On a hot July day in 1187 a crusader army marched away from its water supply toward a village on a ridge, and by evening the kingdom of Jerusalem had effectively ceased to exist. Three months later the city itself surrendered, and its Christian population was allowed to leave on payment of a ransom, which is not what had happened when the crusaders took it in 1099.

The claim this article defends is the victory with restraint thesis: that the military achievement and the conduct that followed it are separate things, that both are documented, that neither is flawless, and that the combination is the actual root of a reputation which has survived in two hostile traditions for eight centuries.

The two errors this article steers between are both common. The first presents flawless chivalry, which requires ignoring the executions at Hattin and the conduct of several sieges. The second dismisses the restraint as propaganda, which requires ignoring that the men reporting it were the enemy.

This article owns the wars. The man belongs to the article on Saladin as sultan of Egypt, the period and the strategic argument to the article on Ayyubid Egypt, and the assessment of whether the reputation is deserved to the article on Saladin as hero or conqueror.

The framing to hold throughout comes from the article on Ayyubid Egypt. These campaigns were possible because Egypt’s revenue funded them, and a commander who could replace a lost army faced an enemy who could not, which is the whole strategic content of the period.

The Saladin Crusades Table

The table below is this article’s findable artifact, setting out the campaigns and their outcomes against the evidence and the honest limit on each.

Element	What happened	The outcome	The honest limit
The frontier before 1187	Two decades of raiding, truces and inconclusive engagements along the Syrian and Palestinian border	Neither side could force a decision	Individual encounters are reported inconsistently by both traditions
Montgisard 1177	A crusader force caught his army dispersed and inflicted a serious defeat in southern Palestine	He escaped with difficulty and rebuilt	The scale of the loss is exaggerated in crusader accounts
The provocation	Raids from a lordship east of the Jordan on caravans and reportedly toward the Red Sea	Supplied the occasion for the campaign of 1187	The sequence of provocations and truce violations is disputed
Hattin, July 1187	The crusader field army left its water source, was surrounded on waterless ground and destroyed	The kingdom lost its army in a single day	Numbers on both sides are unusable
The aftermath of Hattin	Most captured nobles ransomed, military order members executed, one lord executed personally	Established both the clemency and its limits in one episode	Motives for the executions are inferred rather than recorded
Jerusalem, October 1187	The city surrendered after a short siege on negotiated terms	Inhabitants ransomed and permitted to leave rather than massacred	Ransom numbers and the treatment of the poor vary between accounts
Tyre	A single coastal city held out and became the base for the counterattack	The failure to take it made the Third Crusade possible	Whether it could have been taken earlier is debated
The Third Crusade 1189 to 1192	A large European expedition recovered much of the coast and did not retake Jerusalem	Stalemate	Casualty and force figures are unreliable throughout
Acre 1191	A two year siege ended in surrender, after which the garrison was executed by the crusader commander	The clearest counterexample to any chivalric reading of the conflict	The negotiation that broke down is described differently on each side
Jaffa 1192	A negotiated truce leaving the coast to the crusaders and Jerusalem in Muslim hands, with pilgrimage access	Ended the war without resolving it	The terms are summarized rather than preserved in full

The Two Decades Before

The conflict did not begin in 1187 and the years before it explain why that year was decisive rather than merely successful.

The strategic position after 1174, when he became independent of his Syrian overlord, was that the crusader states sat between Egypt and Syria, and the article on Ayyubid Egypt describes the union of the two as the achievement that made everything else possible.

How did Saladin fight the Crusades?

By assembling a base before risking a decision. He spent over a decade uniting Egypt and Syria, conducted raids and inconclusive campaigns against the crusader states, avoided a general engagement until conditions favoured him, and then in 1187 forced the crusader field army onto ground where it could be destroyed in one day.

The pattern of the intervening years was raiding, truces and sieges rather than decisive battle, since both sides understood that a field army lost is difficult to replace and neither wished to risk one without advantage.

The truces were frequent and functional, and the article on Ayyubid Egypt describes them as the normal condition of the frontier, interrupted by campaigning seasons rather than the reverse.

The defeat of 1177 is worth stating plainly because accounts that present an unbroken record of success omit it. A crusader force caught his army dispersed in southern Palestine and inflicted a serious reverse from which he escaped with difficulty.

Was Saladin defeated by the Crusaders?

Yes, notably in 1177 in southern Palestine, where a crusader force caught his army dispersed and inflicted a serious defeat, and again in several lesser engagements. He also failed to take Tyre in 1187 and failed to dislodge the crusader army from the coast during the Third Crusade. The record is one of eventual success rather than continuous victory.

The lesson he drew is visible in what followed, since the campaigns after 1177 show far more caution about concentration, supply and the choice of ground.

The build up to 1187 involved a series of provocations from the crusader side, including attacks on caravans and reportedly a naval expedition toward the Red Sea, which gave both an occasion and a justification.

The Frontier Between Campaigns

What the border looked like in the years when nobody was campaigning deserves description, since those years were the majority and the accounts skip them.

The truces were formal and specific documents, naming the places covered, the duration and the terms, and both sides observed them well enough that violations were noteworthy rather than routine, which is why the caravan raids that preceded 1187 could function as a justification at all.

The trade continued across the line throughout, and the article on Fatimid trade describes Italian merchants operating in Egyptian ports under agreement while their coreligionists fought the same rulers.

The agricultural life of the frontier districts continued, with villages paying dues to whichever authority held them and occasionally to both, which is the ordinary condition of a contested border anywhere.

The pilgrimage traffic ran in both directions, with Christian pilgrims visiting Jerusalem under crusader rule and Muslim traffic to the holy cities passing through territory the crusaders could threaten.

The local accommodations were extensive, and the settled crusader baronage in particular maintained working relationships with Muslim neighbours that newly arrived Europeans found scandalous, which is one of the fault lines the article’s section on the political setting of 1187 identifies inside the crusader council.

The raiding was the persistent low level exception, conducted by both sides against caravans, flocks and villages, tolerated within limits by everyone and periodically escalating beyond them, and it is a sequence of raids of exactly this kind, conducted from a lordship east of the Jordan against caravans under truce, that supplied the occasion for the campaign of 1187.

The general observation is that a frontier is a place where people live rather than a line on a map, and the populations on both sides of this one spent most of the twelfth century farming, trading and coexisting, punctuated by campaigns that the chronicles treat as the whole of the story.

The Holy Places After 1187

What was actually done with Jerusalem’s sacred sites after the surrender is worth setting out, since it is the substance of the restraint claim and is usually asserted rather than described.

The Muslim sites on the raised platform at the centre of the city had been converted to Christian use during the crusader period, with the shrine treated as a church and the congregational mosque used by one of the military orders.

The reconversion was immediate and ceremonious, with the buildings cleansed, the crosses removed and the congregational prayer restored within days of the surrender.

The Christian sites were not destroyed, which is the point on which the comparison with 1099 turns, and the principal church of the Christian holy places remained standing and in use.

The arrangements for Christian clergy allowed a presence to continue, and access for pilgrims was maintained, which the treaty of 1192 later formalized.

The Jewish return is the third element and it is rarely mentioned. The crusaders had excluded Jews from the city and had killed the Jewish population in 1099; the new arrangement permitted Jewish settlement again.

The endowment of institutions followed, with colleges and other foundations established in the city on the pattern the article on Ayyubid Egypt describes for Egypt and Syria.

The general observation is that what a conqueror does with an enemy’s holy places is among the most legible statements of policy available, and the decision not to destroy them, at a moment when nobody could have prevented it, is the single most concrete piece of evidence in the whole restraint argument.

Reading the Word Crusade

A note on terminology is worth including, since the vocabulary shapes how the conflict is understood on both sides.

The Muslim sources do not describe an opposing crusade. They describe Franks, using a general term for western Europeans, and they describe wars against them in the ordinary vocabulary of warfare against an enemy.

The framing of the conflict as a defensive religious struggle developed on the Muslim side across the twelfth century, promoted particularly by scholars and by rulers who needed to justify campaigning, and the article on Saladin as sultan notes the jurists who supplied those arguments.

The European term itself is later than the events, since the participants described themselves as pilgrims or as taking the cross rather than as crusaders, and the noun is a retrospective construction.

The consequence for a reader is that both sides’ religious framing was constructed and useful, which does not make it insincere and does mean it should not be taken as a neutral description of motivation.

The evidence for mixed motives is abundant on both sides, including the plunder, the territorial ambitions, the internal rivalries and the trade that continued throughout, which the article’s section on diplomacy sets out.

The alternative framing this article prefers is political and strategic, treating the conflict as a contest over territory, revenue and routes in which religious vocabulary supplied the terms rather than the objectives.

The general observation is that the language participants use about a conflict is evidence about their political culture rather than about their reasons, and separating the two is the first requirement for understanding any war conducted in religious terms.

The Ransom Economy

The commercial dimension of captivity ran through every campaign and it explains a good deal of the conduct the article has described.

The principle was that a prisoner of standing was an asset rather than a liability, since his family, his order or his ruler would pay for his return, and killing him destroyed the value.

