers in return for service — which converted the wealth of the Nile valley into cavalry. Saladin's family took the largest shares. The customs revenue of Alexandria and the Red Sea ports came with it, and it was Egyptian money, not Syrian, that paid for everything after 1174.

He also built. The citadel on the Muqattam spur, begun in 1176, moved the seat of government out of the Fatimid palace-city and onto a rock above it, and a single wall was projected to enclose Cairo and Fustat together. Later sultans ruled from that citadel for six hundred and fifty years.

Hattin

Drafted from Hillenbrand, The Crusades: Islamic Perspectives (1999)expansion

On 4 July 1187 the field army of the kingdom of Jerusalem marched from Saffuriya across waterless country towards Tiberias, was harassed all day, camped without water, and was destroyed the next morning below the twin hills called the Horns of Hattin. The king of Jerusalem was captured. So was the True Cross, which the army had carried with it and which was never recovered.

The kingdom had almost no one left to garrison anything. Acre fell within a week, then Sidon, Beirut, Ascalon; Jerusalem surrendered on 2 October after a short siege. The terms were a ransom per head, generously and inconsistently applied — Saladin's own brother bought a thousand of the poor their freedom, and a substantial number who could not pay were enslaved anyway. Hillenbrand is careful here, because the contrast with 1099 is the single most-repeated item in the popular literature and it is usually made by people who want one of the two events to be representative of a civilisation.

What Hattin did to the Ayyubids specifically was make Saladin's authority unarguable inside his own house. He had spent twelve years being described by Zangid partisans as a servant who had stolen his master's inheritance, and the charge was hard to answer while he was fighting Aleppo instead of the Franks. After 1187 nobody made it in public. The jihad rhetoric that his chancery had been producing for a decade — his secretaries al-Fadil and Imad al-Din were professionals at it — finally described something that had happened.

The one thing he did not do was take the coast. Tyre held, because Conrad of Montferrat arrived in it by sea at exactly the wrong moment for Saladin, and Tyre was the port through which the counter-attack came.

The war after the victory

Drafted from Hillenbrand, The Crusades: Islamic Perspectives (1999)expansion

The Third Crusade was the price of Hattin. Western Europe sent three kings; two of them arrived. From August 1189 the Franks besieged Acre for two years while Saladin besieged them in turn, an arrangement in which both armies starved and neither could withdraw, and when Acre finally surrendered in July 1191 Richard I had some two thousand seven hundred prisoners executed outside the walls because the negotiated payments were late.

Richard beat Saladin in the field at Arsuf in September and again near Jaffa in 1192, took the coast from Acre to Jaffa, marched inland twice towards Jerusalem and turned back twice without attacking it. The Treaty of Jaffa in September 1192 left the Franks a coastal strip, left Saladin Jerusalem, and guaranteed Christian pilgrims access to it. Saladin died six months later in Damascus, aged about fifty-five, having spent his personal treasury and leaving, by the count of his secretary, one gold piece and forty-odd silver.

The five years after Hattin are the part of the story the romantic version has no use for, and they are the part that explains the polity. Saladin could win a kingdom in a summer and could not hold a siege line for two years, because the army he was holding it with was an iqta' army — officers with revenue assignments in Egypt, Syria and the Jazira who expected to go home for the winter, and cousins with their own territories who could and did leave. The force that took Jerusalem was a coalition of his relatives, and every campaign after 1189 was negotiated with them again.

His reputation among his enemies survived all of it. Hillenbrand's point about the European Saladin — the courteous knight, the Dante placement in Limbo, the Walter Scott figure — is that it was already forming in Frankish writing during his lifetime, and that it says considerably more about what the West needed a noble Saracen for than about the man his own chroniclers argued over.

A family, not a state

Drafted from Humphreys, From Saladin to the Mongols: The Ayyubids of Damascus 1193-1260 (1977)aside?What aside means

Saladin divided his territories among his sons and brothers, and the Ayyubid lands were never afterwards a single polity. Egypt, Damascus, Aleppo, Hama, Homs and the Jazira were held by different members of the house, each with his own army and revenue, acknowledging a senior sultan whose seniority was frequently disputed and occasionally fought over.

This was a deliberate Turco-Iranian inheritance practice rather than a failure — the same patrimonial logic that divided the Seljuk empire and would divide the Mongol one — and it had a genuine advantage: it kept the family invested rather than in revolt. Hama stayed in the hands of one branch for a century and a half and outlived the sultanate that produced it. The arrangement also had a cost that shows up in every chapter after this one: an Ayyubid prince could make his own truce with the Franks, his own alliance with a cousin against another cousin, and on two occasions his own alliance with the Franks against a cousin.

