CHAPTER 3 Commerce and Conflict from 1087 to 1123

Testing the Zirid-Norman Partnership

During the 1060s and 1070s, the Zirids had seen little success against the Normans in Sicily. Nonetheless, other Zirid expeditions against Christian lords and merchants were more effective. Near the beginning of 1087, for example, Tamim ibn al-Muʿizz sent the Zirid fleet on a raid in the eastern Mediterranean that would have profound consequences for the Zirid-Norman relationship in the coming decades. On this voyage, the navy captured a family of eastern Christian (possibly Armenian) officials who had fallen out of favor with their Byzantine lords.¹ On their arrival in Mahdia, the family requested that they be brought before Tamim, who enlisted their services as accountants. One young administrator from this family, a man named George, rose quickly in the Zirid government and was eventually appointed governor of Sousse, likely around the year 482H (1089–90).² George of Antioch, as he is commonly called today, would come to play a pivotal role in dictating the relationship between the Zirid, Norman, and Fatimid courts on his defection to Palermo in the early twelfth century. For the moment, however, he remained a servant of the Zirid emir.

Other examples of Zirid raiding can be inferred from later sources.³ Geoffrey Malaterra mentions a tense but ultimately nonviolent encounter between the Normans and a Zirid fleet near Taormina in eastern Sicily in 1079. In Geoffrey’s narrative, the Normans are immediately wary of nearby Zirid vessels, which had been “plundering across the sea in the habit of pirates.”⁴ The immediate reaction of Roger I and his soldiers to the presence of Zirid ships near Sicily was to assume they were pirates—an indication of the prevalence of these raiders in the waters around the island. And it was not just the Normans who were wary of Zirid vessels. The pervasiveness of Zirid raiding in the central Mediterranean was such that two rival city-states, Pisa and Genoa, united for an expedition aimed at conquering Mahdia and putting an end to the dynasty.⁵ This 1087 expedition is detailed in a triumphant Latin poem called the Carmen in Victoriam Pisanorum, which explains how the Zirids’ expansive pillaging justified this expedition:⁶

[Tamim]⁷ with his Saracens was devastating Gaul,

He was taking captive all the peoples who hold Spain,

And on every bank of the sea he was disturbing Italy;

He was pillaging Roman territory as far as Alexandria.

There is not a place in the whole world nor island in the sea,

Which the dreadful faithlessness of [Tamim] has not disturbed;

Rhodes, Cyprus, Crete, and at the same time Sardinia,

Were disturbed, and with those, noble Sicily.

Hence the captives were shouting out to the redeemer very loudly,

And through the whole world they were weeping most bitterly;

They cry out to the Pisans with a wretched lamentation;

They were stirring up the Genoese with a mournful cry.⁸

This poem cites the piratical activities of Tamim’s ships, which took Christian captives and pillaged cities across the Mediterranean, as being the driving force for the expedition against Mahdia.⁹ Although this poem is likely exaggerating to some degree the scope of the Zirid menace for the sake of bolstering the mission of the Pisans and Genoese, there is nonetheless a kernel of truth to these claims, as is evident from the unification of two rival city-states for this expedition.¹⁰ This poem indicates that Tamim, after negotiating a truce with the Normans in the late 1070s or early 1080s (likely sometime after the 1079 encounter mentioned in Malaterra), focused his raiding on non-Norman forces like the Pisans and Genoese. As a result, these Italian cities united to retaliate against Zirid aggression. The cause of this expedition was such that Pope Urban II provided spiritual support for the attack, which helped inspire the Pisans to wear pilgrim badges on the expedition nearly a decade before the First Crusade.¹¹

Pisan and Genoese ships on this expedition first arrived at the intermediate island of Pantelleria. Muslims stationed on the island sent carrier pigeons to Mahdia warning of the impending attack, although this was to no avail, for Tamim and his forces were away on campaign elsewhere in Ifriqiya. The subsequent Pisan-Genoese attack on the suburb of Zawila was a successful and bloody affair, as described in the Carmen:

There was no house nor road in all of [Zawila],¹²

Which is not red and discolored with blood.