The scale after Hattin was enormous, with a very large number of captives taken in a single day and the market for ransoms consequently flooded, which drove prices down sharply.

The mechanism at Jerusalem was the same principle applied to a population rather than to individuals, with rates set by status and the total treated as the price of the city.

The exceptions define the system. Men who could not be ransomed, because their order forbade it or because their conduct placed them outside the arrangement, were killed or enslaved, which is why the military orders were treated as a category apart.

The poor were the other exception, since a person with no ransom value could be enslaved and sold, and the article’s section on Jerusalem records that this happened to those who could not pay.

The moral reading available to a modern reader is uncomfortable and should be stated rather than avoided. A system in which prisoners survive because they are worth money is more humane in its effects than one in which they are killed, and it is not humane in its principle.

The practice on both sides was identical, and crusader accounts describe the same calculations about the same categories of prisoner.

The general observation is that clemency and commerce are entangled in this period in a way that resists separation, and any assessment of a commander’s conduct has to reckon with the fact that mercy and profit frequently pointed in the same direction.

What the Fighting Was Like

An account written from chronicles risks describing campaigns without describing combat, and a short treatment of the physical experience is worth including.

The armour was the defining condition for the crusader side. Mail over padding, a helm, a shield and weapons, worn in a Levantine summer, produces heat exhaustion within hours without water, which is the mechanism the article’s Hattin sections describe.

The archery was the defining condition for the other side. Composite bows shot from horseback at a rate a foot archer could not match, aimed frequently at horses rather than men, since an unhorsed knight is out of the battle.

The armour worked, which is the counterintuitive finding. Crusader accounts describe men walking with several arrows lodged in their mail and continuing, and the archery’s effect was cumulative exhaustion and the loss of horses rather than immediate casualties.

The charge, when it landed, was decisive and terrifying, and Muslim accounts describe the impact of formed heavy cavalry in terms that leave no doubt about why avoiding it was doctrine.

The infantry did most of the dying on both sides, as it does in every period, and the accounts pay it correspondingly little attention.

The heat and thirst are the conditions the sources return to most often, and the campaigning season was constrained by them as much as by politics.

The disease killed more than the fighting, particularly in the long siege camps, and the article’s section on Acre describes conditions that were lethal without any enemy action at all.

The general observation is that the experience of medieval warfare was overwhelmingly discomfort, thirst, illness and waiting, punctuated rarely by violence, and the chronicles invert those proportions because the violence is what they consider worth recording.

The Siege of Acre in Detail

The two year investment that dominated the middle of the Third Crusade deserves fuller treatment, since it was the largest single operation of the whole conflict.

The opening was improbable. A crusader force far too small for the task began besieging a major fortified port held by a substantial garrison, and it should have been destroyed within weeks.

The sandwich that developed is the peculiarity. The Muslim relief army arrived and besieged the besiegers, producing three concentric layers, and the arrangement then persisted for nearly two years.

The reason it persisted is naval. The crusader camp was supplied by sea and could not be starved out, and the harbour was contested, so neither the garrison nor the besiegers could be reduced by blockade.

The conditions were appalling for everyone. Disease, hunger and winter mud killed far more people on all sides than the fighting, and the accounts describe camps that had become permanent squalid towns.

The reinforcement flow decided it. European contingents arrived continuously across two years and Muslim contingents arrived seasonally and dispersed, so the balance shifted steadily.

The arrival of the European kings in 1191 with fleets and siege equipment ended it, and the garrison negotiated surrender when the walls could no longer be held.

The aftermath is the mass execution the article’s earlier section describes, which occurred within sight of the relief army and which ended any further negotiated surrenders.

The general observation is that the longest operation of the war was decided by shipping and by reinforcement rates rather than by any decision either commander made, which is the recurring finding of this whole article.

The Sources on Each Side

Naming the principal witnesses is useful, since the section on reading the two traditions treats them collectively and a reader should know what kinds of text they are.

The Muslim eyewitness material comes chiefly from the judge who accompanied the later campaigns and wrote a biography, whose account of Hattin is secondhand and whose account of the Third Crusade is direct.

The chancery secretary supplied a second contemporary account with documentary access, in an ornate style, covering a longer span, which the article on Saladin as sultan describes.

The Syrian chronicle tradition adds a third perspective, less admiring and in places critical, particularly regarding the campaigns against Muslim rivals.

The crusader material includes an anonymous itinerary of the English king’s campaign, chronicles compiled in the crusader states, letters despatched to Europe during the fighting, and continuations of an earlier Latin history of the kingdom.

The Old French continuation tradition is particularly valuable for the internal politics of the kingdom, since it reflects the settled baronage’s perspective rather than the newly arrived crusaders’.

The Byzantine and Italian material adds commercial and diplomatic detail, and the article on Fatimid trade describes the merchant networks that generated some of it.

The general observation is that this conflict is documented by participants, by contemporaries in both languages and by administrative material, which makes it one of the best evidenced wars before the modern period and which is why an article of this length can be written about campaigns eight centuries old.

The Two Commanders Compared

Setting the two principal figures of the Third Crusade beside each other is worth doing, since the comparison is usually made in terms of character and is more useful in terms of position.

The English king was a soldier of exceptional tactical ability, commanding an expedition with a defined objective, a finite duration and no political constituency in the theatre he was fighting in.

The Muslim sultan was a strategist and a coalition manager, commanding a permanent position with unlimited duration and a coalition that required continuous political management.

The consequence is that they were playing different games. One needed a result within a campaigning window and could accept high risk to get it; the other needed only to survive the expedition and could accept indefinite delay.

The tactical record favours the first, since he won the field engagement, took the coastal cities and outmanoeuvred his opponent repeatedly at the operational level.

The strategic record favours the second, since Jerusalem remained in his hands, the expedition went home, and the position established in 1187 survived essentially intact.

The reputational records diverge in an instructive way. The English king’s is complicated at home by his absence and by the executions at Acre; the sultan’s is enhanced by his own enemies, which the article’s section on restraint sets out.

The general observation is that comparing commanders by ability is less informative than comparing their positions, since a general with a secure base fighting a temporary expedition needs only not to lose, and a commander who needs only not to lose is playing a considerably easier game than one who must win before his army goes home.

What Would Change This Picture

Naming the evidence that would revise the account is the fairest test of a military narrative built from partisan chronicles.

Muster records from either side would transform everything, since every numerical claim in this article and everywhere else rests on chronicle figures that nobody can check.

Archaeological survey of the Hattin battlefield has been undertaken and further work on the distribution of finds would firm up the reconstruction of the final positions considerably.

Administrative records from the Ayyubid campaigns, on supply, pay and the composition of contingents, would settle the coalition questions the article treats structurally.

The text of the 1192 treaty would remove the small uncertainties in the section on it, and no copy survives.

Documentary evidence on the local Christian populations would fill the largest gap in the whole subject, since their position is inferred from the general legal framework rather than observed.

Correspondence from the crusader council before Hattin, beyond the retrospective chronicle accounts, would establish whether the decision to march was as contested at the time as it was later reported to have been.

None of these is likely except further archaeological work, and setting them out marks the boundary. The chronology, the terrain, the outcomes and the surrender terms are secure. The numbers, the intentions and the internal debates on both sides are reconstructions, and this article has tried to mark which is which.

Why These Campaigns Are Worth Studying

A closing argument, since medieval campaigns can seem a specialist interest and these repay general attention.

The first reason is Hattin itself, which is among the clearest examples in military history of a battle decided by logistics and terrain before contact, and which is therefore an unusually good teaching case.

The second is the documentation. Two hostile literate traditions describing the same events is rare before the modern period, and the article’s section on reading them shows how much that configuration permits.

The third is the strategic argument the article on Ayyubid Egypt makes, which generalizes well beyond this conflict: identify who has the secure revenue base and you have identified who wins a protracted war.

The fourth is the reputation question, which is a case study in how a historical figure’s standing is actually built, and the answer here is that it was built by his enemies.

The fifth is the corrective the article on Acre supplies, since the most celebrated chivalric encounter of the medieval period contains a mass execution of prisoners, and holding both facts together is the discipline that makes this subject useful rather than decorative.

The sixth is the diplomatic record, which contradicts the framing of total religious war more directly than any argument could, and which the section on negotiations sets out.

The general recommendation is that wars are best studied through their constraints rather than their battles, and this conflict supplies water, timber, shipping, harvest seasons and coalition politics as the actual determinants of a struggle everyone involved described in religious terms.

The Political Setting of 1187

The crusader kingdom’s internal condition before Hattin is part of the explanation and it is usually omitted, which makes the defeat look purely military.

The succession had been disputed. The kingdom had passed through a leper king, a child heir and a contested accession, and the king who marched to Hattin held a throne several of his barons regarded as illegitimate.

The factions were deep and personal, dividing the settled baronage of the kingdom from newer arrivals, and the military orders from the crown and from each other.

The truce policy was the specific fault line. The settled barons generally favoured accommodation with the Muslim powers, since they lived there and had to; newer arrivals and the orders generally favoured confrontation.

The caravan raids that provided the occasion for 1187 were the work of a lord acting against the crown’s own truce, and the crown could not or would not restrain him.

