14
Few events of the Middle Ages are better known than the storming of Jerusalem on 15 July 1099. Inspired by the preaching of Pope Urban II, a rag-bag group of adventurers had taken up the cross three years earlier with the promise of redeeming Christianity’s most holy city from the hands of the Muslims. After years of toil, which saw their numbers diminish fourfold, they’d now achieved the unthinkable. Their success would send shockwaves throughout Europe, North Africa and the Middle East, and leave a legacy which is felt to this day.
The story is a good one, regularly recounted in tones either triumphal or censorious by modern historians.1 But what often gets lost along the way is the specifically Norman dimension of the First Crusade. For not only did a large proportion of the crusaders hail from Normandy and southern Italy, but the very endeavour itself was informed by earlier Norman activity. It was the successes of the Conqueror and Guiscard which had first shown the western European aristocracy what could be achieved by risky ventures in foreign lands. And often a direct connection can be drawn between these early Norman conquests and the First Crusade.2 One of the most prominent of the crusaders was Bohemond of Taranto, whose exploits stand centre stage here. Denied an appropriate inheritance in Italy and the Adriatic, Bohemond was only too happy to march east in search of fame and fortune.
Yet the Norman contribution went well beyond Bohemond and his men. Another leading figure was the southern French count Raymond of Saint-Gilles, who had been married to Matilda of Sicily, Bohemond’s cousin. Raymond looked jealously on at the Hauteville successes, and crusading offered an opportunity to match them. Equally prominent are participants with connections to the duchy of Normandy. Foremost among these are Robert Curthose, the Conqueror’s eldest son, and Odo of Bayeux. Robert’s cousin, Robert II of Flanders, was another important recruit. This latter Robert enjoyed ties of his own to the Hautevilles, as his sister had wed Roger Borsa. And two other prominent crusaders, the brothers Godfrey and Baldwin, were the younger sons of Eustace of Boulogne. Finally, Stephen of Blois, perhaps the grandest of all the princes to join the expedition, was married to the Conqueror’s daughter Adela.
Earlier Norman exploits thus furnished the model for the First Crusade. This is not to say that the movement was crudely materialistic. For most participants, religious zeal and material gain (not to mention fame and honour) were either side of the same coin.3 This is what made the venture so enticing – it allowed knights and noblemen to seek their fortune in the name of God. The Conqueror had already sought papal endorsement for his invasion of England, and Guiscard and Roger had similarly framed their Sicilian campaigns in terms of the defence and expansion of the faith. The crusades represent a development of such practices, with papal approval now taking on a decisive role.
Yet if ideas about holy war had been percolating in Europe for some time, the immediate impetus for crusade came from the East – more specifically, from Alexios’ Byzantium. In the 1070s, Pope Gregory VII had already contemplated sending a papally endorsed army to relieve the emperor. The situation became even more critical in the early 1090s. At this point, a wave of attacks from the nomadic Pechenegs served to draw Alexios’ attention to the western provinces.4 In response, Abu’l-Qasim and the Seljuk Turks struck at the rich imperial lands to the north and west of their base at Nicaea (İznik). Sensing weakness, other Turkish groups began to raid deep into the remaining provinces of Anatolia. Matters turned from bad to worse in late 1092, when the Baghdad-based sultan Malik-Shah died. Alexios had enjoyed good relations with Malik-Shah, but the ensuing succession dispute created considerable instability, which the increasingly independent Turkish rulers of Asia Minor were happy to exploit.
As failure followed failure, the mood in the capital darkened. Unrest was already visible in the early 1090s and came to a head in 1094, when conspirators began plotting to oust Alexios and replace him with his nephew, John Komnenos. It increasingly looked as though Alexios would go the way of his predecessors. It was in this context that the emperor made the fateful decision to appeal to Pope Urban II for aid.5 In a sense, this was an extension of the tactics long employed by the Byzantine emperors, who’d successfully pitted Norman knights and Anglo-Scandinavian Varangians against eastern Turks for many years. Yet the results were to prove very different indeed. Neither Alexios nor Urban could have predicted the enthusiasm with which their call would be taken up, nor could they have imagined the long-term consequences of their actions.