It is the same arrangement the Seljuks and the Carolingians both ran their own successions on: the family holds the whole, and each prince holds a share of it in his own right rather than as an official answering to one throne. What distinguishes the Ayyubid case is that the confederation was never reduced to a single line by a winner. Al-Adil and later al-Kamil came close, each reassembling most of it for a decade or so, and each time the reassembly lasted exactly as long as the man who made it.

There is a temptation to read seventy-nine years of cousins as decline from Saladin, and the chronology does not support it. The dynasty's largest territory and steadiest revenues came after 1200, not before, and the treaty that returned Jerusalem to Christian rule and the treaty that had taken it were signed by men of the same family forty-two years apart.

The brother who put it back together

Drafted from Humphreys, From Saladin to the Mongols: The Ayyubids of Damascus 1193-1260 (1977)peak

Saladin's sons fought each other for seven years. Their uncle al-Adil — Saladin's younger brother, who had governed Egypt for him and negotiated with Richard's envoys — took Damascus in 1196, Egypt in 1200, and by 1202 held or controlled almost everything the family had, having deposed or pensioned most of his nephews without a general war.

Humphreys' Ayyubids of Damascus is largely a study of what he did with it, and the answer is: as little fighting as possible. Al-Adil made truces with the Frankish states and kept them, on the working assumption that the coast was not worth the cost of taking and that Italian merchants in Acre and Alexandria were worth more as customs revenue than as enemies. Egyptian trade with Venice and Pisa grew under a ruler whose brother is remembered for the jihad. He divided the realm again in his own lifetime, giving Egypt to al-Kamil, Damascus to al-Mu'azzam and the Jazira to al-Ashraf — reproducing the structure rather than ending it — but he did it while alive and able to arbitrate, which is why it held.

This is the polity at its widest and richest: Egypt, most of Syria, the Jazira as far as Mayyafariqin, the Hejaz with the two holy cities, and Yemen under a separate Ayyubid branch that had been sent there in 1174 and rarely did what Cairo said. Nobody has published a figure for how much ground the confederation covered, and the figure would be misleading if anybody had, because the parts of it obeyed Cairo to different degrees.

The reputational cost was immediate and has never quite been paid off. A ruler who declines to fight the Franks and taxes their merchants instead was, to the religious scholars whose writing survives best, a disappointment after 1187 — and the modern nationalist reading, which needs Saladin and has no use for his brother, has kept that verdict in circulation.

Fools and simpletons

Drafted from Lane-Poole, A History of Egypt in the Middle Ages (1901)peak

In 1219 the sultan of Egypt offered a crusading army the entire kingdom of Jerusalem as it had stood before Saladin took it, in exchange for one Nile port. The offer was refused as inadequate.

The siege

The Fifth Crusade had landed at Damietta and settled in for a siege that went on for eighteen months. Reinforcements kept arriving — the counts of Nevers and La Marche, the earls of Winchester, Arundel and Chester — while inside the walls the garrison steadily dwindled until, of about fifty thousand men, four thousand were still able to stand to arms.

The contest was too unequal to last, and al-Kamil could see it. So he asked for terms, and named them: the whole of the kingdom of Jerusalem as it was before 1187, if Damietta were spared.

Lane-Poole's reaction to what happened next is one word long. Incredible as it appears, the exchange was refused as inadequate, and further concessions were demanded.

Who refused it

The refusal had two constituencies and neither was military.

Cardinal Pelagius, the papal legate, had been elected commander-in-chief. Filled with the exaltation of a pilgrim of the cross, he would have no traffic with the infidels — not on these terms or any. The other party held that Damietta was too valuable a commercial centre to give up, which is a merchant's objection to being handed a kingdom.

Against them stood the king of Jerusalem himself and the northern knights, who urged the advantages of the exchange and were overruled. The cardinal carried the day and it was resolved to press the war to the uttermost.

The verdict that survives came from a man who had been at Acre with Richard and knew what Saracen armies could do. When Philip Augustus heard that the crusaders had refused to take a kingdom in exchange for a city, he said: they are fools and simpletons.

Then eighteen months of nothing

Damietta fell by assault and the remnant of the garrison was massacred. Muslim alarm was such that the walls of Jerusalem and other Palestinian cities were hurriedly pulled down, so they could not serve the enemy as strongholds.

And then the crusaders sat in Damietta for a year and a half quarrelling among themselves, and did not move until July 1221.

They had also chosen the wrong door. Damietta was a valuable port but it was not the base for an advance on Cairo. The road every invader from Cambyses and Alexander to Amr and Amalric had used ran from Pelusium to Bilbays and, desert march aside, presented no obstacles. Between Damietta and the capital lay a network of canals and Nile branches offering a dozen obvious defensive positions — a series of traps for an army that did not know the country.

