God's War: A New History of the Crusades Read online

  The bishops were appealing to the west for reinforcement. In fact, the army was receiving a constant stream of reinforcements from as far apart as Italy, England and Denmark, many travelling with the fleets that arrived in the Levant in 1097–8 providing the besiegers with vital sustenance.24 At no time was the army of God entirely cut off from the west or its Greek paymasters. In the need to replace its devastating casualties as well as its chronic problems of supplies may lie the decision of Tatikios, early in February 1098, to leave Antioch, as he claimed, to seek food and more troops.25 Although later condemned as a doubledyed coward by western observers intent on constructing a justification for the failure to cede Antioch to the Greek emperor, Tatikios may have harboured perfectly legitimate motives. The supply chain to Antioch had broken down; direct consultation with the imperial authorities might have improved matters. Tatikios left his staff behind at Antioch. There were rumours that he had struck a deal with Bohemund, granting him control over the Cilician cities of Mamistra, Tarsus and Adana. This would fit the actual rather than the imagined relations between the Greek general, veteran of commanding western troops in the Balkans, and the probably Greek-speaking Bohemund. They had travelled together in the vanguard to Antioch and remained in close contact. At this stage, Bohemund must have appeared one of the most philhellene of the western princes. A Greek story had Bohemund warning Tatikios to leave in order to avoid an assassination plot hatched against him by the other commanders. At the time, Tatikios’s departure made little impact; contemporary letters fail to mention it. While it is possible that Bohemund may have taken some role in engineering Tatikios’s withdrawal and it is certain that his absence suited the Norman’s schemes, conspiracy charges against either party lack evidence untainted by later propaganda, political posturing or special pleading.

  Of greater importance was the military result of the new sense of community. Thanks to Bohemund’s tactics and disciplined cohesion on the battlefield, the relief army of Ridwan of Aleppo was heavily defeated near the Lake of Antioch, some miles north-east of the city, on 8 February 1098. As at Dorylaeum, the battle near al-Bara in 1097, at Antioch itself later in 1098 and Ascalon in 1099, the fate of the crusade rested on the chance and skill of fighting. The Christian troops increased in effectiveness as numbers dwindled to a hardened core of veterans accustomed to strenuous pitched battles between massed forces of cavalry and infantry. The determination for victory existed in direct correlation to the consequences of defeat. In terms of morale, this gave the Christians an advantage. The victory over Ridwan temporarily steadied Christian resolve, while the arrival of an English fleet in early March allowed the blockade of the city to be tightened by the building of a new fort opposite the Bridge Gate, protecting vital access to the port of St Symeon. However, acute problems of food, horses and morale soon returned. Even the weather was terrible, reminding Stephen of Blois of home: ‘What some say about the impossibility of bearing the heat of the sun throughout Syria is untrue, for the weather here is very similar to our winter in the west.’26 Sharp encounters with the Antioch garrison sapped men and energy without disturbing the stalemate. Some optimism for the future may have been derived from the negotiations with Egyptian ambassadors in February and March 1098 and the dispatch of Christian envoys to accompany the Egyptians back to Cairo. By April, all gates of the city faced Christian blockade. However, news of a fresh Muslim relief force exposed the army’s continuing peril.

  During the spring of 1098, Kerbogha, atabeg of Mosul, assembled a large coalition against the western invaders from as far apart as Damascus, Anatolia and northern Iraq. Collecting allies as he went, Kerbogha was taking the opportunity afforded by the crusaders’ disruption of local power structures to create a new overlordship in Syria, ostensibly loyal to the Seljuk sultan in Baghdad. His alliance included elements hostile to the Fatimids of Egypt and Ridwan of Aleppo as well as to the westerners and their Armenian associates. The attempt to capture Edessa during the three-week siege in mid-May and the seizure of other cities and towns in the region point to a strategy in which relief of Antioch formed only a part. The atabeg’s objectives may be judged from the protraction of negotiations with Yaghisiyan’s son; the price of Kerbogha’s assistance was high. The actual outcome of the fighting of 1098 led to the establishment of Christian power in northern Syria, yet, until his defeat before Antioch, Kerbogha’s assault on Syria offered the opposite prospect of a revived Turkish authority over the region. As with the defeat of al-Afdal of Egypt at Ascalon in August 1099, the westerners’ victories in 1097–9 altered the political complexion of the Near East as much by denying the alternative outcomes of Seljuk or Fatimid révanches as by the establishment of their own limited hegemony.