Alexios’ request arrived at a large church council at Piacenza in early March 1095. Urban responded favourably to the initial entreaty, and by summer was starting to drum up support for an expedition in his native France. These efforts were crowned by the Council of Clermont later in the year, at which Urban gave a stirring speech calling those present to march east to free the Holy Land.6 The pope had good reason for backing the expedition. It promised to distract attention from interminable conflicts between himself and the German emperor (the so-called Investiture Contest). It also offered the prospect of rapprochement between the eastern and western churches, which had long been a papal aspiration. Amid the growing popularity of pilgrimage, eastern affairs became even more pressing, as disruption in Asia Minor now threatened to cut Christianity’s most holy city off from its most devout worshippers.
Precisely because Urban’s appeal proved popular, it took time for the crusading forces to assemble. Not all, however, were willing to wait. Inspired by the apocalyptic preaching of a man called Peter the Hermit, a group of minor noblemen set out before the main armies in the so-called People’s Crusade of 1096. These groups committed a series of gruesome Jewish pogroms before departing Europe – a reminder of the perennially fine line between religious zeal and bigotry.7 Despite clashes with fellow Christians in Hungary, they arrived in Constantinople in early August and were ferried across the Bosporus. Shortly thereafter, however, the motley army was cut down by Turkish forces outside Nicaea. Many died or were enslaved. Some of the survivors now opted to await the arrival of the main force.
Of the four princely armies to follow, three came from France and the Low Countries. One of these – that of Robert Curthose – was Norman through and through, and many of the crusaders that made up the other armies enjoyed Norman associations. But perhaps the most significant Norman contribution would come from southern Italy, in the form of the restless Bohemond. Much like Curthose, Bohemond had drawn the short straw where dynastic succession was concerned, so stood to gain much from the expedition. The ‘Deeds of the Franks’ (Gesta Francorum) – an account of the crusade, written by an anonymous associate of Bohemond – reports that the Norman prince first got wind of the movement while besieging Amalfi, in early summer 1096, with his lord (and half-brother) Roger Borsa. Inspired by the news, Bohemond immediately pledged his involvement, cutting up his own cloak to form crosses for the men who flocked to join him. Modern historians have long suspected that this impressive coup de théâtre was premeditated. But we should not be too cynical. The anonymous report is corroborated by that of Geoffrey of Malaterra, who says that Roger was forced to abandon the siege because Bohemond diverted so much of his army. Geoffrey was a well-informed observer (and no fan of Bohemond). Moreover, it’s not until August of this year, that we see Bohemond liquidating his landed assets to help fund the expedition.8 Urban’s preaching in support of the crusade had been focused on France, so it may well have been spring or early summer 1096 before the full details reached southern Italy.
While Roger Borsa must have been happy to see the back of his brother, the circumstances cannot have pleased him. The anonymous chronicler claims that the two parted on good terms, but Geoffrey records the duke’s frustration (as well as that of his uncle Roger I). That Bohemond was able to recruit so much of the army should not come as a surprise. Italo-Norman aristocrats knew full well the wealth and prestige to be won from foreign ventures. And with the conquest of Sicily now complete, there was little prospect of further gains in Italy. Fame and fortune would now be won in the East.
Bohemond probably always had more in mind than saving the Holy Land. After the failure of the Balkan campaigns of the 1080s, he had unfinished business with Alexios. Geoffrey reports that Bohemond embarked on the expedition precisely because he’d long had designs on the eastern empire (Romania, as Geoffrey calls it, in keeping with Byzantine tradition). Taking imperial service may seem like an unusual way to bring the empire down. But Bohemond knew how easily gamekeepers could turn poachers, particularly in the volatile world of eleventh-century Byzantium.
