God's War: A New History of the Crusades

by Christopher Tyerman

  In this political turmoil, where power rested with military warlords with varying claims to legitimacy, the western army appeared neither as distinctive nor as threatening as it thought itself. With the main contest for power in the Near East being fought in Iran, hundreds of miles to the east, the westerners’ targets – Cilicia, Antioch, Edessa, Jerusalem – were peripheral. As Tutush had discovered, rule of Syria counted for little against the forces of Iraq and Iran. Given the nature of their enterprise, the Christian expeditionary force rarely constituted a genuine threat to local dynasts. Despite the loss of Nicaea and defeats by the crusaders in 1097, the sultanate of Rum and the power of the Danishmends remained intact if dented. Only where Turkish authority had already eroded or collapsed – as in Cilicia or parts of northern Syria, including Antioch – did the crusaders threaten existing structures of authority. This new, fanatical, single-minded force apparently of Byzantine mercenaries fitted easily into a world dominated by armies of foreign hirelings – Kurdish, Turcoman or Armenian. The First Crusade was well suited to contemporary Near Eastern politics.

  Such insights were far from apparent to the members of the expedition as they set out to cross Anatolia in late June 1097. Within days of leaving the area of Nicaea, the army was almost defeated, its vanguard overrun and nearly destroyed, by Kilij Arslan’s field army. Four years later, similar western contingents were serially annihilated by local forces, leaving merely uneasy rumours of their fate. Such could easily have been the fate of the 1097 host. The battle, conventionally called of Dorylaeum, but actually fought over twenty-five miles to the north, attracted vivid if confused memories, recalling fear (‘huddled together like sheep in a fold… we had no hope of surviving,’ remembered one), recognition that it had been a close-run thing and certainty that victory had been God-given.8

  By the early morning of 1 July 1097, the Christian vanguard, perhaps 20,000 strong, comprising the contingents of Bohemund, Robert of Normandy, Stephen of Blois and Robert of Flanders, with the Byzantine troops under Tatikios, had advanced about forty-five miles south-east of Nicaea, reaching a valley just under three miles north of the modern Bozuyuk. There they were confronted by Kilij Arslan and his new ally the Danishmend emir. The Turkish force was mounted and probably outnumbered the western knights in the vanguard, which had become detached from the main body of crusaders under Raymond of Toulouse and Godfrey of Bouillon, around 30,000 strong, who were still some three miles away when battle was joined. On seeing the size of the Turkish army, Bohemund, his generalship skill already recognized, ordered the infantry, priests and other non-combatants to make a defensive camp, awkwardly with its back to a marsh, while the mounted knights advanced towards the enemy. Immediately things went badly, the mobile Turkish mounted archers driving the knights back to the camp, which became assailed on all flanks. Surrounded, the vanguard fought ferociously in dogged, bloody close combat, sustained by their close formation, lack of alternatives and a burgeoning esprit de corps.9 After more than five hours, the vanguard faced massacre until the arrival of the main force under Godfrey and Raymond compelled the Turks to break off their assault to engage in a series of running fights across the field before abandoning the fray when the Provençals, led, some said, by Adhemar of Le Puy, threatened to encircle them. In the pursuit, lasting some days, the westerners looted the sultan’s camp, seizing gold, silver, horses, asses and camels (employed as pack-animals for the rest of the booty), cows and sheep. While proving the crusaders’ mettle, the battle had exposed failures of command and the danger still posed by the Turks; disaster had come perilously close.