The council of war before the march reproduced every one of these divisions, and the crusader accounts describe accusations of cowardice and of treachery flying between men who disagreed about whether to leave the water.

The consequence is that the fatal decision was a political outcome rather than a military judgement, taken by a king who could not afford to look hesitant in front of barons who doubted his right to rule.

The general observation is that armies are commanded by people with political positions to protect, and a commander who cannot afford to appear cautious will not be cautious, which is a mechanism that has destroyed a great many armies in a great many periods.

The Aftermath in Europe

How the news was received in the west explains the scale of the response and adds a dimension the campaign narrative does not.

The news travelled by ship and reached Europe within months, and the accounts describe a reaction of shock disproportionate even to the loss of a kingdom.

The reason is that Jerusalem was not simply a city. Its recovery in 1099 had been understood as a divine vindication, and its loss required an explanation that implicated the sins of Christendom rather than the competence of its army.

The papal response was immediate and the preaching of a new expedition began at once, with taxation levied across several kingdoms specifically to fund it.

The scale of the response reflected the shock, and the article’s section on the Third Crusade describes three of the greatest rulers in Europe taking the cross, which had not previously happened.

The literary consequence was substantial and long lived, and the article’s section on legends describes the body of romance that grew around the campaigns over the following centuries.

The reputational consequence for the enemy commander is the one that concerns this article, and the section on restraint sets out why testimony from a shocked and hostile Europe carries the weight it does.

The general observation is that the reception of an event in a distant culture is a fact about that culture, and the European reaction to 1187 tells you a great deal about what Jerusalem meant in the medieval west and comparatively little about what happened in Galilee.

The Village and the Ridge

The battlefield itself deserves description, since the ground did most of the work and reading it is how the engagement becomes intelligible.

The setting is upland Galilee, a plateau of basalt country west of the lake, dry in summer, with the water sources known and few, and a distinctive twin peaked hill beside a village giving the battle its name.

The distance from the crusader camp at the spring to the objective was a day’s march, which sounds modest and is not for an armoured army in July with no water on the route.

The terrain between is broken and stony, hard on horses, and offers no shade, which compounds every effect of heat and thirst.

The lake was visible from the higher ground, which is a detail the sources record and which is worth pausing on: an army dying of thirst could see a freshwater sea it could not reach.

The final position on the slopes below the twin peaks gave a defensible feature and no water, which converts a defensible position into a trap once an army is surrounded on it.

The fires are the last element. Dry summer scrub burned upwind of an exhausted army adds heat, smoke and confusion at the moment when discipline was already failing.

The reconstruction is unusually confident because the terrain has not changed and the water sources are still where they were, which lets a modern reader stand where the decision was made and see why it was fatal.

The general observation is that battlefield terrain is the most reliable evidence a military historian has, since it does not exaggerate, has no side and cannot be edited by a chronicler, and this battle is the case in this series where it explains the most.

What Hattin Did Not Settle

Identifying the limits of the decisive victory is the clearest way to see why the war continued for another five years and another century after that.

It did not eliminate the crusader presence, since Tyre held and the coastal position survived, which the article’s section on that city treats as the hinge of everything afterward.

It did not remove the European capacity to intervene, since the crusader states were an outpost of a much larger world and the destruction of a local army prompted reinforcement rather than concession.

It did not solve the naval problem, and the section on that constraint explains why coastal cities supplied by sea remained beyond reach.

It did not resolve the coalition problem, since the army that won it dispersed with its plunder and had to be reassembled for every subsequent season.

It did not produce a settlement, since the loss of Jerusalem guaranteed a European response and the response guaranteed years of further campaigning.

It did not change the fundamental arithmetic that the article on Ayyubid Egypt sets out, which is that a rich base beats a frontier over time and that time is measured in generations rather than campaigns.

The general observation is that decisive battles are decisive about the thing they destroy and about very little else, and the gap between the completeness of the victory at Hattin and the incompleteness of what followed is among the clearest illustrations available in medieval military history.

The Century After

Tracing what happened to the position he left completes the account and connects this article to the rest of the cluster.

The immediate aftermath was the succession conflict the article on Ayyubid Egypt describes, during which the crusader states recovered ground while the family fought over the inheritance.

The middle decades saw the crusader invasions of Egypt that the article on crusader invasions treats, in which the strategic logic ran the other way and the enemy attacked the base rather than the frontier.

The negotiated transfer of Jerusalem in 1229, which the article on Ayyubid Egypt records, returned the city to Christian control by treaty for a period, which is a striking reversal of everything 1187 had meant.

The recovery of the city later in the century and the final elimination of the crusader states in the 1290s belong to the Mamluk regime, which the article on the Mamluk sultanate describes.

The pattern across the whole century is that the position established in 1187 was neither consolidated nor lost, and the frontier remained roughly where the treaty of 1192 had left it for decades.

The completion required a regime with the succession problem solved, which is exactly what the article on the Mamluk sultanate argues the Mamluks supplied by abandoning family rule.

The general observation is that the achievement of one reign is a position rather than a conclusion, and the reputation attaching to the man who established it frequently obscures the century of work that followed before the position produced a result.

Reading a Battle Account

The methodological problem these campaigns present is worth setting out, since battle narratives are among the least reliable documents a historian handles.

The first difficulty is that nobody present sees a battle. A participant sees a few dozen metres of it, a commander sees his own position and receives reports, and the coherent narrative of an engagement is always constructed afterward.

The second is numbers, which are unusable throughout this period. Both traditions inflate the enemy and understate their own, medieval writers used large numbers rhetorically, and no muster rolls survive to check against.

The third is causation, since accounts written after the event explain the outcome by whatever the writer considers explanatory, which for these sources means divine favour, treachery, or the character of commanders.

The fourth is the retrospective tidiness. Hattin is described in both traditions as a sequence of decisions leading inevitably to a result, and the council of war that preceded it shows that nothing about the outcome looked inevitable at the time.

The tests that work are the ones this series applies throughout. Terrain is checkable against the ground. Chronology is usually secure. Agreement between hostile traditions on specific events carries weight. And statements against a writer’s own interest, such as the crusader accounts recording their own divided council, are the most reliable material available.

Applying them to Hattin produces a secure account of the ground, the march, the water and the outcome, and an unreliable account of numbers, of individual conduct on the ridge and of what any commander intended.

The general observation is that a battle is best reconstructed from its logistics and its terrain rather than from its narrative, since the first two are physical facts and the third is a story someone told afterward.

The Wider Muslim World

The campaigns were fought by a coalition and its politics deserve attention, since the crusader states were not the only constraint on what could be attempted.

The contributions came from across the family confederation the article on Ayyubid Egypt describes, with relatives bringing contingents from Syrian and Mesopotamian territories and expecting a share of the direction as well as the plunder.

The allied and subordinate rulers of upper Mesopotamia supplied further forces, and their participation depended on continuing negotiation rather than on any obligation they could not decline.

The Abbasid caliph in Baghdad supplied recognition and nothing else, which the article on the Fatimid and Abbasid rivalry identifies as the standard arrangement: an endorsement that cost the caliph nothing and was worth having anyway.

The criticism from within the Muslim world persisted throughout, and the article on Saladin as sultan records the complaints that a man claiming to fight for the faith had spent a decade fighting Muslims.

The pressure to disperse was constant. Contingents pressed to go home after each season, and the accounts of the later campaigns record the sultan arguing with his own commanders for one more month more often than they record him arguing with the enemy.

The consequence for the war is visible at Tyre and in the failure to press the coastal sieges, and the article’s section on the peace of 1192 gives coalition exhaustion as the principal reason for the settlement.

The general observation is that a coalition army is a political instrument before it is a military one, and its commander spends more effort holding it together than directing it, which is the constraint that shaped the second half of this conflict.

The Local Christian Population

An element usually absent from accounts of these campaigns is the position of the eastern Christians who lived in the territory being fought over.

The distinction is fundamental and is frequently missed. The Christians of Syria, Palestine and Egypt belonged to eastern churches, and the article on Chalcedon and the Egyptian church split describes the divisions that separated them from Latin Christianity centuries before the crusades.

The crusader establishment did not recognize those churches. Latin hierarchies were installed over local Christian populations, and eastern clergy found themselves subordinated in their own dioceses.

The consequence for loyalties was that local Christians had no obvious reason to prefer crusader rule to Muslim rule, since the latter had guaranteed their worship and property under the framework the article on Copts under early Islamic rule describes for four centuries.

The evidence for what they actually did is thin and mixed, with instances of cooperation with both sides and no discernible general pattern, which is what one would expect from communities trying to survive a war fought around them.

The treatment after 1187 is documented for Jerusalem, where the eastern churches were permitted to remain and arrangements were made for their clergy, which contrasts with the Latin arrangements they had lived under.

The Egyptian dimension belongs to the article on religion and tolerance in Ayyubid Egypt, and the general position is that a war against Christian invaders exposed local Christians to suspicion regardless of their conduct.

The general observation is that the framing of the crusades as a conflict between Christendom and Islam omits the Christians who lived in the disputed territory and belonged to neither party, which is a substantial omission given that they were the majority Christian population there.