If the decision to join the crusade had been spontaneous, Bohemond was sensible enough to spend the coming months preparing for the expedition. He had less far to travel than his northern counterparts and knew the route well, so there was no rush. Bohemond left from Brindisi in October 1096 and, in a striking repeat of the campaigns of the 1080s, landed near Valona on the Ilyrian coast. In doing so, he parted ways with the other crusading groups which travelled through Italy. They all sailed straight to Dyrrachion, where they met up with imperial forces and traversed the Via Egnatia. Bohemond’s deviation is significant, particularly in light of Geoffrey’s testimony. He may have been weighing up a strike against the emperor. One account has it that Bohemond requested Godfrey of Bouillon (Eustace’s son) not to enter into friendship with Alexios, but rather to join him in moving against the capital. And Anna Komnena similarly recalls Alexios’ suspicions that Bohemond and Godfrey were scheming.9 If Bohemond had yet to declare – or even commit to – his peaceful intentions, this would also state why his men clashed with the emperor’s Pecheneg auxiliaries, when they first joined the Via Egnatia.
In the end, Bohemond decided to stick with Urban’s plan. If Godfrey could not be diverted, there was little hope of removing Alexios. The emperor was, however, right to be suspicious of Bohemond’s motives. To allay such fears, Bohemond now left his men in the hands of his nephew Tancred and headed straight to the capital. Both Bohemond and Alexios had reason to be wary; but both needed each other. By means of deft diplomacy, the emperor convinced Bohemond that it was in both of their interests to cooperate; and Bohemond, for his part, committed to the Byzantine cause, swearing an oath of loyalty. This oath reportedly contained a provision that any traditional Byzantine territories would revert to Alexios, but that Bohemond could expect to receive lands of his own beyond Antioch.
For Alexios, such promises were essential. From his perspective, the crusade remained first and foremost a mission to save the empire. More men may have arrived than he’d anticipated – and more fuss made of Jerusalem as their final destination – but the westerners were still exalted auxiliaries, the heirs of Hervé, Robert Crispin and Roussel de Bailleul. He was therefore insistent that they swear oaths of loyalty, committing to restore any old Byzantine territory they were to conquer. For the idealistic incomers, Alexios’ expectations were uncomfortable (and often resisted). Godfrey only agreed to oaths similar to those sworn by Bohemond under pressure, while Raymond of Saint-Gilles refused to swear outright. Bohemond, however, was playing a different game. He had no fear of subordination to Alexios. Previous Norman leaders had taken similar oaths; and even Guiscard had entered into (nominal) Byzantine service when he’d become nobilissimus in 1074. So rather than avoid swearing, Bohemond sought to wring the greatest possible concessions for doing so. Initially, he may have floated the possibility of being made commander-in-chief of the imperial army in the East. In the end, he seems to have extracted the promise that he would receive substantial lands beyond Antioch. Here we see Bohemond’s aims in their purest form: for him, crusade was a means to an end.10
With the oaths sworn (or not, in the case of Raymond), the massive crusading army was ready to lumber into action. Their first target was Nicaea. The city had been lost to the Turks in the early 1080s. In recent years, it had been a thorn in Alexios’ side, enabling attacks deep into the remaining Byzantine lands on the western Anatolian coast. Yet it was not simply Alexios’ interests which dictated the move. If the crusading forces were to receive supplies from Constantinople in the coming months, they would need to control the city. Nicaea was the key to Asia Minor.
Nicaea was a large and important centre, now the capital of Kilij Arslan’s Sultanate of Rum (one of the many splinters of Malik-Shah’s massive empire). Kilij’s forces had already seen off the People’s Crusade the previous year, but the main force presented an altogether more serious challenge. When the crusaders arrived, the sultan was off campaigning against the Danishmends in the East. But he soon returned, attempting a relief of the city on 16 May. This was driven back. Even so, it took another month before Nicaea capitulated. Alexios carefully stage-managed the entire operation. Nicaea’s submission was agreed in advance; however, imperial forces were allowed to scale the walls, creating the impression that the Byzantines had won the day. The concern was that the western troops would otherwise resent the generous terms offered to the city’s inhabitants (most of whom were old Byzantine subjects), particularly after a hard-won siege.