What the family did

While the crusaders argued, al-Kamil built. A little south of Damietta he threw up strong fortifications at a village he afterwards enlarged into a city and named al-Mansura, the Victorious.

And he summoned his relatives — which is the part that matters for this state. The Ayyubid confederation spent most of its existence as a set of princes fighting each other over Damascus, Aleppo, Homs and the Jazira, and the family quarrel is usually the explanation for everything it failed to do. Faced with a Christian army holding an Egyptian port, one and all laid aside their rivalries and rallied to him. Al-Mu'azzam of Damascus had come from the first.

The confederation had exactly one thing that could unite it, and it was an invasion of Egypt. That is the mechanism of Ayyubid survival and also its limit: a family that combines only under attack cannot conduct a policy, and eight years later al-Kamil would hand Jerusalem back to a Christian emperor by treaty, without consulting the relative who thought he owned it.

Jerusalem, given back

Drafted from Humphreys, From Saladin to the Mongols: The Ayyubids of Damascus 1193-1260 (1977)peak

The Fifth Crusade went to Egypt rather than Palestine, on the correct reasoning that Egypt was where the money and the army were. It took Damietta in 1219 after an eighteen-month siege. Al-Kamil offered, twice, to hand over Jerusalem and most of the kingdom's old territory in exchange for withdrawal; the papal legate Pelagius refused, both times, and in 1221 marched on Cairo at the season when the Nile rises. Al-Kamil opened the dykes. The Crusade was cut off in the flood and surrendered on terms of withdrawal.

Eight years later he gave Jerusalem away for nothing very much. Frederick II arrived in 1228 excommunicated, with a small force and no intention of campaigning; al-Kamil was in the middle of a quarrel with his brother al-Mu'azzam of Damascus, wanted a Frankish emperor as a counterweight, and signed the Treaty of Jaffa in February 1229. It gave the Franks Jerusalem, Bethlehem, Nazareth and a corridor to the coast for ten years. Muslims kept the Haram al-Sharif, its administration and its worship. Both men were denounced by their own clergy — Frederick by the pope, al-Kamil by preachers from Aleppo to Baghdad, and the sermons after the news reached Damascus are among the most bitter things in thirteenth-century Arabic writing.

Two things follow. The first is that the treaty is unintelligible without the appanage structure: al-Mu'azzam had died in 1227 and Jerusalem was in territory al-Kamil was trying to take from his nephew, so the city he was giving away was a bargaining chip in a family quarrel rather than a national possession. The second is that it worked. Al-Kamil got Damascus, held the confederation together until his death in 1238, and the ten years ran out with no crusade against him.

Jerusalem returned to Muslim control in 1244, taken not by an Ayyubid army but by Khwarazmian cavalry — refugees from the Mongol destruction of their own empire, hired by al-Kamil's son, who sacked the city on their way through. The same year they and the Egyptian army destroyed the Frankish field army at La Forbie. It was the last Ayyubid victory of that scale, and it was won by men the Mongols had dispossessed, six years before the Mongols arrived themselves.

The Khwarazmians

Drafted from Lane-Poole, A History of Egypt in the Middle Ages (1901)contraction

By the 1240s the Ayyubid confederation was at war with itself, and both sides had allies of the other religion.

Jerusalem had been handed to Frederick II by treaty in 1229 and had stayed Frankish, undefended, for fifteen years. The princes of Damascus and the smaller Syrian appanages were fighting the sultan in Cairo, and had made common cause with the kingdom of Jerusalem to do it. So the position in Syria was an Ayyubid sultan of Egypt on one side, and Ayyubid princes of Syria plus the Franks on the other.

Who was available

Al-Salih Ayyub, in Cairo, needed soldiers, and there was a supply of them wandering in upper Mesopotamia: the Khwarazmians, the remnant of an empire the Mongols had destroyed, driven west with nowhere to go and nothing to do but fight for whoever would pay.

He called them in.

In September 1244 they took Jerusalem, massacred seven thousand of the inhabitants — Lane-Poole's word for them is helpless, and they were: the city had no walls worth the name and no garrison — and the Holy City returned to Islam, this time permanently.

The following month the combined armies of the Franks and the Syrian Muslims met the Egyptians and the Khwarazmians near Gaza and were destroyed. It was the worst Frankish defeat since Hattin, and it was inflicted on an army in which Frankish knights and Damascene cavalry fought side by side.

What the victory was worth

Al-Salih Ayyub recovered Damascus in 1245 and Ascalon in 1247, and by the end of the decade he had restored his kingdom to the extent it had held under his father and his grandfather.