Two Legends Worth Dismantling

Certain stories about these campaigns are so persistent that stating their status is useful.

The first is the personal duel or meeting between the two commanders, which never happened. They corresponded, exchanged envoys and gifts, and negotiated for months, and the two men never saw each other.

The second is the story of the sword demonstration, in which one commander cleaves an iron bar and the other divides a floating silk cushion or a falling veil. It appears in later literature rather than in contemporary accounts and is an invention illustrating a contrast between two military cultures.

The third is the disguised visit, in which one commander enters the enemy camp incognito to observe or to treat the other, which belongs to the same body of romance.

The fourth is the knighting story, in which the Muslim commander receives European knighthood, which appears in medieval European literature attempting to absorb an admired figure into its own institutions.

The mechanism producing all of these is worth naming. A culture that admires an enemy generates stories bringing him closer to itself, and the article on Saladin as sultan describes the same process producing invented European ancestry.

The value of the legends is as evidence about their inventors rather than about their subject, and they tell you a great deal about what twelfth and thirteenth century Europe wanted its own knightly ideal to look like.

The general observation is that the most famous stories about famous events are usually the least reliable, since a story is repeated in proportion to how well it satisfies rather than to how well it is attested.

Diplomacy During the War

Negotiation ran continuously alongside the fighting and it is among the most revealing bodies of evidence for how the participants actually regarded each other.

The envoys moved constantly. Messages, proposals and counterproposals passed between the camps throughout the Third Crusade, and his brother conducted much of the negotiation personally, which the article on Ayyubid Egypt notes as one of that man’s roles.

The proposals canvassed were startling in their range. Partition schemes, arrangements for shared access to Jerusalem, and at one point a marriage between the English king’s sister and the sultan’s brother, with the couple to rule the disputed territory jointly.

The seriousness of that last proposal is debated and it was certainly discussed, which tells you something about how the participants understood the conflict that the rhetoric on both sides does not.

The courtesies exchanged are recorded on both sides: gifts of horses, fruit and snow, the loan of physicians during illness, and the return of captured non combatants.

The interpretation this article offers is neither cynical nor sentimental. Practical rulers conducting an expensive war negotiate continuously because a settlement is always cheaper than a campaign, and the courtesies are part of the negotiating relationship rather than separate from it.

The limits are visible in the same record, since negotiations broke down repeatedly, the Acre executions occurred in the middle of them, and both sides used the exchanges to gather intelligence.

The general observation is that the diplomatic record contradicts the framing of total religious war more directly than any other evidence, and the article on Ayyubid Egypt makes the same point about the 1229 transfer of Jerusalem by treaty a generation later.

What the Campaigns Cost Egypt

An article in a series about Egypt should ask what the country paid, since the article on Ayyubid Egypt argues that Egyptian revenue funded everything described here.

The financial burden was continuous across two decades, since armies in Syria and Palestine were paid, supplied and replaced from Egyptian resources.

The mechanism was the fiscal machinery the article on taxation describes, unchanged in its essentials, extracting from an agricultural base whose yield the flood determined.

The revenue assignment system the article on Ayyubid Egypt describes distributed part of that burden directly, since soldiers holding assignments collected from cultivators and campaigned on the proceeds.

The absence of the ruler was a further cost of a different kind, and the article on Saladin as sultan notes that Egypt was governed by his brother and by officials for most of the period after 1174.

The direct exposure was limited to the crusader attacks of the 1160s and 1170s, since after that the fighting was in Syria and Palestine, which means most Egyptians paid for a war they never saw.

The return on that expenditure is the article on Ayyubid Egypt’s argument. Egypt emerged as the leading power in the region with a fortified capital, a professional army, an endowed religious establishment and a strategic position it held for centuries.

The general observation is that the costs of a distant war fall on people who have no part in it and no say in it, and that the accounts written about such wars concentrate on the people who fought them rather than on those who paid.

The Naval Problem

Sea power decided more of this conflict than any account focused on land campaigns suggests, and the Muslim disadvantage was structural rather than a matter of neglect.

The constraint is the one the article on Fatimid trade identifies. Egypt has almost no usable shipbuilding timber, so every fleet the country has ever built depended on imports, frequently from the same Mediterranean powers the fleet existed to deter.

The Fatimid fleet had decayed before the Ayyubid takeover, and rebuilding it was expensive and slow, though the effort was made and Egyptian squadrons operated during the campaigns.

The crusader advantage was that the Italian maritime cities supplied shipping continuously, which meant a besieged coastal city could be provisioned and reinforced indefinitely while a besieger sat outside it.

The consequences appear at every coastal siege. Tyre held because it could be supplied by sea. Acre held for two years for the same reason until the arrival of the European fleets reversed the balance. And the crusader recovery of the coast in 1191 and 1192 was a naval operation supported by land forces rather than the reverse.

The strategic implication is that the crusader states could not be eliminated by a power without command of the sea, which the article on Ayyubid Egypt notes was accomplished only in the 1290s under a later regime.

The economic dimension compounds the military one, since the article on Fatimid trade describes Italian merchants trading profitably in Egyptian ports throughout, which means the same shipping that supplied the enemy also carried the commerce that funded the war.

The general observation is that a land power fighting a sea supplied enemy on a coastline is engaged in a contest it cannot win by land victories, and every coastal siege of this conflict demonstrates the point.

The March Down the Coast

The crusader movement from Acre southward in 1191 is the clearest demonstration in the whole conflict of what a disciplined army could do and what it could not.

The formation was the achievement. Infantry and crossbowmen on the landward flank, cavalry inside them, baggage on the seaward side, and the fleet keeping pace offshore supplying water and provisions.

The harassment was continuous. Mounted archers worked the column for days, shooting at the infantry screen, aiming at horses, and attempting to provoke a charge that would break the formation open.

The discipline held, which is the remarkable part, since the natural response of heavy cavalry under fire is to charge and the whole Muslim tactical system depended on that response.

The engagement at Arsuf came when the pressure finally produced a charge, and on that occasion it was made in enough strength and with enough of the army committed that it succeeded rather than exposing the column.

The consequence was a crusader victory in the field, which is worth stating clearly in an article about a Muslim commander’s campaigns, and which did not alter the strategic position.

The reason it did not is the article’s recurring theme. A tactical defeat costs an army that can be replaced very little, and the Muslim force withdrew, reassembled and continued the campaign of denial, while the crusaders had gained a coastal march they could have made anyway.

The general observation is that the coastal march shows the two systems in their purest form: an army that could not be broken while it kept formation and could not achieve anything decisive while it did, opposed by an army that could not break it and did not need to.

Scorched Earth

The systematic destruction of resources ahead of the crusader advance is among the most consistently applied elements of the whole campaign and it deserves separate treatment.

The instruments were the destruction of wells and cisterns, the burning of crops and fodder, the dismantling of fortifications that could not be held, and the removal of populations from the line of advance.

The logic is straightforward. An army that must carry everything it consumes moves slowly and can operate only within a limited radius of its supply, and denying local resources shrinks that radius drastically.

The application before Jerusalem is the decisive instance, and the article’s section on why the city was never retaken describes the fouled water on the approaches as the practical reason a siege was impossible.

The dismantling of fortifications is the counterintuitive part. Places that could not be garrisoned securely were slighted rather than defended, on the reasoning that a fortress taken by the enemy is worse than no fortress at all.

The cost fell on the population of the affected districts, and this is the element that celebratory accounts omit entirely, since destroying wells and crops in inhabited country has consequences that do not appear in a campaign narrative.

The comparison with the crusader practice is that both sides did it, and the accounts of the period record devastation as a normal instrument of war on every side.

The general observation is that pre modern warfare in a marginal environment is largely a contest over water and fodder, and the commander who understands that will beat the one who is thinking about battles, which is the pattern this whole conflict displays.

Siege Engineering

The technical business of taking a fortified place is what most of this war consisted of, and describing it corrects an impression left by accounts that concentrate on battles.

The instruments were mining, stone throwing engines, towers and rams, and the choice among them depended on the ground, the walls and how long the besieger could stay.

Mining was the most effective against a wall built on soft ground, since a tunnel driven under a foundation and then collapsed by burning its props brings the wall down, and this is how the breach at Jerusalem was made.

The counter to mining was countermining, and both sides employed engineers capable of detecting a tunnel and driving another to meet it, which produced underground fighting of a kind the sources describe with some relish.

The stone throwing engines of the period had improved considerably, with counterweight designs capable of damaging masonry rather than merely harassing defenders, and both sides used them extensively at Acre.

The defence rested on wall thickness, on flanking towers that allowed fire along the face of the wall, and on the ability of a garrison to sortie against the siege works.

The decisive factor in most sieges was neither engineering nor assault but relief and supply, since a besieger who cannot be dislodged and a garrison that cannot be resupplied produces a surrender on terms regardless of how strong the walls are.

The Acre siege demonstrates every element at once, with mining, engines, sorties, a relieving army camped behind the besiegers and a naval blockade contested by both sides across two years.

The general observation is that siege warfare is a contest of logistics conducted with engineering, and the reason this war took two centuries to resolve is that the crusader states’ fortifications converted every campaign into that kind of contest.

