On 20 August, ten days after the deadline for the exchange of the first instalment of prisoners and ransom money, Richard called Saladin’s bluff. However much he wanted the relic and the money, Richard knew that further delay would only undermine the preparedness of his army for the tough campaign in prospect. He later described what happened:
On Saladin’s behalf it had been agreed that the Holy Cross and 1,500 living persons would be handed over to us, and he fixed a day for us when all this was to be done. But the time limit expired, and, as the pact which he had agreed was entirely made void, we quite properly had the Saracens we had in custody – about 2,600 of them – put to death. A few of the more notable were spared, and we hope to recover the Holy Cross and certain Christian captives in exchange for them.17
Ibn Shaddad was in Saladin’s camp a few miles away when the massacre occurred. Understandably, his account is more vivid.
When the king of England saw that the sultan hesitated to hand over the money, the prisoners and the Cross, he dealt treacherously towards the Muslim prisoners… He and all the Frankish forces, horse and foot, marched out at the time of the afternoon prayers on Tuesday 27 rajab (20 August). They… moved on into the middle of the plain. The enemy then brought out the Muslim prisoners for whom God had decreed martyrdom, about 3,000 bound in ropes. Then as one man they charged them and with stabbings and blows with the sword they slew them in cold blood.
The stunned Muslim advance guard watched helplessly while they sought orders from the sultan. By the time they tried to intervene, the killing was over. Next day they inspected the corpses and, Ibn Shaddad added, ‘were able to recognize some of them’.18
Richard I’s butchery of his Muslim captives was an atrocity not uncommon in war. It was not an act of random sadism, less so, for example, than Saladin’s own execution of the Templars and Hospitallers after Hattin. It ranks, perhaps, with Henry V’s slaughter of his French prisoners at Agincourt in 1415, except that then the battle was still in progress. Even Ibn Shaddad recognized that Richard’s action contained logic: revenge for Muslim killing of surrendering Christians during the siege of Acre ‘or that the king of England had decided to march to Ascalon… and did not think it wise to leave that number in the rear’.19 Richard and his apologists, and many observers not noted for their sympathy towards him, insisted on the justice of the killings, even their legality. One favourable source declared that, without the agreement with Saladin, the lives of the defeated garrison were forfeit jure belli, ‘under the rights of war’.20 The contrast between the control displayed when Acre fell and the cold-blooded savagery of the mass execution weeks later showed Saladin clearly the sort of adversary he faced. The sultan probably recognized the massacre for what it was, a deliberate act of policy for which his own actions were in part responsible. Over the following weeks he treated captured Christian soldiers with summary execution, occasionally allowing their corpses to be mutilated out of revenge. More seriously for the strategy of the war, he recognized that Richard had raised the stakes for all subsequent besieged Muslim garrisons. To avoid such consequences, Saladin moved quickly in September to dismantle the fortifications of Ascalon, one of Richard’s prime objectives, the key port in southern Palestine and the bastion on the route to Egypt. Yet Saladin understood the game both were playing and, judging from the respectful tone used of Richard by his close colleague Ibn Shaddad, appreciated the king’s skill at it. Whatever his public emotions, only a fortnight later he authorized his brother to resume face-to-face talks with Richard.21
By that time, Saladin’s own options had diminished severely in the face of the Christian advance. Five days after the massacre, on 25 August, Richard had fully assembled his forces and began the march south, along the coast road past Mt Carmel, Haifa, Caesarea and Arsuf to Jaffa. The eighty miles from Acre to Jaffa proved hazardous and exhausting. In the debilitating summer heat, with little shade, the Christians marched for the most part under arms to resist the repeated attacks of Saladin’s troops. The sultan shadowed the host, constantly harrying the line, especially the rearguard. This was entrusted to the Hospitallers. The Templars were in the vanguard. Between them were four separate divisions: the Angevins and Bretons; Guy of Lusignan, his Jerusalem followers and the Poitevins; the Anglo-Normans under the king; and the French, under Hugh of Burgundy and Henry of Champagne. The infantry and archers were divided into two columns, one of which marched on the landward side to provide outer defence for the knights against attack from mounted Turkish archers, while the other accompanied the baggage train on the seaward side. The large Christian fleet shadowed the army offshore, affording rest, food and protection. Before the crusaders lay a scorched landscape, its forts levelled, its crops burnt. Frequent and intense skirmishes cost both sides dear. Richard himself, constantly rallying the lines, was wounded. Progress was slow, barely five miles a day. However, as long as the battered Christian army remained intact it posed an increasingly menacing threat to the ports of Jaffa and Ascalon and thus Saladin’s