  News of the approach of Kerbogha’s massive army reached the besiegers at Antioch in late May when it was only a few days’ march away. Although they were well informed of the diplomatic efforts to dislodge them, each relief attempt had taken the westerners by surprise. Kerbogha’s appearance was the nastiest yet, catching the Christians between a huge hostile field army and the impenetrable walls of Antioch. At a crisis meeting of the high command on 29 May, Bohemund was again entrusted with the leadership: if he could capture the city, he could keep it for his own provided no aid came from the Greek emperor, this last proviso reflecting the unease at Bohemund’s ambition felt by Adhemar of Le Puy and Raymond of Toulouse, who harboured his own designs on the city. This agreement did nothing to stem the panic. Desertions multiplied, the most prominent being that of Stephen of Blois. Only nine weeks earlier, he had boasted to his wife of his appointment to a prominent role in the communal leadership, in his words ‘lord, guardian and governor’, perhaps in charge of administrative matters or coordinating supplies.27 Stephen fled on 2 June, yet, within hours, Antioch had fallen.

  The legendary quality of so many incidents during the First Crusade is nowhere more evident than in the story of how Bohemund and an Armenian dissident in Antioch, Firuz, collaborated in allowing the crusaders to penetrate the walls of the city at a point under the traitor’s command on the night of 2–3 June 1098. It appears that Bohemund had been hatching the scheme for some time, probably before the meeting of 29 May. Contact across the front line at Antioch was common, especially with local Armenians. Bohemund and his followers possessed a linguistic advantage for this: on the night of the agreed commando-style raid on his section of the walls, they were able to converse with Firuz in Greek.28 However, the small force which established itself under cover of dark on the inside included Godfrey of Bouillon and Robert of Flanders; Tancred, Count Raymond and Bishop Adhemar had also been let into the secret and were instrumental in rousing the main army to exploit the incursion the following morning. The element of surprise devastated the civilian population waking to uproar and the sounds of massacre. The overwhelmed garrison immediately withdrew to the citadel, leaving the city below to be plundered at will by the invaders. Resistance crumpled. Yaghisiyan panicked and fled, fearful, perhaps, of reprisals for his oppressive regime; within hours he had been assassinated by local Christians, the coup de grâce possibly delivered by an Armenian butcher.29

  The fall of Antioch reflected the growing self-discipline and tenacity of the western army rather than any military resourcefulness or technological supremacy. There was minimal use of siege engines or artillery, surprisingly, in view of their use at Nicaea and later sieges at Ma ‘arrat al-Nu ‘man (December 1098), Arqah (March–May 1099) and Jerusalem. By contrast, at Antioch the Christians appeared reactive, often, as in confronting relief expeditions, dangerously so. While, at least after the foraging battle of December 1097, excelling in set-piece battles, there was no evident superiority in western tactics or equipment. The impression left by their siege of Antioch is of a frustrated bankruptcy of ideas to resolve the struggle, more effort being expended simply in keeping the army intact. Yet the solid bravery as well as occasional gaudy heroism suggests the crusaders’ continued belief in their cause. They celebrated their dead as
martyrs, in Stephen of Blois’s words in March 1098, whose souls were borne to the joys of paradise.30 Conviction alone was insufficient; fear, hunger, military incompetence or realism provoked despair and desertion, witness Count Stephen himself. Without idealism, the whole enterprise would have foundered long since.

  The rapid collapse of Antioch once its defences had been breached showed that the defenders had become as weakened as their attackers. When the slaughtering had ceased, the new rulers of Antioch faced a grim prospect. The Muslim garrison still held the citadel above the city; there was little food and fewer horses. Before supplies could be collected from elsewhere, just a day after the capture, the vanguard of Kergogha’s army arrived on the plain to the north of the city, having swept aside the Christian forward defences. By 7 June the city was invested; over the following week, all remaining Christian outposts beyond the city were overrun and a Muslim camp established close to the citadel to coordinate attacks from that sector. Heavy fighting all day around the citadel on 10 June caused another collapse in morale; that night panic spread among the crusaders, many fleeing the city using rope ladders to earn western opprobrium as furtivi funambuli, shifty rope-dancers.31 For those who remained, a sense of hopelessness was inescapable. Unable to summon help, unaware of any possibility of relief by land or sea, heavily outnumbered, ill-equipped and near to starvation, the Christians, now reduced to perhaps less than 30,000 including non-combatants, had reached their lowest ebb in fortune. Only desperate measures could avert ruin.