There were so many wretched corpses of the Saracens,

Which now exhale a stench throughout the one hundred thousand [of them].

The one city is laid waste; they hurry to [Mahdia],

And they stretch out to leap to the high palaces;

Where King [Tamim] was standing, miserable enough,

He who was despising God, as he [thought himself] unconquerable.¹³

The Carmen reports that the attack on Zawila was a devastating success, a perspective confirmed in an Arabic elegy (qaṣīda) that describes the attack from the perspective of the Zirids. It refers to the Pisans and Genoese as “serpents” whose numbers were so great that they seemed like “clouds of locusts” or “swarms of maggots.”¹⁴ Despite their numbers, however, the Pisans and Genose were unable to penetrate the main fortress of Mahdia and instead turned to diplomacy to resolve the stalemate. They first offered the city to Roger I of Sicily, for they were unable to commit the forces to hold it themselves. Due to previous contracts between Roger and the Zirids, he refused, thus “upholding his legal obligations” to the Zirid emir.¹⁵ The Pisans and Genoese then turned to Tamim, who had evidently returned from his expeditions in Ifriqiya, to negotiate a price for leaving the city. The final sum they agreed was substantial, ranging (depending on the source) from thirty thousand to one hundred thousand dinars. This treaty also specified that Tamim was to release his Christian captives and that the Zirid emir would grant favorable trading privileges to the Pisans and Genoese.¹⁶ With these conditions fulfilled, the Pisans and Genoese returned to Italy with the spoils of war.¹⁷ Likely traveling with the victors was one of Tamim’s sons, who was taken as a hostage as part of this treaty to ensure that the Zirids would cease their raiding.¹⁸

The consequences of the 1087 Pisan-Genose campaign for the Zirid dynasty are unclear. There are no surviving documents that attest to the state of the Zirid treasury at the end of the eleventh century, which makes it difficult to quantify the impact of the payment (whatever the sum) and the trading privileges (whatever their details) specified in this treaty. The actions of the Zirid emirs in the years after the attack, however, indicate that the attack did not cripple the dynasty. In 498H (1104–5) Tamim repelled a Byzantine fleet that had sought to prevent a Zirid fleet from leaving Mahdia, presumably because of raids it had been conducting in the eastern Mediterranean. The resulting battle was a victory for the Zirids in which Tamim’s sailors killed “a great number” among the Byzantines.¹⁹

Tamim’s ambitions were not restricted to plundering the Mediterranean; he also campaigned in Ifriqiya to recover lands that had been lost during the reign of his father.²⁰ These expeditions were concentrated on the coastline against the cities of Sousse, Sfax, Gabès, Tunis, and Tripoli, as well as the island of Djerba. Amid these campaigns, Tamim had to contend with some internal dissent, including revolts in Sfax and Gabès, the former of which involved the capture of his son (and eventual successor) Yahya. When the dust settled at the end of Tamim’s reign in 501H (1107–8), the Zirid dynasty held Mahdia, Sousse, Sfax, and the agricultural lands that surrounded these cities. The rest of coastal and inland Ifriqiya was under the control of other dynasties. Although the lands over which Tamim governed were geographically limited, his expansive raiding had brought him into contact with Christian powers across much of the central and eastern Mediterranean.