Victory at Nicaea was crucial. It opened the Anatolian plateau to the crusaders; it also marked the first significant expansion of Byzantine territory in a generation. For Alexios, it was an opportunity to cement control of the crusade. A master of carrot-and-stick politics, he now rewarded the army’s leaders generously, securing oaths from many of those who’d previously escaped or resisted these. Among this group was Bohemond’s nephew Tancred, who’d stayed with the main Italo-Norman force when his brother hurried on to the capital.11
As the army marched further inland, it divided in two. The crusading force was far larger than most western and Byzantine armies of the era, and would be easier to provision if divided into constituent parts. Yet division was not without risks. Smaller forces risked being picked off. And sure enough, Kilij Arslan soon sought to exploit the situation, attacking the smaller vanguard led by Bohemond near Dorylaion (Şarhöyük). The fearsome Norman charge initially checked the Turkish attack, but it was the arrival of reinforcements from the second (larger) force that finally turned the tide. The battle had been a close-run thing, particularly for Bohemond and his men. But victory left the crusaders free to march unopposed through central Anatolia.
Alexios’ gamble had paid off – or so it seemed. The main Byzantine enemy, Kilij Arslan, had been decisively defeated; Nicaea had been won; and the rest of the Anatolian plateau would soon follow. Yet with territorial gains came complications. The towns and cities captured reverted to Byzantine control, as per the earlier oaths and agreements. Yet many of the crusaders started to become frustrated by the lack of spoils. The first to part ways with the main force here was Baldwin of Boulogne, the youngest son of Eustace. Baldwin helped secure Tarsos for Alexios against the wishes of Tancred, who’d initially secured the city. Soon thereafter, Baldwin received his quid pro quo, when he took Edessa. Though Baldwin may initially have acknowledged Alexios’ loose overlordship, he clearly had his sights set on greater things, and Edessa would soon become an autonomous polity – the first crusader state in the Middle East.12 It may seem odd for Baldwin to abandon the expedition so early. Yet here we must bear in mind that freeing Jerusalem was just one of the aims of the crusade; only from the perspective of 1099 did this become the defining goal of the expedition. It had been Alexios’ plea which spurred Urban to action, and Baldwin had played a key role in liberating the Christians of Asia Minor. He may well have believed that he could best serve the Christian cause from here.
It’s clear that Alexios remained an active player in developments. For while the crusaders had been bulldozing their way through the Anatolian plateau, his own forces were securing the western and southern coasts. This was a carefully co-ordinated operation, which brought almost all of Asia Minor back under the imperial banner. Yet, with the (partial) exception of Bohemond, the crusaders had not signed up to be imperial stooges. And Baldwin’s own gains provided a model for any looking to break ranks.
These tensions came to a head during the long siege of Antioch. The ancient city lay about halfway between Constantinople and Jerusalem, just south of the fulcrum between Asia Minor and the Middle East, where the crusader army tacked south after its long eastward march. Having made steady progress in 1097, they arrived outside the city walls in October. Antioch was one of the largest and best fortified cities of the Middle Ages, and even the massive crusader army couldn’t invest the entire perimeter at once. Supplies therefore continued to stream in to the defenders. The only options available to the crusaders were to risk assault, hope for an enemy traitor or wait for submission.
The result was a war of attrition – the longest and most trying episode of the First Crusade. The problem for the crusaders was that they were now far beyond Alexios’ supply lines. This meant that while the defenders continued to receive regular sustenance, they did not. In the New Year, they saw off a large relief force from Damascus, but were still left facing starvation. The living resorted to eating their horses; and when these ran out, reportedly turned on their fallen comrades. Many died. Many more deserted, including the Conqueror’s son-in-law, Stephen of Blois.