It is worth pausing on that, because it is the last time an Ayyubid sultan could be said to be winning. The instrument that did it was a stateless army of displaced mercenaries hired to fight the ruler's own relatives, and the most celebrated result — the recovery of Jerusalem — was achieved by men who had no interest in Jerusalem and were paid to be there.

Then he fell ill. The despatch announcing that Louis IX had landed in Egypt reached him while he was too sick to ride, and he had himself carried to the war in a litter.

He died during the campaign, and what his household did next is the subject of the next chapter.

Mansura, and the regiment that stayed

Drafted from Humphreys, From Saladin to the Mongols: The Ayyubids of Damascus 1193-1260 (1977)end

The last effective Ayyubid sultan of Egypt, al-Salih Ayyub, distrusted his relatives with reason and built his power on men who had none: a purchased regiment of Kipchak Turkish slave soldiers, quartered on the Nile island of Rawda, and called the Bahriyya after it. He bought them in unusual numbers, promoted them over his kinsmen's officers, and by the 1240s the Egyptian army was theirs.

Louis IX's crusade landed at Damietta in June 1249 and took it without a fight. Al-Salih Ayyub, already dying, was carried to Mansura and died there in November; his widow Shajar al-Durr concealed the death for weeks and ran the war with the senior emirs while an heir was fetched from the Jazira. In February 1250 the Franks tried to force the canal at Mansura, got a vanguard into the town, and lost it in the streets. By April the army retreating towards Damietta was destroyed and the king of France was a prisoner.

Then the winners turned round. Turanshah, the heir who had arrived from Hisn Kayfa, began replacing the Bahriyya's commanders with his own men, and in May 1250 they killed him — in front of the army, after a banquet, having chased him into the Nile. Shajar al-Durr was proclaimed sultan, ruled for eighty days, married the emir Aybak and handed him the title. The men who had just saved Egypt from a crusade had taken it.

The dynasty was not conquered. It was replaced from inside by its own soldiers, which is the third time that had happened to a state in this region in three hundred years. The Syrian branch outlived the Egyptian one by ten years: al-Nasir Yusuf held Aleppo and Damascus, claimed Egypt, failed to take it, and in 1260 rode away from Hulegu's army without fighting. He was executed in Mongol custody, and the Mongols who had taken his cities were beaten at Ayn Jalut by his father's regiment.

The prince of Hama

Drafted from Lane-Poole, A History of Egypt in the Middle Ages (1901)afterlife

When Baybars was at the height of his power in the 1270s, his orders ran from the fourth cataract of the Nile to the Pyramus in Anatolia, and eastward along the Euphrates. The bedouin of the deserts were his auxiliaries, the sharifs of Mecca and Medina were under his control, the king of Yemen sent him presents and the ruler of Abyssinia asked him for a patriarch. All Syria obeyed him.

Lane-Poole lists two exceptions. The coastal cities the Franks still held — and the principality of Hama.

Hama was Ayyubid. Saladin's family had been swept out of Egypt in 1250 and out of Syria by the Mongols and then the Mamluks in 1260, and one branch of it was still sitting in a Syrian city seventy years later, ruling as the Mamluk sultanate's vassal and being left alone.

A prince who wrote the books

The most interesting man that arrangement produced is Abu'l-Fida, prince of Hama in the early fourteenth century.

He is one of the two great compilers of his generation: a universal history that remained a standard reference in Europe into the nineteenth century, and a geography of the known world that was still being translated and argued over six hundred years later. He was also a working soldier of the Mamluk state, and Lane-Poole notes in a footnote to his account of the last great battle against the Ilkhanate that both Abu'l-Fida and the historian al-Nuwayri took part in it personally.

So the description of the fighting at Marj al-Suffar in the standard sources was written by two men who were in it — one of them the reigning prince of a principality, holding the last piece of Saladin's inheritance, riding in somebody else's army.

What that is

It is not a survival in the heroic sense. Nobody in Hama was plotting a restoration. The Ayyubid princes there took their orders from Cairo, sent their troops when Cairo asked, and were rewarded with the thing their cousins in Egypt had lost: a small state, governed in peace, for another three generations.

The line ended in 1341, when the sultan removed Abu'l-Fida's son and put Hama under a governor like everywhere else. That is ninety-one years after the Ayyubids lost Egypt and eighty-one after they lost Damascus.

Saladin's dynasty is remembered for one campaign and one city. The longest that any part of it actually lasted was in a town in central Syria that he never fought for, under the protection of the slave soldiers who had destroyed it, and its last notable member is remembered not for governing but for a geography book.