The Normans also expanded their territorial gains in the late eleventh century, though on a much larger scale than the Zirids. By 1091, Roger I and his Norman allies had conquered the entirety of Sicily as well as the nearby island of Malta.²¹ These conquests put a small group of Norman administrators in control of a diverse population that was primarily Muslim but also contained substantial minorities of Greek Christians and Jews. On the whole, Roger reached peace settlements “swiftly and amicably” with Sicilian Muslims, who were allowed to practice their religion freely.²² Indeed, Muslim soldiers had been integral in assisting the Normans with their conquest of the island, and, as a result, many Muslim administrators remained in powerful government positions in the nascent Norman government.²³

Nonetheless, Muslims under Norman rule still suffered economic hardship because of their confessional identity. The Norman administration imposed on Muslims (and other non-Christians) a poll tax that was reminiscent of the jizya but had considerable variation across the island. Many of the newly installed Christian landlords in Sicily preferred to rent their lands to fellow Christians.²⁴ Muslims on the island also witnessed an infusion of wealth for the Latin church. Throughout the late tenth century, Roger I established bishoprics and sponsored church infrastructure on the island with the approval of the papacy. He also provided incentives for Christian settlers to migrate to Sicily. These changes in the society and political hierarchy of Sicily caused Muslims to emigrate from the island to other Muslim lands (including Ifriqiya), as documented in ecclesiastical registers for churches with records from the eleventh century.²⁵

Alongside this influx of Christian immigrants to Sicily was a change in rhetoric that the Normans brought to their administration, which elevated Roger I as the “restorer of Christianity” to the island.²⁶ A charter from 1087, for example, praises him for leading the campaigns through which “the Christian name is uplifted, the clergy and the Christian people are enlarged, [and] truly, the multitude of Saracens is scattered.”²⁷ Roger’s investment in ecclesiastical infrastructure also brought him substantial authority over clerics in his lands. In an unprecedented move for the rulers of Christendom, Pope Urban II pledged to Roger and his legitimate heirs “not to appoint legates to his lands without his prior consent, to entrust the count himself with the oversight of churches instead of a legate when he felt this was appropriate, and [to give] him the right to veto the attendance of bishops at papal councils.”²⁸ These privileges laid the foundation for later conflict between the Kingdom of Sicily and the papacy when Roger II and later popes both sought to assert their rights over clerics on the island.

In the years after the Norman conquest of Sicily and Malta, relations with the Zirids remained stable as people and goods flowed between the dynasties’ ports.²⁹ An episode from the chronicle of Ibn al-Athir testifies to the strength of their treaties. According to this narrative, envoys from Baldwin of Bologne, a leader of the First Crusade, came to Sicily and informed the Normans of his plan to take Ifriqiya (presumably en route to Jerusalem) and thus to become their neighbor. Roger repudiated this attempted alliance with a massive fart and suggested that this flatulence was a better course of action than invading Ifriqiya. He explains this rationale:

If [the Crusaders] come to me, I shall require vast expenditure and ships to convey them to Ifriqiya and troops of mine also. If they take the territory it will be theirs and resources from Sicily will go to them. I shall be deprived of the money that comes in every year from agricultural revenues.³⁰ If they do not succeed, they will return to my lands and I shall suffer from them. Tamim will say, “You have betrayed me and broken the agreement I have [with you].” Our mutual contracts and visits will be interrupted.… Between me and the people of Ifriqiya, however, are oaths and treaties.³¹

Roger I valued having the Muslim Zirids as his trading partner and was unwilling to involve himself in the conquest of their lands. Nonetheless, Ibn al-Athir quotes Roger I as foreshadowing the eventual Norman conquests of Zirid lands by having him say, “The lands of Ifriqiya will be waiting for us and when we find the strength, we will take it.”³²

Although this story presents an evocative instance of potential collaboration between the Normans and leaders of the First Crusade, it would be unwise to take this story at face value. The meeting between Roger I and the envoys of Baldwin of Bologne occurs only in the chronicle of Ibn al-Athir, who had a concerted goal of presenting the Franks (i.e., Latin Christians) as a monolithic force that sought to invade all the lands of Islam.³³ This story fits all too well with the idea of collusion between different groups of Franks to attack Muslim lands in both North Africa and the Middle East. Even if this story is apocryphal, however, Ibn al-Athir’s inclusion of information about treaties between the Zirids and Normans is evocative. It reinforces the narrative of the 1087 Pisan-Genoese expedition, in which Roger refused to take Mahdia on the basis of his agreements with the Zirids, and solidifies the idea of the Normans valuing their economic alliance of the Zirids over the idea of military conquest. The inclusion of this story in Ibn al-Athir’s chronicle further indicates that the pretense of the Normans refusing to attack the Zirids because of their treaties was (at the very least) feasible to his audience.