It was in this context that relations with Alexios finally broke down. Byzantine efforts to secure supplies for their allies repeatedly failed – whether through want of trying or sheer difficulty is unclear. In January 1098, Alexios’ representative, Tatikios, left the army, never to return. Bohemond was quick to exploit the situation.13 In Tatikios’ absence, Bohemond painted Alexios as a traitor, blaming him (not entirely without reason) for the army’s present predicament. Because Alexios had failed to keep his side of the bargain, Bohemond argued, the crusaders were no longer bound by their oaths. This was essential for Bohemond’s purposes, for he now had his eyes set firmly on Antioch, a city he’d committed to restoring to Alexios.
Having undermined the emperor’s authority, Bohemond made contact with the captain of a portion of the wall, an Armenian called Firuz, who agreed to let the crusaders in. Bohemond then suggested to his fellow leaders that whichever of them could secure the city should receive it in reward. Some were wary – and rightly so. Bohemond was not a man to be trusted. But desperate to break the deadlock, they eventually relented.
With all the arrangements in place, Bohemond now made good his plan. He and his men entered the city via Firuz’ section of wall on the night of 2/3 June, then threw open the gates to their fellow crusaders. The result was a bloodbath second only to that witnessed at Jerusalem the following summer. Yet scarcely had the crusaders entered the city than they found themselves facing the prospect of being under siege themselves. For a large force under the command of Kerbogha was rapidly closing on Antioch. Kerbogha was the governor (or atabeg) of Mosul to the east, and the army had originally been intended to relieve the siege. It was much larger than the depleted crusader force; what’s more, it was fresh and well-provisioned. Initially, the crusaders sought to defend the walls. But they were desperately short of provisions and Kerbogha was more than happy to starve them out.
Running out of options, the crusaders decided to risk everything on a battle. Their spirits were raised by the discovery of the Holy Lance – the spear that pierced Christ’s side upon the cross. Not all were convinced of the relic’s authenticity (the German and Byzantine emperors both already claimed to possess the Holy Lance). But for an army in need of a miracle, it provided much-needed encouragement. For three days, the crusaders embarked on a barefoot procession from church to church within the city, entreating divine favour. Spiritually prepared, they then marched out with the lance on 28 June. Initially taken aback, Kerbogha gathered his forces and struck. Yet the nerve of the crusaders held. Facing fiercer resistance than they’d anticipated, the Turks began to panic; and soon enough, they’d broken ranks, leaving the field to the determined westerners. The latter had been heavily outnumbered – perhaps by as much as two to one. Yet thanks to divine intervention – or so it seemed – they’d triumphed against the odds.14
If the crusaders had been united in their travails, they were now divided in victory. For Bohemond insisted on having Antioch turned over to him, as per the promises exacted before the final assault. But other members of the force, most notably Raymond of Saint-Gilles – who may not have been party to the agreement – claimed that these were pre-empted by the earlier oaths to Alexios. For some time, the army remained at an impasse. Yet Bohemond refused to budge, so Raymond eventually accepted the inevitable and moved on with the rest of the princes.