whole position in southern Palestine even without a direct assault on Jerusalem. Saladin’s loss of sea-power was proving as significant as the land defeat at Acre. Recognizing the urgency of stopping the crusaders’ advance, on 4 September, with the Christians nearing the plain of Arsuf, Saladin agreed to Richard’s request to reopen negotiations. For Saladin it offered a chance to buy time to allow more reinforcements to arrive; for Richard it formed part of his consistent strategy of allying diplomatic with military pressure. On 5 September, Richard, with the jilted Humphrey of Toron as his interpreter, held a private interview with al-Adil which ended in acrimony, with Richard sticking to his demands for a return of the pre-1187 kingdom of Jerusalem.22 Nothing was achieved except to convince Saladin that his only option was to risk a pitched battle. With the failure of diplomacy as well as Fabian tactics, Saladin was compelled to try to convert what should have been the undoubted advantages of a home base, easy access to supplies and manpower, a sympathetic population and local knowledge into an immediate decisive victory.
On 7 September, just south of the Forest of Arsuf and north of the town itself, the increased pressure of the Turkish attacks forced Richard to halt the march and turn his column to face the enemy, as Saladin intended. The tactics of each side were clear. Turkish light cavalry would harass the Christian line to provoke a disorganized counter-charge which, by breaking the crusaders’ formation, would open them to piecemeal slaughter by fast-moving mounted archers. Failing that, a series of feints to draw a more concerted charge would have a similar, if riskier effect, exposing the Christians to Turkish counter-attack, provided the Turks avoided becoming trapped by the full force of a concerted enemy cavalry assault. The Christians’ aim was to withstand the Turkish archers, using the screen of infantry to shield the waiting cavalry, until the Muslims were committed to close combat on tiring horses, at which point a mass cavalry charge would be launched to annihilate the enemy and sweep them from the field. The two chief problems for Richard were to survive the hail of arrows and missiles for long enough, without taking too many casualties, to make this effective; and to maintain sufficient control over his separate divisions to ensure that, when it came, the cavalry charged as one to guarantee maximum impact.
Once battle was joined about nine o’clock in the morning, the Christian lines were pounded incessantly for hour after hour but they held. Just as Richard was preparing a decisive encircling attack on all fronts at once, the bruised and battered Hospitallers, on the left (i.e. northern) flank, goaded beyond endurance and worried at the loss of horses, charged, taking with them the French division on their right. Richard immediately grasped the tactical imperative and ordered a general attack that threw the Turks back. As Saladin regrouped, Richard, having kept the Anglo-Norman brigade in reserve as a rallying point around the royal banner, the dragon standard, the ensign of English monarchs at least since Harold Godwinson, managed to restore order to his lines, preventing them breaking up in pursuit of the enemy.23 He was thus
able to repulse the Turkish counter-attack and, in the final mêlée, launch a series of renewed charges of his own that eventually forced the Turks from the field. After a brief rest, the Christian army resumed its march, reaching Jaffa on 10 September.
This was no revenge for Hattin. Saladin’s army had escaped destruction. Some accounts, including Richard’s own brief despatch,24 discounted any climactic quality to the battle, portraying it merely as a sharper, more intense contest among many that had marked the Christian march south. One report suggested that the Christians had only lost just over 100 horses. The most prominent crusader killed, or at least the most mourned, was James of Avesnes, who had first arrived at the siege of Acre in September 1189. He later became the star of a secular chivalric cult, a familiar hero of uplifting anecdotes used by crusade preachers, poets and chroniclers in the thirteenth century.25 On the Muslim side, Ibn Shaddad identified only three important casualties. Yet, whatever the emphasis, Saladin had precipitated a direct attack, engaging larger numbers than previously, and he had been decisively repulsed, if not defeated. His object of halting the crusaders’ march to Jaffa, the port of Jerusalem, had failed. Ibn Shaddad recorded how inconsolable Saladin appeared on the evening of his defeat. Although his army remained intact and continued to shadow the Christian advance, it was now seemingly powerless to prevent its progress. After Arsuf, although the Christian army was still isolated in hostile territory, far from its base, its size, confidence, naval support and cohesion meant that militarily Saladin was reduced to reactive tactics; he was unable to dictate the course of events. This led him into uncharted territory with his restless allies, facing the prospect of keeping a field army at a plausible fighting strength for an indeterminate period. Saladin’s success had been based on offering the military askars, or standing armies, of the rulers of the Near East a share in profits: land, revenues, booty. Now he had little immediately to offer but doubt, debt, struggle and prayer.