The Ascent of Yahya ibn Tamim and the Regency of Roger II

The deaths of Tamim ibn al-Muʿizz in 501H (1107–8) and Roger I in 1101 had little impact on the relationship between the Zirids and Normans. The trading partnership between the two dynasties persisted even as their next generation of rulers, Yahya ibn Tamim and Roger II, both took steps to extend their power in the central Mediterranean. Yahya continued the policy of aggressive Mediterranean raiding that had become a staple of Zirid foreign policy under his father. Meanwhile, Roger II looked to cultivate support for his regime by intervening on behalf of captive monks and asserting control over clerics in Sicily. In the years prior to 511H (1117–18), the most meaningful interaction between the Zirids and Normans was the defection of George of Antioch from Ifriqiya to Sicily, a move that had lasting consequences for the Mediterranean policy of Roger II.

During the late eleventh century, Roger I used his many children, both legitimate and illegitimate, to bind the nascent Norman administration to himself.³⁴ This plan for political centralization was hindered by the death of his last living son in 1091, which made it imperative for Roger to have more male children to continue his bloodline. His third and last wife, Adelaide del Vasto, bore him two male children: Simon (b. 1093) and Roger II (b. 1095). When Count Roger died in 1101, Adelaide became regent to the elder Simon and then, on his death in 1105, to Roger II. Ruling over the entirety of Sicily and Calabria in the name of her youngest son, Adelaide faced the formidable task of maintaining the administration of large and demographically diverse lands. Nonetheless, she was a member of a prominent Ligurian noble family and proved a capable ruler. She suppressed at least one major revolt that cropped up during the beginning of her reign, and she adopted the royal titulature of her late husband with the grandiose title of “the great lady, the queen of Sicily and Calabria, the protector for the religion of Christianity.”³⁵ The decisiveness with which Adelaide governed kept the borders of Sicily and Calabria intact for Roger II during his minority.³⁶

Adelaide’s regency also brought about meaningful political and cultural changes to the Norman administration.³⁷ For the majority of her reign, Adelaide ruled from the eastern port city of Messina. Although Adelaide and Roger II in theory laid claim to the entire island of Sicily, their sphere of influence was largely restricted to the northeastern portion of the island and to southern Calabria in Italy. To broaden the power of her court, Adelaide encouraged Lombard immigrants to settle in Sicily, which continued a process of Latin Christianization that Roger I had encouraged.³⁸ The settlement of Christians in western Sicily paved the way for the transfer of the center of Norman power in Sicily from Messina to Palermo in 1112.³⁹

Roger II grew up in this changing environment, though little is known about the details of his childhood beyond a few likely apocryphal tales. Alexander of Telese, a chronicler who held Roger II in great esteem, foretells the child’s aptitude as a ruler in an anecdote about a fight he had with his older brother Simon: “When they fought, each with [their] companies of boys gathered together, Roger the younger had the upper hand. Whence, deriding his brother Simon, he said, ‘Certainly it is more fitting that I, triumphant, should hold the honor of ruling after the burial of our father than you. Wherefore, when I become master, I will appoint you either as a bishop or even the Roman Pope, for which you are more competent.’ ”⁴⁰ Beyond fanciful stories like these, it is possible to draw some broader conclusions about Roger II’s upbringing and education. Around 1105, Adelaide appointed a man named Christodoulos as emir, a position that made him “a sort of prime minister” of Roger’s lands and a tutor for the young count.⁴¹ Christodoulos was a native Sicilian, likely a Latin-Christian convert from Islam or Greek Orthodoxy, who had been raised in the island’s hybrid Arabo-Greek culture. Under the tutelage of Christodoulos and Adelaide, Roger II was exposed to a variety of languages and cultures that would benefit him as he prepared to govern Sicily and Calabria.⁴² Alexander of Telese also records (in a somewhat less fanciful though still likely embellished account) that a young Roger II was keen to study the inner mechanisms of government. As a youth, he reportedly looked over records of public taxation to ensure that the riches of his treasury were properly spent.⁴³