As with Baldwin the previous year, we must be wary of interpreting Bohemond’s actions as a betrayal of the crusade. From the start, the expedition had been a means of acquiring land and influence. And he may, like Baldwin, have felt that he’d done his bit for his fellow Christians. Asia Minor was safely back in Byzantine hands. Bohemond would now hold Antioch, ensuring that Jerusalem, should it be taken, would not be cut off from its Christian allies to the north and west. A significant portion of Bohemond’s force remained with him, establishing new lordships. The resulting principality of Antioch would therefore be the most ‘Norman’ of the crusader states; and its rulers remained Bohemond’s descendants right up until its conquest in 1268 by the Mamluks.15
We should not, however, exaggerate the Norman character of Bohemond’s polity – nor should we underestimate that of the other crusader states. The crusading forces were extremely fluid, with men passing between lords and companies at many points during the campaign. By 1098, many of Bohemond’s men wouldn’t have hailed from southern Italy, while a significant number of those who’d made their way east with him may have joined others. In fact, a considerable degree of Norman influence can be detected within the later crusader kingdom of Jerusalem to Bohemond’s south. Much of this can be put down to the part played by Curthose and his men in the conquest and colonisation of the region. But some must be laid at the door of Tancred, Bohemond and their followers; and more still to later arrivals.16 Beyond Antioch’s name – it was the only crusader state to be named a principality, like the Lombard states Bohemond knew from southern Italy – the Italo-Norman contribution to the crusader states presents something of a paradox. It was clearly significant during the first generation or two of settlement. But thereafter the connection was severed, and the Sicilian realm is notably under-represented in later crusading expeditions. This ambivalence was the result of both accident and design.17 As the movement evolved, crusading became a strongly familial affair, with generation after generation from the same dynasty often participating.18 While in much of France and the Low Countries this encouraged involvement, in southern Italy it had the reverse effect. Here Bohemond had been the black sheep of the Hauteville family, so neither Roger Borsa nor Roger II (Borsa’s successor) had any reason to exalt in his legacy. Moreover, most of Bohemond’s men settled with him in the Middle East, leaving few Italian heirs to follow in their footsteps. The significant Muslim minority in Sicily further complicated matters. Roger II and his successors took a notably flexible and pragmatic approach when it came to relations with these groups and this may have served to dampen crusading zeal.
Political developments were equally important. In 1113, Baldwin of Jerusalem – the same Baldwin who’d begun his career as count of Edessa – would marry Adelaide del Vastro, the dowager-countess of Sicily (Roger I’s wife, the mother of Roger II). Baldwin had long enjoyed good relations with the Normans. He was the son of Eustace of Boulogne, and his first wife, Godevere of Tosny, hailed from the duchy. The match promised to extend these contacts to the Normans of Italy. A condition of the marriage was that any children of Adelaide with Baldwin would succeed him as king; failing that, the throne would pass to Adelaide’s adult son, Roger II of Sicily. The latter detail was crucial. Adelaide was not a young woman at the time – she must have been at least in her late thirties – and Baldwin did not have any children of his own, so there was a good chance that the throne of Jerusalem would soon be left vacant.
Yet Italian influence in the Holy Land would prove fleeting. Over the winter of 1116/17, Baldwin fell severely ill. Faced by the prospect of an absentee monarch, the local magnates rallied together to block Roger II’s succession in preference for that of Baldwin’s older brother, Eustace III of Boulogne (who was a sometime crusader himself). Baldwin himself seems to have been amenable to the plan (relations with Adelaide were strained). And soon after his recovery, Baldwin sent Adelaide rudely packing – without her rich dowry. William of Tyre reports that Roger II then ‘conceived an undying hatred of the kingdom and its inhabitants’, one which lived on among his heirs.19 If in parts of Europe crusading zeal became a familial trait, in Sicily crusading cynicism proved similarly heritable.
Equally significant were developments in Antioch itself. Here the death of Bohemond’s son and eventual heir, Bohemond II, in 1130 created considerable uncertainty. The young prince was in his early twenties and only left a two-year-old daughter, Constance, as heir. In order to stabilise the situation, Constance was married to Raymond of Poitiers – the great-nephew of Raymond of Saint-Gilles – changing the political orientation of the principality significantly. Symptomatic of the shift is the fact that Constance and Raymond gave up claims to Taranto, Bohemond’s original Apulian patrimony, which he and his son had so carefully maintained. But just as we should not exaggerate early Norman influence, so we should not underestimate its longevity. Most of the early settlers remained in place, both in Antioch and Jerusalem; and there is little evidence for the wave of Poitevin newcomers sometimes associated with Raymond. Indeed, marriage made Raymond a Hauteville rather than Constance a Poitevin, and it’s telling that their eldest son bore the distinctively Italo-Norman name Bohemond.20
The precise Norman contribution to the Holy Land remains hard to gauge. From at least the later tenth century, Norman and French identities were not hermetically sealed. Much like a modern inhabitant of Leeds can be a proud Yorkshireman and Englishman, so it was possible to be both Norman and French (or Frankish). Byzantine sources had always called the Normans Franks (much as English sources called them French); Arabic sources now did likewise. And it was as Franks that the crusaders themselves identified. Many hailed from France; and those who did not were mostly Romance speakers with claims to Frankish pedigree. In this context, there was little to distinguish Robert Curthose from Baldwin of Boulogne – or Norman from French (or even Italian) influence.