JERUSALEM?
With the second crisis of the Palestine war surmounted, the Christian army established itself at Jaffa and the surrounding area in the weeks after the battle of Arsuf. The crusaders were now within striking distance of their goal. However, Richard’s grasp of strategy was subtler than many of those he commanded. An intelligent reading even of the history of the First Crusade would have revealed the need to secure as well as capture the Holy City. As with the leaders of the 1090s expedition, Richard, as well as Guy, the local barons and the military orders, understood that Egypt could determine the fate of Palestine. Equally, experience of campaigning in the west demonstrated that the chances of a successful siege increased if the attackers were in control of the surrounding region. For all these reasons, Ascalon occupied a central place in Richard’s calculations. Before he left Acre, he had intended to take the port, probably in order to make it his base for any operations against Jerusalem and Egypt.26 At the very least, occupation of Ascalon would hinder Saladin’s ability to reinforce his field army in southern Palestine. Alive to the danger, Saladin forestalled him by demolishing its fortifications shortly after the battle of Arsuf. Richard’s failure to persuade his allies at Jaffa in mid-September 1191 to prevent this has been seen by Richard’s apologists, then and now, as a crucial tactical error.27
This did not remove the strategic choices. While turning Saladin’s tactics against him by a series of foraging and harrying raids that led the Turkish army to withdraw from the coastal plain around Jaffa, Richard was still faced with alternative policies based on Jerusalem or Ascalon/Egypt. To placate his followers and wrong-foot Saladin, he tried to ride the two horses simultaneously. In October, while still holding out the prospect in a newsletter of taking Jerusalem by mid-January 1192, Richard floated a plan to the Genoese for a joint invasion of Egypt the following summer, which would require more men and ships to accomplish.28 How seriously he meant this scheme is unknowable. It may simply have been a ploy to keep the Genoese sweet while he remained in alliance with their rivals the Pisans. However, suggestions of an invasion of Egypt usefully added to the diplomatic pressure on Saladin. Negotiations with al-Adil intensified. Both sides seemed willing to explore a wide, even bizarre, range of possibilities. In mid-October Richard apparently offered his sister, the widowed Joan, to be one of al-Adil’s wives as part of a deal based on a Muslim–Christian condominium in Palestine ruled by al-Adil and Joan under Saladin’s suzerainty. The Christians would receive the coast, as Joan’s dowry, and free access to Jerusalem, while the Muslims retained nominal sovereignty over the whole and direct rule over the hinterland. Although later claimed by Richard to have been a joke, perhaps because on hearing of the plan his sister flew into a Vesuvial rage so characteristic of her short-tempered family, the suggested terms, the question of marriage and suzerainty apart, outlined a partition very similar to what was finally agreed a year later in the Treaty of Jaffa (2 September 1192). To complicate matters further, at the same time, the disaffected Conrad of Montferrat, who had held aloof from a campaign dominated by his rival Guy’s overlord, began to seek a separate treaty with Saladin, using the sultan’s old adversary, the fluent Arabic speaker Reynald of Sidon, as intermediary. Conrad apparently proposed swapping Acre for Sidon and Beirut, already promised him under the arbitration of July 1191, in order to establish a new Lebanese state for himself. Saladin could afford to keep both camps talking until the serious issues of the war in Palestine were resolved.29