During Roger II’s formative years, his counterpart in Ifriqiya, Yahya ibn Tamim, was developing a reputation for aggressive naval campaigns in the Mediterranean. Yahya assumed control of the Zirid throne in 501H (1107–8) on the death of his father Tamim. He inherited a small strip of coastal territories in Ifriqiya that included Sfax, Sousse, and Mahdia. Other dynasties governed the other major cities of Ifriqiya: the Banu Khurasan in Tunis, the Banu Dahman in Gabès, the Banu al-Rand in Gafsa, and either the Banu Khazrun or the Banu Matruh in Tripoli (the sources are ambiguous).⁴⁴ Yahya’s right to rule amid these rival dynasties was symbolically bolstered by the Fatimids, to whom he pledged his allegiance at the beginning of his reign and with whom he exchanged ceremonial gifts.⁴⁵

During Yahya’s reign, Zirid galleys attacked Christian powers in Lucania, Salerno, Amalfi, Sardinia, Genoa, and southern France.⁴⁶ The force of these raids was such that, according to Ibn Khaldun, Yahya was able to extract tribute from some of his victims.⁴⁷ Zirid raiding also extended east to Byzantine waters, although these raids were not always successful. During one expedition in 503H (1109–10), for example, the Byzantines intercepted a Zirid fleet and captured six ships. Despite this defeat, Ibn al-Athir reports that the strength of the Zirid military was such that in subsequent years “there was no defeat for the army of Yahya by sea or land.”⁴⁸ A poem written by the Zirid panegyrist Ibn Hamdis in 509H (1115–16) helps to corroborate the extent of this raiding, for it celebrates the arrival of a Byzantine envoy bearing gifts and a request for peace, which Yahya accepted.⁴⁹ Significantly, this extensive Zirid raiding campaign in the central and eastern Mediterranean did not extend to the lands of Roger II. The commercial treaties put in place by Tamim ibn al-Muʿizz and Roger I persisted under their sons at the beginning of the twelfth century. While coastal raiding brought with it the promise of plunder and wealth for the Zirid treasury, in the case of Sicily, this material gain was less lucrative than the previously established trading partnership.

Although the focus of Yahya’s military efforts was on Mediterranean raiding, he also campaigned in Ifriqiya during the first several years of his reign. In 501H (1107–8), he captured the castle of Kelibia on the coast of Cap Bon and, several years later, appointed two of his sons as the governors of Sfax and Sousse.⁵⁰ These appointments proved unpopular, however, as the people of Sfax rebelled against their new governor and sacked the royal palace. It was only by spreading dissent among the ranks of the people that Yahya managed to break their unified front and quash the uprising. Several years later, in 509H (1115–16), Yahya married his daughter Badr al-Dawja to the Hammadid emir al-ʿAziz in an attempt to preserve the truce that his predecessors had cultivated with their cousins in Bougie.⁵¹ As with the Normans, it was in the best interests of the Zirids to maintain friendly relations with their Hammadid neighbors to the west.