  From this extreme crisis emerged the visionary politics that characterized the rest of the campaign until Jerusalem was won. According to the story generally accepted by immediate eyewitnesses, on the very night of the panic and desertion, a Provençal priest, Stephen of Valence, beside himself with terror at what seemed to him the imminent fall of the city, while praying in the church of the Virgin Mary experienced a vision of Christ, the cross, Mary and St Peter (traditionally the first bishop of Antioch and the city’s patron saint). Christ assured Stephen that the beleaguered Christians would receive His aid in five days provided they demonstrated their faith through prayers, ceremonies and penitence for their sinfulness. After initial scepticism, insisting that Stephen swore to the truth of his statement on the Gospels, Adhemar of Le Puy exploited the vision by instituting more morale-stiffening religious ceremonies and persuading the princes to renew their oaths to stay with the expedition. More dramatically, in an almost simultaneous report, a poor Provençal pilgrim, Peter Bartholomew, claimed to have received over the previous months a number of visions of St Andrew (in the Gospels the brother of St Peter) in which the saint had urged penitence on the crusaders and, as a sign of God’s favour, had indicated where the Holy Lance that had pierced the side of Christ on the cross was buried in the cathedral of St Peter. Peter’s story conveniently matched Stephen’s by its promise of a sign of divine aid in five days. Adhemar and many others thought Peter a fraud, yet desperation and the advocacy of Raymond of Toulouse persuaded them to verify the story. On 14 June Peter and twelve others dug around in the floor of the cathedral until, as evening fell, Peter himself discovered what he and his fellow diggers took to be the point of the Lance sticking out of the ground at the bottom of the excavations. The discovery transformed the army’s mood from terrified inertia to awed encouragement, allowing the leaders to organize a military breakout with some prospect of success, further celestial sightings accompanying the preparations for battle hardly coincidentally containing saintly instructions to further penance and military discipline.32

  The objective reality of these visions or the authenticity of the Holy Lance are immaterial. The visions fitted contemporary models of such encounters, the visual iconography of the celestial messengers borrowing from contemporary art. A scrap of metal found beneath an old, much-renovated church after a day’s digging does not stretch credibility or credulity. What mattered in June 1098 was the crusaders’ belief. Although at first Bishop Adhemar and, later, others, particularly among the followers and propagandists of Bohemund, regarded Peter Bartholomew as a charlatan, not least perhaps because they may have seen a ‘Holy Lance’ on display in Constantinople, the visions provided the leadership with a precise plan to salvage the crusade. The link with reviving morale and military purpose is clear. One eyewitness suggested that Peter Bartholomew was only believed once the connection between the Lance and the defeat of Kerbogha had been made explicit; all associated the Lance with the subsequent success in battle. The visions fitted into the wider use of religious theatre and public ceremonial of penance to reinvigorate the army. An Armenian observer, writing within eighteenth months of the events, recorded the fervent prayers of the Christians in the cathedral of St Peter, designed to stiffen resolve, without mentioning the Lance.33

  While Stephen of Valence provided the leaders with an opportunity to impose strict discipline dressed as instructions from Christ, their heavenly commander, the poor layman Peter Bartholomew, fitted less easily into this hierarchical order. Even so, Raymond of Aguilers, guardian of the Holy Lance after its removal from the cathedral, implied Peter’s association with Count Raymond’s entourage and, cynics noted, his political interests. Evidently no protégé of Adhemar of Le Puy, Peter may have attached himself to the newly arrived William bishop of Orange, who helped dig up the Lance or perhaps one of Count Raymond’s vassals, Peter Raymond of Hautpol, mentioned as someone to whom Peter should report St Andrew’s instructions.34 Peter claimed to have been as far as Edessa and Cyprus in search of supplies for the crusade, pointing to aristocratic employment or contacts. He knew sections of the Latin liturgy by heart, so much, indeed, that he was able to forget some and still be able to display a degree of knowledge impressive in a supposedly humble peasant; some later chroniclers thought him a clerk. Although apparently universally accepted in the summer of 1098, Peter’s visions later appeared partisan, directed at elevating Raymond of Toulouse to leadership and forcing the princes to abandon their newly won possessions in northern Syria in favour of the march to Jerusalem. However, Peter Bartholomew was not necessarily a Provençal stooge. Although ultimately emerging as the champion of the popular demand for the invasion of Palestine, for months after June 1098 Raymond of Toulouse was as reluctant to cede position in Syria as Bohemund; during this period Peter’s repeated heavenly messages voiced the wishes of those unable to benefit from the profits of Syrian lordship, those, by virtue of the Christian success, denied local forage, plunder and booty. Only when Raymond decided to place himself at the head of the popular party at Ma ‘arrat al-Nu ‘man in January 1099 did Peter’s visions directly serve the count’s purposes, to his cost. At the siege of Arqah (February–May 1099), factional in-fighting, allied to doubts harboured even by supporters such as Raymond of Aguilers, forced a crisis of confidence. The theatricality of the visions was challenged by an equally potent set of judicial rituals devised to determine whether Peter was inspired or a fraud. A protracted trial instigated by Arnulf of Choques, Robert of Normandy’s chaplain, culminated on Good Friday 1099, 8 April, in an ordeal by fire. Although Peter satisfied his supporters by emerging alive from a corridor of flames, thirteen feet long, four foot high and only one foot wide, he succumbed to his injuries a few days later.35