  While the largely nomadic Turkish forces of Kilij Arslan and the Danishmends could not be destroyed in one set-piece defeat, such a reverse undermined the sultan’s authority, especially over the towns of Anatolia with large Christian populations, and his recently constructed prestige amongst his Turkish supporters. Towns and cities across Anatolia repudiated the sultan, many welcoming the passing crusaders. The sheer size of the westerners’ force invited respect. In crossing central Anatolia, its main enemy was heat by day and, in the uplands, sharp cold by night, thirst, lack of supplies and fatigue. There was an almost unstoppable haemorrhaging of horses: one veteran suggested they lost most of them in a few weeks after the battle, a potentially fatal blow. While Tancred and Baldwin of Boulogne separately pressed on directly towards Konya, the main army, moving at times as little as five miles a day and never much more than ten, took a detour into more fertile territory around Pisidian Antioch to the south to allow effective foraging. Reunited at Konya in mid-August, the great army was showing signs of depletion. Goats, sheep, even dogs were pressed into service as pack animals, their backs soon lacerated with sores, while knights rode cows or walked. In high and late summer temperatures could soar to over 30 degrees. Some recalled hundreds dying, mainly of thirst; the true figure may have been thousands. New-born babies were abandoned by their mothers. The march across Anatolia scarred the memories of the survivors. The leaders were not immune, Raymond of Toulouse falling so gravely ill that he received the last rites and Godfrey of Bouillon apparently being attacked and injured by a bear.10 The requirements of food and water were paramount, determining the routes and behaviour of the army. Without cowed Turkish opponents and the general uprising against the sultan in Christian towns, the westerners could hardly have survived. The only serious military resistance encountered was at Ereghli (Heraclea) about a hundred miles east of Konya, around 10 September.

  After Heraclea, the great army divided, a decision displaying awareness of regional political conditions, local geography and topography, diplomatic opportunities, and prospects for collective and personal gain. Byzantine interests remained influential. Confronted by the formidable barrier of the Taurus mountains, the ordinary route for travellers to Syria led south-east through the steep, narrow pass known as the Cilician Gates (at its tightest about thirty yards across) down into the fertile Cilician plain, past Tarsus, Adana, Mamistra to Alexandretta and the Belen pass through the Ammanus mountain range, thence to northern Syria and Antioch, a journey from Heraclea of some 220 miles. This route was taken by two separate contingents, led by Tancred and Baldwin of Boulogne, acting on orders of the whole command or, possibly, as surrogates for Bohemund and Godfrey of Bouillon. Despite bitter, at times violent rivalry once in Cilicia, these forays in September and early October cleared opponents from the southern flank of the main army and established access to significant areas for supply and forage while denying them to the Turks of Antioch. Although Tancred and Baldwin came to blows at Tarsus, the latter being implicated in the Muslim massacre of 300 knights sent by Bohemund to reinforce his nephew,11 and again at Mamistra, they left behind sympathetic local rulers and garrisons: Baldwin’s at Tarsus, Tancred’s at Mamistra and possibly Baghras in northern Syria. The evidently self-interested campaigns of these two youthful, well-connected, skilful but landless adventurers materially aided the attack on Antioch and the protection of longer-term western interests in Syria.

  The main army turned northwards from Heraclea towards Caesarea in Cappadocia (Kayseri), crossing the mountains by steep but broader passes before turning south-east to Coxon (Goksum) and Marasch, through defiles almost as precipitous and narrow as those of the Cilician Gates, to approach Antioch from the north, a distance from Heraclea of just under 400 miles, a long, agonizing trek through inhospitable, barren high country, the road reaching over 5,500 feet, with the risk of snow at higher altitudes. The march from Heraclea to Antioch took about seven weeks, averaging about eight miles a day; losses in the Taurus mountains were great. The reason for this apparent diversion lay in the need to encourage Armenian Christian support and to secure the hinterland of Antioch by clearing the Taurus approaches of hostile Turks. Byzantine authority was restored in some places; at Comana, the Italian Norman Peter of Alipha or Aups, a veteran in Byzantine service, assumed command of the city ‘in fealty to God and the Holy Sepulchre and to our leaders and the emperor’.12 At others local
Christians resumed control, as the Armenian Simeon at an unnamed town in Cappadocia, and Tatoul at Marasch. Tatoul was a supporter of the Greek emperor; Simeon had accompanied the army providing it with local knowledge and political contacts. The diversion to Caesarea and Marasch thus served Greek interests by liberating local Christians from Turkish dominion under the imperial aegis. Militarily, approaching from the north isolated Antioch by freeing the mountain cities, capturing the strategically important city of Artah, which controlled the city’s eastern approaches, and establishing a presence in the fertile Ruj valley, east of the Orontes, on which Antioch stands. Combined with the Cilician activities of Tancred and Baldwin, the westerners stood well placed to attack Antioch.