The Campaign That Followed Hattin

The months between July and the end of 1187 were the most productive of the whole war and they are usually compressed into a sentence, which conceals how the conquest actually worked.

The mechanism was that the cities had no garrisons, since almost every fighting man in the kingdom had marched to the relief and had been captured or killed on the ridge.

The sequence ran along the coast and through the interior at speed. Acre surrendered within days, then a string of coastal ports, then the inland towns, most of them on terms rather than by assault.

The terms offered were consistently generous and consistently practical. A place that surrendered kept its inhabitants and paid; a place that resisted risked the standard consequences of a storm, and almost none chose to resist.

The speed is the point. A campaign that would ordinarily have taken a decade of siege work took four months, because the field army had been removed and fortifications without garrisons are buildings.

The garrisons he installed were the constraint on going further, since every place taken required men left behind, and an army shrinks with every success of this kind.

The exceptions were few and they mattered. Tyre held, as the article’s section on it describes, and a small number of inland castles held out for months or years afterward.

The plunder was enormous and it had consequences, since men who have acquired a great deal of movable wealth become considerably less interested in a winter siege.

The general observation is that the destruction of a field army and the conquest of a territory are the same operation separated by a few months, which is why both sides after 1187 refused to risk a second general engagement and the war became a matter of sieges.

The Army He Commanded

The force that fought these campaigns deserves description, since its composition explains both what it could do and what it could not sustain.

The core was the professional cavalry the article on Ayyubid Egypt describes: Kurdish and Turkish military families, household troops attached to individual commanders, and the purchased and trained soldiers whose institution that article traces back to the ninth century.

The contingents came from across the empire. Egyptian forces, Syrian troops from the cities of the family confederation, Mesopotamian contingents from allied or subordinate rulers, and Bedouin auxiliaries all appear in the campaign accounts.

The command structure was the family confederation in the field, with relatives leading their own troops, which the article on Ayyubid Egypt describes as a system that expands quickly and coordinates slowly.

The financing was Egyptian above all, and the revenue assignment system that article describes meant that many soldiers held land whose revenue supported them, which creates a specific problem discussed below.

The seasonal limitation is the constraint that governs everything. Contingents assembled for a campaigning season and expected to disperse afterward, since men holding revenue assignments needed to return to them and allied rulers had their own territories to attend.

The consequence appears repeatedly in the record. Sieges that could not be maintained through a winter, armies that dissolved after a successful season, and commanders pressing for release, all of which the failure at Tyre illustrates.

The contrast with the crusader forces is instructive. The military orders and the settled nobility were permanently in the theatre, while a European expedition was temporary but intensely committed while it lasted, which is a different set of constraints producing a different tempo.

The general observation is that medieval armies are not standing institutions in the modern sense, and campaigns are shaped by when men will assemble and when they will insist on going home at least as much as by strategy.

The Crusader States as an Opponent

Describing what he was actually fighting is necessary, since the crusader states were a peculiar military system and their peculiarities shaped every campaign.

The territory was a long coastal strip with an inland extension, four principal states, and no strategic depth anywhere, which meant that a defeat in the field exposed everything at once.

The manpower was the permanent constraint. A settler population of limited size held territory among a far larger local population, and reinforcement arrived from Europe irregularly, seasonally and by sea.

The compensation was fortification of a very high order. Castles built and rebuilt across the twelfth century were among the most sophisticated in the world, and reducing them one by one would have taken decades even against no field opposition.

The military orders supplied the professional core. Permanent, celibate, disciplined bodies of fighting men supported by European estates gave the crusader states a standing force that did not go home after a campaign, which is why they were treated as a category apart after Hattin.

The naval position favoured them decisively, since the Italian maritime cities supplied shipping, reinforcement and provisions, and no Muslim power in the region could contest the sea reliably.

The political character was fragmented. Four states with their own rulers, the military orders answering to their own masters, and Italian merchant communities with their own interests produced a system that struggled to agree on a strategy, which the council before Hattin illustrates.

The general observation is that the crusader states’ survival for two centuries rested on fortification and sea supply rather than on numbers, and their vulnerability was that a single lost field army exposed everything simultaneously, which is exactly what 1187 demonstrated.

Hattin

The battle of July 1187 is the decisive event of the whole conflict and its mechanism is unusually clear, which is why it is worth setting out carefully.

The strategic problem for a Muslim commander was that crusader field armies were formidable in a pitched battle. Heavy cavalry charging in formation was extremely difficult to stop, and the crusader states’ castles made a campaign of conquest slow and expensive.

The solution required getting the crusader army to fight on ground and in conditions of his choosing, which meant drawing it away from water and supply into open country in summer.

What was the Battle of Hattin?

The engagement of July 1187 in which the crusader field army, having marched away from its water supply across arid ground in summer heat to relieve a besieged town, was surrounded, exhausted and destroyed near a village on a ridge in Galilee. It cost the kingdom of Jerusalem its army in a single day and left its cities without garrisons.

The bait was a siege. He attacked a town whose lady was the wife of a leading crusader commander, which created a political pressure to relieve it that overrode military caution.

Why did the crusader army leave its water supply before Hattin?

Because a council of war was divided and the argument for relieving the besieged town prevailed over the argument for staying at the spring. Political pressure, questions of honour and distrust between commanders all played a part, and the decision committed a heavily armoured army to a summer march across arid ground toward a waiting enemy.

The march was the error and it was recognized as such at the time, since crusader accounts describe the debate in the council and record that the decision was contested.

The conditions did the work. A day of marching in armour in July heat without water, harassed by mounted archers, with the enemy firing the dry scrub to add smoke and heat, produced an army incapable of fighting effectively by the time it reached the ridge.

The engagement itself was consequently short. The infantry broke away toward water and was cut off, the cavalry made repeated charges from the ridge and could not break out, and the army was destroyed as a formation.

The captures were extensive and included the king of Jerusalem, several leading nobles and the relic of the True Cross, which the crusader army carried into battle.

What happened to the True Cross after Hattin?

It was captured with the army and taken to Damascus, and its subsequent fate is unknown. Negotiations for its return appear in the diplomacy of the following years and none succeeded. Its loss was felt in Europe as a disaster comparable to the military defeat, since the relic had accompanied crusader armies as a battle standard.

The strategic consequence was total for the kingdom. An army destroyed in one day cannot be replaced, the cities had sent their garrisons to the field, and the following months saw place after place surrender because there was nobody left to defend them.

After the Battle

What happened to the prisoners is where the article’s central tension appears for the first time, and both halves of it are documented.

The treatment of the noble captives was standard for the period and generous within it. The king of Jerusalem was well treated, given water in a reported episode, and eventually released, and most of the aristocratic prisoners were ransomed.

The execution of the military order members was a separate decision. Members of the two great fighting orders were executed after the battle, on the reasoning that they were irreconcilable enemies who could not be ransomed and would return to the field.

How did Saladin treat Crusader prisoners?

Inconsistently, and the record is documented on both sides. Noble captives after Hattin were treated well and ransomed, and prisoners at Jerusalem were released or ransomed rather than killed. Members of the military orders were executed after Hattin as irreconcilable opponents, and ordinary captives were frequently enslaved, which was standard practice for both sides.

The personal execution of one lord is the sharpest episode. The lord of a fortress east of the Jordan, responsible for the caravan raids and the reported Red Sea expedition, was executed by his own hand after being brought before him.

Why did Saladin execute Reynald of Chatillon?

Because that lord had repeatedly attacked caravans in violation of truces and had reportedly sent an expedition toward the holy cities of the Hijaz, which made him an exceptional case rather than an ordinary prisoner. The execution was announced as a judicial act against a specific record of conduct rather than as treatment of a captured enemy.

The distinction drawn at the time is the one worth recording. The king was spared and told he had nothing to fear; the lord was executed for a documented record of truce breaking and for the expedition toward the holy cities, and the two decisions were made in the same tent within minutes.

The general observation is that clemency and severity were both instruments, applied according to a reasoning that the sources record and that a modern reader may find harsh or find intelligible, and the article on Saladin as hero or conqueror weighs which.

Jerusalem

The taking of the city in October 1187 is the event on which the cross cultural reputation principally rests, and the comparison that gives it force is with 1099.

The military situation was hopeless for the defenders. The garrison had been destroyed at Hattin, the city was full of refugees from the surrounding country, and the defence was organized by a knight who had arrived to evacuate his family.

How did Saladin capture Jerusalem?

By a short siege in October 1187 followed by a negotiated surrender rather than by storm. The city had lost its garrison at Hattin and could not be defended. After his engineers breached a section of wall, the defenders opened negotiations, and terms were agreed under which the inhabitants paid ransoms and were allowed to leave.

The siege lasted under two weeks. Engineers worked against a section of the northern wall, a breach was opened, and at that point the defenders sought terms rather than face a storm.

The negotiation is the interesting part and it was conducted under threat. The defending commander reportedly warned that if terms were refused the defenders would destroy the city’s holy places and kill their prisoners before dying in the breach, and terms were agreed shortly afterward.

The terms were ransom rather than massacre. Each inhabitant paid a set sum according to status, those who paid were escorted to crusader held territory, and those who could not pay were in principle enslaved.