  Visions and miracles articulated the aspirations of the mass of soldiers and pilgrims who, without prospect of lasting profit from conquest, concentrated on fulfilling their vows at Jerusalem. They also mirrored the changing military and political choices facing the princes. Peter Bartholomew’s equivocal fate did not mean the end of the dialogue between the terrestrial and the heavenly followers of Christ. Further visions of Stephen of Valence and another Provençal priest, Peter Desiderius, confirmed the centrality of relics, liturgy and penance in fixing the cohesion of the army of God on its march to Jerusalem, now transmitted through the politically safer medium of clergy rather than an uncomfortably radical layman. One strand of visionary politics in the final stages of the march elevated the Provençal cult of t
he lost leader, Adhemar of Le Puy, who had died, possibly of typhoid, at Antioch on 1 August 1098, part of an intense response to the deaths of comrades, many of whom soon reappeared to their friends in visions and dreams, witnesses to continued support from the other world. The presence in the army of Bishop Adhemar’s cross and cloak provided unassailable relics of unity and leadership; the dead bishop’s words inspired the troops at Jerusalem; his presence, some reported, assisted the final assault.36 During the siege of Jerusalem, another vision of Peter Desiderius, in which Adhemar urged a penitential procession around the walls of the city, legitimized a religious framework for persuading the army to attempt the final assault. At least it did so in retrospect. While the general accretion of miracle stories, relics and religious ceremonies from the sieges of Antioch onwards is undeniable, the neatness of the visionary prophecies and saintly interventions, their correlation with political and factional conflicts, the orderly narrative of celestial advice and the precise association of relics to visions in some accounts suggest crafting at the study desk as much as experience over the camp fire. Yet the importance of the miraculous and the holy, witnessed by participants’ letters, lay in the power of the perceived transcendent to transform events.

  Whatever later doubts and manipulation, the discovery of the Holy Lance and the injection of religious ceremony into the political discourse of the army contributed to the startling victory achieved over Kerbogha’s much greater forces by the Christian breakout of Antioch on the morning of 28 June 1098. This was Bohemund of Taranto’s finest hour. The day before, Peter the Hermit and an interpreter, Herluin, probably an Arabic-speaking Italian-Norman, had visited Kerbogha’s camp, a gesture of defiance as much as an attempt to negotiate or spy. Next day, under Bohemund’s direction, the Christian army, employing flexible, close-ordered, well-disciplined and tightly coordinated columns, first engaged, then threw back, outflanked and finally routed the forward divisions of the Muslim army before Kerbogha’s main force became involved. Much of the fighting was between infantry, at close quarters, as the Christians lacked horses. Despite greatly outnumbering the Christians, Kerbogha’s coalition disintegrated once the forward positions had been destroyed. Kerbogha fled ignominiously, leaving his camp, its prisoners, women, non-combatants, footsoldiers and booty open to the victors’ pleasure. The spoils were impressive: tents, camp equipment, livestock, beasts of burden, horses, camels, gold and silver, considerable supplies of food and drink. All Muslims found were killed. Unlike their co-religionists in Antioch three weeks earlier, the women were not raped; instead ‘the Franks… drove lances into their bellies’.37 Such unrestrained lethal warfare, characteristic of earlier medieval western conflicts against Vikings, Slavs or Magyars, had largely been replaced in the west by limited aristocratic internecine skirmishing. Its return in the aftermath of the battle of Antioch marked exultant, exhilarated release from weeks of terror.