As we might expect, Bohemond’s own career did not end with the conquest of Antioch. His machinations had won him a major prize, but they’d also earned him a bitter enemy. Antioch had been in Byzantine hands as recently as 1078, and Alexios was not about to roll over for the Norman turncoat. Bohemond, for his part, continued to have ambitions of his own concerning the eastern empire. The years immediately after 1098 were ones of consolidation. In August 1100, however, Bohemond was unexpectedly defeated and captured by the Danishmend Turks. There was much wrangling over his release, during which Tancred ably held the fort in Antioch. Upon Bohemond’s return in 1103, he was able to make little headway with expanding his domains. And after being defeated by the Turks again in the summer of 1104, he decided to change tack.
Bohemond now set out for France, where he began to recruit forces for a ‘crusade’ against Alexios. This was in part a response to the challenge posed by the empire to his new principality. Alexios was Bohemond’s immediate neighbour to the west and was angling for the restitution of Antioch, which he saw (not without reason) as rightfully his own. Bohemond’s move also reflects his longer-term ambitions. In the early 1080s, he and his father had attempted to take Constantinople; and in 1096/7, he’d entertained another strike against Alexios. Now Bohemond made a final bid to topple the eastern emperor. Successful crusaders were treated with awe and reverence upon their return to Europe, and Bohemond was no exception. The crusade had done much to burnish his (already formidable) reputation, and his recruiting efforts met with considerable success. Bohemond was also able to win the hand of Constance, the daughter of the French king Phillip I.21
By 1107, Bohemond was ready to make his move. He had unfinished business with Alexios. And rather than returning to Antioch and striking at the neighbouring Byzantine territories to his north and west, he now attacked at the empire via the Adriatic – just as he had a quarter of a century earlier. As then, Bohemond claimed to be acting in the name of the true Byzantine emperor, this time a purported son of Romanos IV Diogenes. Bohemond’s army was probably larger this time, but his luck was no better.22 For unlike in 1082, Alexios was now in a position of strength. Bohemond launched his invasion from Brindisi in early October, landing at Aulon. After ravaging the countryside, he then prepared to besiege Dyrrachion. The late season, however, meant that siege operations could only begin in earnest the following spring. By this point, Alexios had had ample time to prepare his defence. The emperor now marched west with a large force from Thessaloniki. But rather than seek battle, Alexios blockaded the Normans between his army and the walls. The hunter now became the prey.
Alexios had evidently learned the lessons of previous encounters. It was hard to beat the Normans in battle, but easy enough to outmanoeuvre them. For Bohemond and his men, the situation soon became desperate. By late summer, he was forced to sue for peace. The resulting Treaty of Diabolis saw Bohemond accede to almost all of Alexios’ demands, agreeing to acknowledge Byzantine overlordship of Antioch and restore disputed territories in Cilicia. For the proud Hauteville, this was humiliation. Rather than return to Antioch and see the city pass into Byzantine hands, he therefore headed back to southern Italy, where he died three years later. This was a final snub to Alexios, and it ensured that the terms of the treaty would never be implemented. In Antioch, Bohemond’s nephew Tancred continued to rule in his stead; and eventually, his own son Bohemond II would take the reins. If Bohemond died a failure, his legacy thus lived on. Antioch was to prove one of the longest lived of the crusader states, influencing the politics and society of the Middle East into the late thirteenth century. And Bohemond’s fellow Hautevilles remained a potent force in the central Mediterranean. The only question was, where would they strike next?