How did the 1187 surrender terms compare with the 1099 conquest?

Sharply. When the crusaders took the city in 1099 they stormed it and killed a very large part of its Muslim and Jewish population over several days. In 1187 the city surrendered on terms, the inhabitants were ransomed and escorted to safety, the holy places were not destroyed and no general massacre occurred. Europe registered the contrast immediately.

The implementation was imperfect and this is where honest accounts diverge from celebratory ones. Many who could not pay were enslaved, and the numbers freed by charity, whether by the sultan himself, by his brother or by wealthy inhabitants, vary considerably between sources.

The contrast with 1099 is nonetheless the fact that mattered at the time and afterward. European writers made the comparison themselves, which is the corroboration effect the article on Saladin as sultan describes, and a reputation established by that comparison is not something admirers manufactured.

The treatment of the holy places is a further element. Christian sites were not destroyed, arrangements were made for Christian clergy to remain, and the Jewish population was permitted to return to a city from which the crusaders had excluded them.

The strategic consequence was to guarantee a European response, since the loss of Jerusalem could not be ignored in the way the loss of a frontier castle could.

Tyre and the Missed Opportunity

The one coastal city that did not fall is the strategic hinge of everything that followed, and its survival deserves examination.

The situation after Hattin was that place after place surrendered because no field army existed to relieve them, and a rapid sweep through the interior and along the coast took most of the kingdom within months.

Why did Saladin fail to take Tyre?

Because it was reinforced at the critical moment by a European nobleman who arrived by sea and organized the defence, because its position on a peninsula made it defensible against a besieger without naval superiority, and because the army had been dispersed after a long campaigning season and could not press the siege effectively in winter.

The reinforcement is the contingency. A European nobleman arrived by sea at exactly the moment the city was negotiating surrender, took command, and refused the terms that had almost been agreed.

The geography did the rest. A city on a peninsula connected by a narrow causeway can be attacked only along that causeway or from the sea, and the article on Fatimid trade explains the standing Egyptian problem of naval weakness.

The dispersal of the army compounded it, since a force that has campaigned all season and taken enormous plunder is difficult to hold together for a winter siege.

The consequence was decisive and it is recognized as such in both traditions. A single defended port gave the European expedition a landing place, and the article’s later sections on the Third Crusade describe what was done with it.

The counterfactual is worth stating without pretending it can be settled. Had Tyre fallen in 1187, a European relief expedition would have had no base on the coast and would have faced an opposed landing, which is a materially different proposition.

The general observation is that the failure to complete a conquest is frequently more consequential than the conquest itself, and the pattern recurs across military history wherever a single defended point survives a general collapse.

The Third Crusade

The European response occupied the last four years of his life and it is the phase in which the reputation was made in the west.

What was the Third Crusade?

The European expedition launched after the loss of Jerusalem, involving the rulers of England, France and Germany. The German emperor died en route, the French king returned home early, and the campaign was conducted mainly by Richard of England from 1191. It recovered much of the coast, failed to retake Jerusalem, and ended in a truce.

The scale was larger than previous expeditions and reflected the shock of the news. Three of the greatest rulers in Europe took the cross, which had not happened before.

The German contingent never arrived intact, since the emperor died crossing a river in Anatolia and his army dispersed, which removed the largest force before it reached the theatre.

The French king arrived, participated in the siege of Acre and returned home shortly after its fall, leaving the campaign to the English king.

The English king is the figure the western tradition remembers, and the two years of his campaign produced the encounters on which the chivalric literature was built.

The military character of the campaign was siege warfare and coastal operations rather than decisive battle, and neither side sought a second Hattin, for opposite reasons.

Did Saladin fight Richard the Lionheart?

Extensively, across two years from 1191 to 1192, though the two never met in person despite persistent legend. Richard won the field engagement at Arsuf and took several coastal cities; Saladin avoided a decisive battle, shadowed the crusader march, denied supplies and twice let Richard approach Jerusalem without being able to besiege it.

Acre

The siege that dominated 1189 to 1191 deserves its own treatment, both for its length and for how it ended.

The situation was peculiar. A crusader force besieged the city, a Muslim army besieged the besiegers, and the arrangement persisted for nearly two years with fighting, disease and starvation on all sides.

The relief attempts failed repeatedly, and the arrival of the European kings in 1191 tipped the balance, since the additional forces and siege equipment made the position untenable.

The surrender was negotiated, with terms involving payment, the release of prisoners and the return of the True Cross.

What was the massacre at Acre in 1191?

After the city surrendered on terms, the agreed payment, prisoner release and return of the True Cross were not completed to Richard’s satisfaction within the deadline, and he had the garrison and their families, some thousands of people, executed outside the walls in view of the Muslim army. It complicates any chivalric reading of the conflict.

The negotiation that broke down is described differently on each side, with the crusader account alleging deliberate delay and the Muslim account alleging impossible terms and bad faith.

The execution itself is not disputed by either tradition, and the number, though unreliable, is in the thousands.

The consequence for the war was to end any prospect of negotiated surrenders elsewhere, since a garrison that expects execution fights to the end.

The consequence for the historical reading is the one this article insists on. The most famous chivalric encounter of the crusading era contains a mass execution of prisoners by the crusader hero, and any account of the period that omits it is not describing the period.

The general observation is that the vocabulary of chivalry described an ideal that neither side consistently met, and the value of studying this conflict is precisely that both the ideal and the departures from it are documented.

Why Jerusalem Was Never Retaken

The crusader failure to recover the city despite twice approaching it is the strategic puzzle of the Third Crusade and it has a clear answer.

Why did Richard never attack Jerusalem?

Because taking the city was militarily feasible and holding it was not. Jerusalem lies inland, away from the coast the crusader fleet supplied, in country where the water sources could be poisoned and the supply line cut. A besieging army would have been isolated, and once taken the city would have required a garrison the expedition could not spare.

The advance was made twice, in early 1192 and again later that year, and on both occasions the army halted within a short distance and turned back after debate.

The supply problem was decisive. The crusader position depended on the coast and on ships, and an army inland was dependent on a vulnerable overland line through country the enemy controlled.

The water was the specific vulnerability, and the wells and cisterns on the approaches had been destroyed or fouled, which is the same instrument that had worked at Hattin.

The garrison question was the second element. Even a successful siege would have left a city requiring a permanent garrison in hostile country, and most of the expedition intended to go home.

The internal debate is recorded on the crusader side and shows the local barons arguing against the advance while the newly arrived crusaders pressed for it, which is a division of interest between those who would leave and those who would remain.

The Muslim position was to avoid battle and let the geography work, which is the correct response and which the earlier sections show him applying consistently after 1177.

The general observation is that the objective of a campaign and the sustainable objective of a war are often different things, and an expedition that could have taken Jerusalem and could not have kept it was correct to stop.

The Peace of 1192

The settlement that ended the fighting is the least celebrated part of the story and among the most instructive.

The position by mid 1192 was exhaustion on both sides. The crusader army had been in the field for years and its leader had pressing reasons to return home; the Muslim coalition had campaigned continuously since 1187 and its commanders were pressing for release.

Did Saladin and Richard make peace?

Yes, in a truce concluded in 1192, negotiated over months through envoys and never in a personal meeting. The terms left the crusaders holding a coastal strip, left Jerusalem in Muslim hands, guaranteed pilgrim access and set a term of a few years. Both presented it as acceptable and neither could present it as victory.

What did the Treaty of Jaffa actually say?

The surviving accounts summarize rather than preserve it. The crusaders retained a strip of the coast between two ports, Jerusalem remained under Muslim control with guaranteed access for Christian pilgrims, certain fortifications were to be dismantled, and a truce was set for a term of a few years. The details are reported with minor variations and no text survives.

The negotiation is well documented and it ran for months, conducted through envoys including his own brother, with proposals that at one stage included a marriage alliance between the two ruling houses.

The absence of a personal meeting is worth stating, since the legend of the two commanders meeting is persistent and false. They exchanged messages, gifts and medical assistance, and never met.

The criticism on both sides was immediate. Crusaders who had campaigned for years without recovering Jerusalem were bitter, and Muslim commanders who had expected the coast cleared were equally so, which the article on Saladin as sultan describes.

The defence available on his side is that the coalition was at its limit, that the coastal cities were supplied by sea and could not be reduced without a fleet he did not have, and that a ruler who overextends loses what he holds.

The consequence for the region was a century of coexistence along a coastal frontier, and the article on Ayyubid Egypt notes that the Mamluk regime eventually eliminated the crusader states in the 1290s.

The consequence for him personally was that he returned to Damascus and died within six months, which makes the settlement effectively his last act.

How the Two Armies Fought

The military systems in conflict were genuinely different and the differences explain most of the tactical pattern.

How did crusader and Muslim armies differ in tactics?

Crusader armies relied on a heavy cavalry charge in close formation, devastating when it connected and hard to recall once launched, screened on the march by infantry and crossbowmen. Muslim armies relied on mounted archers who harassed at a distance, declined contact, and sought to break formation or cut the enemy off from water.

The crusader strength was the charge, and a formed body of heavily armoured cavalry striking at speed could break almost any formation it reached, which is why avoiding that contact was the standard Muslim response.