  3. The Siege of Antioch, October 1097–June 1098

  This two-pronged attack on northern Syria territorially reconstituted much of the principality carved out by a renegade Greek commander, Philaretus Brachamius, between 1077 and the Turkish occupation of Antioch in 1085 and had been accomplished with the support of local Armenian lords, with whom there had been contact since Nicaea.13 One of them, Bagrat, persuaded the restless Baldwin of Boulogne, with whom he had travelled since Nicaea, to try his luck further east towards the Euphrates, also in lands once controlled by Philaretus. Leaving the army again after a brief stay in mid-October, with a small contingent of knights Baldwin moved on Tell-Bashir where he was welcomed by local Armenians as their lord. Having established military overlordship of the region up to the Euphrates, in a fashion familiar from Greek and Turkish precedents, in February 1098 Baldwin received an invitation from Thoros, the Armenian ruler of Edessa, forty-five miles east of the Euphrates, to come to his aid against the imminent advance of Kerbogha of Mosul, who was preparing a massive army to relieve Antioch and regain northern Syria from this nascent Franco-Armenian coalition. Baldwin accepted on condition Thoros recognized him as his heir. Arriving at Edessa on 20 February 1098, Baldwin tacitly or actively colluded with local dissidents in the rapid removal of Thoros, failing to intervene as his adoptive father was lynched by the town mob. On 10 March, Baldwin assumed authority over Edessa, establishing the first so-called Frankish state in the Levant. Apart from satisfying Baldwin’s ambition, the addition of Edessa and the lands on either side of the Euphrates to the westerner’s sphere of control proved vital for the survival of the whole expedition. A source of material aid and intelligence to the main army, in May 1098, Baldwin’s presence persuaded Kerbogha of Mosul to pause in his march on Antioch to besiege Edessa. This delay of three weeks was crucial; instead of trapping the western army outside the city walls, Kerbogha’s vanguard arrived just one day after the Christians had finally entered Antioch after an eight-month siege. Baldwin’s campaigns in Cilicia and Syria highlighted the impact of small forces; at Edessa, so his chaplain Fulcher of Chartres recorded, he was accompanied by as few as eighty knights, suggesting a force of a few hundred at most, Baldwin’s success exposing the fragility of local power structures and allegiances which contributed to the wider success of the western army in Syria.14

  * * *

  The siege of Antioch from October 1097 to June 1098 provided the twelfth century with its Trojan War, famed in verse, song and prose, commemorated in stone and glass, the central episode of trial and heroism in epic and romantic recounting of the First Crusade.15 For once, legend was justified. Despite the size of the western armies, Antioch presented a formidable obstacle. Although its garrison was modest, perhaps only a few thousand, the circuit of the walls, studded with scores of towers, ran for about seven and a half miles, much of it over rough, mountainous terrain. Contained within the fortified area of about three square miles was Mt Silpius, near the summit of which perched the citadel, a thousand feet above the main city. Incapable of investing Antioch by complete blockade, the crusaders’ alternative of assault offered little immediate prospect of success as they appeared at this stage to lack sufficient heavy artillery (i.e. great throwing engines such as mangonels or trebuchets) to breach the walls. A lengthy siege was in prospect, the only choice being whether to conduct it at close range or to blockade the city at a distance. Neither bore the certainty or even prospect of success as the governor of Antioch, Yaghisiyan, nominally a client of Ridwan of Aleppo, exerted much diplomatic energy to garner help. While past animosities prevented a concerted Muslim response, time lay on Yaghisiyan’s side, even though many of the outlying garrisons and commanders in the area, often non-Muslim, took the opportunity to throw off the governor’s unpopular rule, some Armenians regarding the westerners as liberators.