In one respect, whatever his misgivings and clever schemes, the immediate course of Richard’s policy was determined for him. To the mass of his followers, the attraction of Ascalon, even Egypt, paled before the allure of Jerusalem, less than fifty miles away from their base at Jaffa. Within Richard’s cosmopolitan army there could only be one strategy that would gain majority support: a march on Jerusalem. This was especially true for crusaders from the west who had joined up specifically to answer the call to recover the Holy City. Many of them had been in Palestine for more than a year, some since 1189. The privations of the siege of Acre were for a purpose defined rhetorically and emotionally in traditional, First Crusade terms. With hindsight many sources argued that Jerusalem, even if captured, would have been impossible to hold against Saladin’s undefeated field army. It was argued that the Christians lacked adequate manpower to secure both Jerusalem and Jaffa and the supply lines to Acre, especially as the bulk of crusaders would depart for their homes in the west. For the local veterans, the lessons of the twelfth-century Outremer spoke loudly. Yet, the only reason any of them, Richard included, found themselves in southern Palestine at the turn of the year was the quest to recapture the Holy City. All Richard’s schemes were ultimately directed to that end. The providential nature of the enterprise had been proclaimed and reinforced at every turn by stories of heroism, reports of visions and, from the perspective of September 1191, success. As Saladin could not afford to bargain away Jerusalem, his most iconic triumph, Richard’s attempt to exert pressure by means other than a frontal assault on the Holy City lacked credibility. If there were too few to take or hold Jerusalem, the same was patently true, if not more so, for any attempt to invade Egypt. In fact, this lack of manpower, added to the factional difficulties within the Christian army, the rival negotiations with Conrad of Montferrat and the overwhelming fact that Richard, unlike Frederick Barbarossa, was not intending to devote the rest of his life to Outremer, indicated that the only chance of success, even in the terms Richard envisaged, lay in a military coup that brought Saladin to his knees. The stalemate after Arsuf required a new triumph. So even the logic of Richard’s own policy dictated an attack on Jerusalem. Given the fissiparous nature of the coalition he led, it had to be conducted as soon as possible, even though it was mid-winter and the weather was foul.
Despite being overruled by the council of leaders when he proposed an expedition to Ascalon in September, Richard understood his position. Whatever his instincts as a secular general, this was no ordinary war, at least in its objective if not conduct. If the army wa
s to stay together and his leadership recognized, both prerequisites for the effectiveness of his continuing diplomacy with Saladin, he had to march towards Jerusalem. The manner in which he did so suggested a serious intent; what precisely that was appeared less obvious. After a brief trip by sea to Acre, on 31 October Richard set out from Jaffa on the Jerusalem road. It had taken the First Crusaders a week to march from the coast to Jerusalem. After two months, Richard’s vanguard had only reached Bayt Nuba, on the edge of the coastal plain, still twelve miles from Jerusalem. Along the way, as well as fighting off repeated Turkish attacks, the Christians had rebuilt the castles of the plain ruined on Saladin’s orders. A six-week stop at Ramla allowed for provisions to be stockpiled. Richard spent Christmas at Latrun, still only a day’s ride from Jaffa. To Saladin, the advance seemed worryingly inexorable in its measured pace, although the dreadful weather contributed to its glacial progress. By the beginning of January 1192, Richard had achieved mastery over the coastal plain between Jaffa and the Judean hills. This may have been the limit of his ambition. But such occupation was also crucial if a realistic attack on Jerusalem were to be attempted. Throughout these manoeuvres, Richard kept up his talks with al-Adil, although sticking to his demands for a complete return to the pre-1187 frontiers. Urgency was added when Reynald of Sidon was seen inspecting the skirmishing between the Turks and crusaders while out riding with al-Adil.30 By November the discussions between al-Adil and Richard and their agents, and their friendly tone, were public knowledge, with open exchanges of gifts, mutual entertainment and feasting. This apparently contradictory behaviour for a warrior of Christ shocked many of Richard’s followers. To avoid the taint of appeasement and show that ‘he lacked not loyalty to God and Christianity’, when the negotiations faltered and fighting resumed, Richard compensated by eagerly slicing the heads off Turks and displaying them as trophies around the camp. Such bizarre turns of behaviour did not pass unnoticed even by Richard’s keenest fans.31
God's War: A New History of the Crusades Page 57