The crusader weakness was mobility and supply. Armoured men and horses in a hot climate require enormous quantities of water, move slowly, and cannot pursue an enemy who declines to stand.

The Muslim strength was mobility and the archery of light horsemen, who could harass a marching column for days, inflict cumulative losses on horses and unarmoured infantry, and withdraw from any charge.

The Muslim weakness was the difficulty of finishing an intact crusader formation, since disciplined infantry and crossbowmen screening the cavalry could hold off archers indefinitely, which is what happened during the march down the coast in 1191.

The discipline problem cut both ways. A crusader army that charged prematurely was lost, and a Muslim army whose contingents pursued plunder or dispersed at the end of a season could not maintain a siege.

How were siege warfare and field battle different in this conflict?

Field battles were rare, decisive and avoided by both sides after 1187, since a lost army could not be replaced quickly. Sieges were the normal business of the war, slow, expensive and dependent on engineering, supply and relief. The crusader states’ castles meant a campaign of conquest proceeded fortress by fortress unless a field army could first be destroyed.

The consequence is the pattern the whole conflict shows. One decisive battle in 1187 that decided a kingdom, and years of siege and manoeuvre around it in which neither side would risk a second.

Reading the Two Traditions

The evidence for these campaigns comes from two hostile bodies of writing and knowing how to use them is essential.

The Muslim sources include the two contemporary biographies described in the article on Saladin as sultan, one of them an eyewitness of the later campaigns, plus chronicles from Syria and Egypt.

The crusader sources include contemporary chronicles, letters sent to Europe during the campaigns, and accounts written by participants shortly afterward.

The agreements between them are the most valuable material and there are more than a reader might expect. Both traditions agree on the outline of Hattin, on the terms at Jerusalem, on the executions at Acre and on the shape of the 1192 negotiation.

The disagreements are systematic and identifiable. Numbers are inflated on both sides, motives are attributed favourably to one’s own commanders, and each tradition explains its own defeats by treachery, bad luck or divine displeasure rather than by enemy competence.

The rule this article has followed is that where hostile traditions agree the fact is probably secure, and where they diverge the divergence usually concerns motive or number rather than event.

The corroboration effect the article on Saladin as sultan describes operates most strongly here. Crusader writers had every reason to present their enemy as barbarous and instead recorded courtesies, which is why those reports carry weight that no amount of admiring testimony would.

The general observation is that a conflict documented from both sides is a gift to a historian, and this one is among the best documented wars before the modern period precisely because everyone involved thought it mattered enough to write down.

The Restraint and Its Limits

The article’s central tension deserves a direct statement, since the whole reputation turns on it.

The restraint is documented. Jerusalem surrendered on terms rather than being stormed, noble prisoners were ransomed rather than killed, the king of Jerusalem was spared, the holy places were preserved, the Jewish population was permitted to return, and courtesies toward opponents are recorded by those opponents.

Why did Saladin release prisoners without ransom?

He did so in several recorded cases, including at Jerusalem where his brother requested and received a number of captives whom he then freed, and where he himself released others. The motives available are piety, calculated magnanimity and the political value of a reputation for it, and the sources do not permit a choice between them.

The limits are equally documented. The military order members were executed after Hattin, one lord was executed personally, those who could not pay at Jerusalem were enslaved, and ordinary captives across the campaigns were sold as slaves in the standard practice of the period on both sides.

The reasoning behind the distinctions is recoverable in outline. Ransomable nobles were an asset, irreconcilable religious warriors were a threat, a truce breaker who had attacked the holy cities was a special case, and the population of a surrendered city was covered by terms.

The comparison this article resists making is with modern standards, since a war in which enslaving prisoners was normal on both sides is not usefully judged by rules that did not exist.

The comparison it does make is with contemporaries, which is the fair test. Measured against the conduct of the 1099 conquest, against the executions at Acre and against the general practice of the period, the record is better than the norm and is not spotless.

The honest formulation is therefore narrow. He was more restrained than most commanders of his time in most circumstances, which is a real distinction, and he was a medieval commander who executed prisoners when it suited him, which is also true.

The article on Saladin as hero or conqueror weighs these against each other and reaches a verdict; this article’s job is to establish that both columns have entries.

Studying the Campaigns

For a student, these wars organize around a small number of dates and one strategic idea.

Be able to date Hattin and the fall of Jerusalem to 1187, the Third Crusade to 1189 through 1192, and the truce to 1192. Be able to explain Hattin as a battle decided by water, heat and ground rather than by the fighting on the ridge. Be able to state the contrast between the 1187 surrender and the 1099 storm. Be able to explain why Tyre mattered. Be able to say why Jerusalem was never retaken, which is supply rather than strength. And be able to give the executions at Acre as the necessary complication to any chivalric reading.

The table in this article carries that structure, and you can save this guide and build your own Egypt timeline free on VaultBook, where running the campaign dates against the Ayyubid political chronology shows how closely the war’s tempo tracked the state of the coalition behind it.

The transferable habit is separating the outcome from the fighting. Battles are frequently decided by decisions made days earlier about ground, water and supply, and Hattin is among the clearest cases available of a battle lost before contact.

A second habit concerns agreement between hostile sources, which the article on Saladin as sultan develops and which this material demonstrates repeatedly.

Myths Worth Correcting

Six claims about these campaigns circulate and each misleads.

The first is that he never lost. He was seriously defeated in 1177, failed to take Tyre, and could not dislodge the crusader army from the coast during the Third Crusade.

The second is that he and the English king met in person. They never did, exchanging envoys, messages and gifts across two years of negotiation and campaigning.

The third is that Jerusalem was taken by storm. It surrendered on negotiated terms after a short siege, which is the entire basis of the comparison with 1099.

The fourth is that his conduct was uniformly chivalrous. The military order members were executed after Hattin, one lord personally, and those who could not pay at Jerusalem were enslaved.

The fifth is that the Third Crusade failed completely. It recovered a substantial coastal strip and secured pilgrim access to Jerusalem, which is why the crusader states survived another century.

The sixth is that the wars were continuous. Truces were the normal condition, campaigning seasons were short, and trade between the two sides continued throughout, which the article on Fatimid trade describes as the standing pattern.

A seventh belief is a framing error rather than a myth, which is treating the conflict as a war of religion in the modern sense. Both sides used religious vocabulary, both allied across the divide when it suited them, and the Muslim commander spent more years fighting Muslims than crusaders, which the article on Saladin as sultan sets out.

The Verdict

The argument this article has defended is that the military achievement and the conduct are separable and that both need stating.

The military achievement was to force a decision at all. Two decades of raiding, siege and truce along the frontier had settled precisely nothing, and the campaign of 1187 produced in a single afternoon the result that twenty years of ordinary frontier warfare had failed to produce.

The mechanism was not superior fighting but superior positioning, since the crusader army was defeated by water, heat and ground before the engagement on the ridge began, and the decision that lost the battle was made in a council of war the previous day.

The strategic foundation was Egyptian, which is the article on Ayyubid Egypt’s central claim restated in military terms: a commander funded from a rich and secure rear could sustain campaigns year after year and replace whatever he lost, while his opponents depended on reinforcement arriving irregularly from a continent away by sea.

The conduct that followed was better than the norm of its period in most respects and not uniformly so, and this article has recorded both columns rather than balancing them.

The reputation rests on the comparison that Europeans made themselves and made immediately. A city surrendered on terms in 1187 set against the same city stormed and its population killed in 1099, with the contrast drawn by the losing side in its own chronicles while the war was still being fought, is evidence of a kind that no amount of admiration from a subject’s own side could ever manufacture.

The limits are those the article has named throughout. A serious defeat in 1177 that accounts of unbroken success omit. The failure to take Tyre, which handed the Third Crusade its landing place and made the following five years possible. The executions of the military order members after Hattin and of one lord personally. The enslavement of those at Jerusalem who could not raise a ransom. And a settlement in 1192 that left the coast in enemy hands and satisfied partisans on neither side.

The qualifications are those stated throughout and they are not incidental. Numbers on both sides are unusable, since no muster rolls survive and medieval chroniclers used figures rhetorically. Motives are inferred from outcomes rather than recorded. The text of the 1192 treaty does not survive and is known only from summaries. And every source describing these campaigns was written by somebody with a side, which is why agreement between the two traditions carries so much more weight than the volume of testimony within either.

What remains is the observation this article opened with. An army marched away from its water on a July morning and a kingdom ended by evening, destroyed less by fighting than by thirst, heat and a council of war that could not agree. The man who arranged it then let the capital of that kingdom surrender on terms rather than storming it, eighty eight years after the same city had been taken the other way. His enemies wrote that down, in their own chronicles, while they were still fighting him, and that is why anybody still knows his name.

Frequently Asked Questions

Q: How did Saladin fight the Crusades?

By building a base before risking a decision. He spent more than a decade uniting Egypt with Syria, which the article on Ayyubid Egypt identifies as the achievement that made everything else possible, and during those years he raided, besieged, made truces and avoided any general engagement. When the moment came in 1187 he manoeuvred the crusader field army onto waterless ground in July heat and destroyed it in a single day. Afterward he reverted to avoidance, shadowing the crusader march during the Third Crusade and declining a second battle he did not need.