  Less clear is why the siege was undertaken in the first place. The Christian army was ill-equipped for siege warfare; the success at Nicaea had been due to Byzantine diplomacy and naval power as much as anything. Given Turkish disunity, negotiation rather than attack might have cleared a path southwards. The warring jealousies of the rulers of the great Syrian cities were not greatly moved by the appearance of the crusaders; accommodation, especially in the context of the westerners’ hardly secret negotiations with Fatimid Egypt, could have been achieved. When, in 1099, the Christian host marched on Jerusalem, there was little suggestion of taking Homs, Damascus or the cities on the Palestinian coast. Perhaps of greater strategic importance than Antioch itself were its ports, Alexandretta, St Symeon and Lattakiah, through which supplies of food (chiefly from Cyprus), war materials and men could reach the Christian army in Syria. Tancred had secured Alexandretta weeks before the main army reached Antioch, and a combination of Greek-sponsored and western fleets had occupied Lattakiah and St Symeon before the land army arrived.

  An agreed objective between the Emperor Alexius and the westerners, evident from Stephen of Blois’s comment to his wife from Nicaea, as a strategic target of the war, Antioch’s role was as much political as military or logistic. Alexius, so his daughter later admitted, had hired the western armies ‘to extend the bounds of the Roman (i.e. Byzantine) empire’, specifically, it seems, the northern Syrian principality based on Antioch which had acted as a semi-autonomous buffer between Byzantium and the Seljuks in the late 1070s and early 1080s.16 Its re-establishment would have greatly helped Alexius reclaim Asia Minor. The care taken by the Christian army to circle Antioch via Cilicia and the Taurus mountains and establish firm relations with Armenian rulers indicated such a policy. Although neither ignorant nor immune to the appeal of Jerusalem, Greek strategy held to more prosaic and customary ambitions, for which the emperor was prepared to lavish money, military aid, naval support and supplies on his western recruits.

  Greek plans for Antioch were complemented by the ambitions and needs of the westerners. By the winter of 1097–8, the resources of many of the lesser and some of the greater leaders of the armies were reaching exhaustion. Negotiations with the Fatimids of Egypt continued: an Egyptian embassy arrived at the crusader camp in February 1098. Agreement with the Egyptians would have reduced pickings in southern Syria and Palestine, making exploitation of northern Syria urgent and necessary. In the winter of 1097–8 elements in the Christian forces eagerly established themselves as de facto rulers of significant tracts of Antioch’s hinterland, even though the dangers of such loose investment of the city were well understood by a high command fearful of the army’s disintegration. Alone, the requirement to find winter quarters to rest and recover from the arduous march across the mountains of eastern Asia Minor scarcely explains the siege of Antioch, especially as the decision taken in October/November 1097 to invest the city closely exacerbated the difficulty of supplying such a large army. There was more to it than logistics. Of particular significance may have been the circumstances of Bohemund, who had attached himself closely to Byzantine interests since arriving at Constantinople in April 1097. It is possible that he hoped for, or even expected, some territorial reward from Alexius in Syria. The expedition into Cilicia by his nephew, Tancred, may have been conducted on his behalf.17 At Antioch his skills as a field commander propelled him to overall military leadership of the expedition in February 109
8 against Ridwan of Aleppo. The near-disaster of the engagement with the forces of Duqaq of Damascus on 31 December 1097 in the Orontes valley, twenty-five miles south of the city, due in part to divided command, persuaded the leaders to appoint a single field commander. This suited Bohemund’s political ambitions. Lacking the forces or the money to compete with Raymond of Toulouse or Godfrey of Bouillon, Bohemund may have long regarded Antioch as a prize to further his own interests, although it was only in May 1098, with the army threatened with annihilation by Kerbogha’s relief force, that he showed his hand. The attempt to capture the city conformed to Greek policy. In theory, it provided a focus for a period of recuperation for the Christian host; it helped keep Egyptian friendship by threatening the Seljuk powers of northern Syria, allowing the Fatimids to recapture Jerusalem itself in July 1098. The siege of Antioch was, therefore, of general political significance as well as reflecting some misplaced confidence. Raymond of Toulouse reputedly argued in favour of the siege on the grounds that God would see them right as He had done at Nicaea and in Anatolia.18 In fact, Antioch almost destroyed the crusade. Yet the extraordinary chain of events forged a harder unity of purpose among the mass of the army, a newly strident militant identity and confidence in divine favour expressed in the willingness of the survivors, great and humble alike, to integrate into their language and behaviour the rhetoric, symbols and theatre of visionary religious enthusiasm.