Q: What was the Battle of Hattin?

The engagement of July 1187 that destroyed the kingdom of Jerusalem’s field army in one day. He besieged a town to draw the crusader army out, and its council of war chose to march to the relief rather than remain at its water source. A heavily armoured army crossing arid ground in summer heat, harassed by mounted archers and with the dry scrub fired around it, arrived near a village on a ridge in Galilee incapable of fighting effectively. The infantry broke away and was cut off, the cavalry could not break out, and the army ceased to exist as a formation.

Q: How did Saladin capture Jerusalem?

By a short siege in October 1187 ending in a negotiated surrender rather than a storm. The city had lost its garrison at Hattin and was crowded with refugees, and the defence was organized by a knight who had come to evacuate his family. His engineers breached a section of the northern wall, and at that point the defenders sought terms, reportedly warning that they would destroy the holy places and kill their prisoners if refused. Terms were agreed under which inhabitants paid ransoms by status and were escorted to crusader held territory.

Q: Did Saladin fight Richard the Lionheart?

Extensively, across two years from 1191 to 1192, though the two never met in person despite one of the most persistent legends of the period. Richard won the field engagement at Arsuf during the march down the coast and took several coastal cities. Saladin declined a second decisive battle, shadowed the crusader column, denied water and supplies, and twice allowed Richard to approach Jerusalem knowing the city could not be besieged successfully. The two exchanged envoys, messages, gifts and even medical assistance across the whole period.

Q: Was Saladin defeated by the Crusaders?

Yes, more than once, and accounts presenting unbroken success are selecting. In 1177 a crusader force caught his army dispersed in southern Palestine and inflicted a serious defeat from which he escaped with difficulty. He failed to take Tyre after Hattin, which proved decisive since it gave the Third Crusade a landing place. He lost the field engagement at Arsuf in 1191 and could not prevent the crusader recovery of much of the coast. The record is one of eventual strategic success rather than continuous victory.

Q: How did Saladin treat Crusader prisoners?

Inconsistently, and both sides recorded it. Noble captives after Hattin were treated well and ransomed, including the king of Jerusalem, who was spared and eventually released. At Jerusalem the inhabitants were ransomed and escorted to safety rather than massacred, and a number were freed without payment. Against that, members of the two military orders were executed after Hattin as irreconcilable opponents, one lord was executed personally for a record of truce breaking, and those who could not pay at Jerusalem were enslaved, which was standard practice on both sides.

Q: What was the Third Crusade?

The European expedition launched after the loss of Jerusalem, involving the rulers of England, France and Germany, which was an unprecedented concentration of European power. The German emperor died crossing a river in Anatolia and his army dispersed before reaching the theatre. The French king participated in the siege of Acre and returned home shortly after its fall. The campaign was conducted mainly by Richard of England from 1191 to 1192. It recovered a substantial coastal strip, failed to retake Jerusalem, and ended in a negotiated truce.

Q: Did Saladin and Richard make peace?

Yes, in a truce concluded in 1192 after months of negotiation conducted through envoys, including his own brother, and never in a personal meeting. The terms left the crusaders holding a strip of the coast between two ports, left Jerusalem in Muslim hands with guaranteed access for Christian pilgrims, provided for certain fortifications to be dismantled, and set a truce for a term of a few years. Both leaders presented it as acceptable, neither could present it as victory, and partisans on both sides were bitterly critical.

Q: Why did the crusader army leave its water supply before Hattin?

Because a divided council of war chose relief over caution. He had besieged a town whose lady was the wife of a leading crusader commander, which created political pressure to march. Crusader accounts record the debate and show that the decision was contested, with some arguing to remain at the spring and let the besieger come to them. The argument for marching prevailed, committing a heavily armoured army to a summer crossing of arid ground toward a waiting enemy. It was recognized as an error at the time.

Q: What happened to the True Cross after Hattin?

It was captured with the army, since the crusaders carried the relic into battle as a standard, and taken to Damascus. Its subsequent fate is unknown. Requests for its return appear repeatedly in the diplomacy of the following years, including in the negotiations over the surrender of Acre in 1191, and none succeeded. Its loss registered in Europe as a disaster comparable to the military defeat itself, since the relic had been the visible sign of divine favour accompanying crusader armies for nearly a century.

Q: Why did Saladin execute Reynald of Chatillon?

Because that lord’s record placed him outside the ordinary category of captured enemy. He held a fortress east of the Jordan and had repeatedly attacked caravans in violation of standing truces, and he had reportedly launched a naval expedition down the Red Sea toward the holy cities of the Hijaz, which was an offence of a different order. The execution was carried out personally, in the same tent and within minutes of the king of Jerusalem being spared and told he had nothing to fear, and the distinction drawn between the two men was deliberate.

Q: How did the 1187 surrender terms compare with the 1099 conquest?

The contrast is sharp and Europeans made it themselves. When the crusaders took Jerusalem in 1099 they stormed it and killed a very large part of the Muslim and Jewish population over several days. In 1187 the city surrendered on negotiated terms, the inhabitants paid ransoms by status and were escorted to safety, the Christian holy places were preserved with arrangements for clergy to remain, and the Jewish population was permitted to return to a city from which the crusaders had excluded them. The implementation was imperfect, since those who could not pay were enslaved.

Q: Why did Saladin fail to take Tyre?

Because a European nobleman arrived by sea at the precise moment the city was negotiating its surrender, took command and refused the terms. Beyond that contingency, the city’s position on a peninsula connected by a narrow causeway made it extremely difficult to besiege without naval superiority, which Egypt chronically lacked for the timber reasons the article on Fatimid trade describes. His army had also campaigned all season and taken enormous plunder, which makes a force hard to hold together for a winter siege. The failure gave the Third Crusade its landing place.

Q: What was the massacre at Acre in 1191?

After a siege lasting nearly two years the city surrendered on terms involving a payment, a prisoner exchange and the return of the True Cross. When those conditions were not completed to Richard’s satisfaction within the agreed deadline, he had the garrison and their families, numbering some thousands, executed outside the walls in full view of the Muslim army. Neither tradition disputes that it happened, though each blames the other for the breakdown in negotiations. It is the episode that most complicates any chivalric reading of the Third Crusade.

Q: How did crusader and Muslim armies differ in tactics?

Crusader armies relied on the charge of heavy cavalry in close formation, devastating when it connected and nearly impossible to recall once launched, screened on the march by disciplined infantry and crossbowmen. Muslim armies relied on mounted archers who harassed at a distance, declined contact, and sought to break formation, exhaust the enemy or separate them from water. The crusader weakness was mobility and the enormous water requirement of armoured men and horses; the Muslim weakness was the difficulty of finishing an intact crusader column that kept its discipline.

Q: Why did Richard never attack Jerusalem?

Because taking the city was feasible and holding it was not. Jerusalem lies inland, away from the coast his fleet supplied, in country where wells and cisterns on the approaches had been destroyed or fouled and where the overland supply line ran through territory the enemy controlled. A besieging army would have been isolated. Even a successful siege would have left a city requiring a permanent garrison that an expedition intending to go home could not spare. The local barons argued against the advance while newly arrived crusaders pressed for it.

Q: Did Saladin ever lose a battle?

Yes. The clearest case is 1177 in southern Palestine, where a crusader force caught his army dispersed and inflicted a defeat serious enough that he escaped with difficulty and the crusader accounts, predictably, exaggerate. He lost the field engagement at Arsuf in 1191 against Richard’s disciplined column. He failed at Tyre, could not prevent the recovery of most of the coast during the Third Crusade, and accepted a settlement in 1192 well short of what he had sought. His career is one of strategic patience recovering from tactical reverses.

Q: How were siege warfare and field battle different in this conflict?

Field battles were rare and decisive, and both sides avoided them after 1187 because a destroyed army could not be replaced quickly, which is exactly what Hattin had demonstrated. Sieges were the ordinary business of the war: slow, expensive, dependent on engineering and on whether a relieving force could reach the besieged. The crusader states’ network of castles meant that without first destroying the field army, a campaign of conquest proceeded fortress by fortress over years. Hattin’s importance is that it removed that requirement in a single afternoon.

Q: What did the Treaty of Jaffa actually say?

No text survives and the accounts summarize rather than reproduce it. The substance was that the crusaders retained a coastal strip between two ports, Jerusalem remained under Muslim control with guaranteed access for Christian pilgrims and clergy, specified fortifications were to be dismantled, and a truce was set for a term of a few years. Details vary slightly between the Muslim and crusader accounts, particularly over which places were included and what was to be demolished, which is the normal condition of medieval treaty evidence.

Q: Why did Saladin release prisoners without ransom?

He did so in several documented cases, most notably at Jerusalem, where his brother requested a number of captives as a gift and then freed them, and where he himself released others including the elderly and some who could not pay. The motives available are religious obligation, calculated magnanimity and an understanding that a reputation for it was politically valuable, and the sources do not allow a choice between them. The article on Saladin as sultan argues that his generosity functioned as policy whether or not it was also temperament.