The military campaigns that Yahya conducted across the Mediterranean and in Ifriqiya did not preclude his patronage of the arts and sciences. Ibn al-Athir extolls Yahya’s virtues and the court of the Zirid emir: “Yahya was just to his people, firmly in control of affairs of state and a manager of all his circumstances. He was merciful to the weak and the poor, giving much in alms to them. He used to favor the men of religion and learning and he himself was knowledgeable in history, ancient lore, and medicine.”⁵² The apparent magnanimity of Yahya, however, did not extend to all members of his administration. George of Antioch, the eastern Christian official who had been captured and then employed in the Zirid administration, had enjoyed a productive relationship with Tamim ibn al- Muʿizz, but the same was not true of his relationship with Yahya. Tensions between the two first cropped up sometime during the reign of Tamim. According to the Egyptian chronicler al-Maqrizi, George’s younger brother, an adolescent named Simon, had been collecting information about the family of the Zirid emir to convey to his older brother. Simon reported some of this information (presumably to George) that Yahya had said, which angered Yahya such that he ordered that Simon be strangled.⁵³ It is unknown whether this order was carried out (the lack of any mention of Simon in future years indicates that it was), but the end result was a bitter feud between George of Antioch and Yahya.

When Yahya succeeded his father in 1108, George feared for his safety. He wrote to the high-ranking administrator and tutor of Roger II, Christodoulos, requesting aid from the Norman administration. Christodoulos obliged, and George managed to escape aboard a galley that had docked at Mahdia in 502H (1108–9). The defection of George of Antioch to Palermo provided Roger II with a capable administrator who knew the inner workings of the Zirid government in Ifriqiya.⁵⁴ The reaction of Yahya to these events, however, is unexplored in the sources. In the short term, George of Antioch’s flight did not lead to enmity or violence between the courts of Palermo and Mahdia. Perhaps Yahya was glad to be rid of an administrator whom he had suspected of conspiring against him? Whatever his reaction, the trading partnership between the two dynasties remained intact. Regular commercial exchange passing between Zirid and Norman ports was of such importance that the defection of a prominent administrator did not inspire either dynasty to forsake their mutual treaties.

George of Antioch’s flight from Ifriqiya further highlights the degree of connectivity between the courts of Palermo and Mahdia.⁵⁵ George had sufficient knowledge of the Sicilian court and was comfortable enough with its administrators to write to them requesting asylum. Likewise, Sicilian officials knew enough about this Zirid official and his standing in the court to welcome him into their administration. This episode thus provides a glimpse of not only tensions between the Zirid emirs and their administrators, but also the degree to which open lines of communication existed between the Zirids and Normans. It was through these lines that the two dynasties undoubtedly communicated frequently about other economic, political, and military matters—the details of which scarcely survive except in extrapolations made from the writings of officials like Abu al-Salt and Abu al-Dawʾ.⁵⁶

Roger II’s coming of age in 1112 had little impact on the relationship between the courts of Palermo and Mahdia. The first documented interaction between Roger II and an Ifriqiyan lord came in 1114 when Hammadid raiders seized a group of monks who were returning from Sardinia to (presumably) the abbey of Montecassino. According to the Chronicle of Montecassino, on hearing news of the monks’ capture, the abbot of Montecassino sent some men with ransom money to the Hammadids. These men, though, were blown off course and landed in Sicily. When Roger II heard of the monks’ mission, he was “led by love of the most Holy Father Benedict” and intervened on their behalf.⁵⁷ Roger told the Hammadid emir that the monks should be returned to their monastery “if he should desire to delight in his love and to enjoy his peace.”⁵⁸ The Hammadid emir al-ʿAziz obliged and released the monks.⁵⁹

This account shows that the Normans and Hammadids had open diplomatic contact during the early years of Roger II’s reign.⁶⁰ While the Hammadids did not hesitate to seize Christian monks in what was likely one of many state-sponsored raids (reminiscent of their Zirid cousins), they were willing to negotiate with Roger II for their release. The response of the Norman count likewise shows his willingness to cooperate with the Hammadids, though there is a threatening subtext to it as well. If the Hammadids refused to return their captives, would it mean that there would be no peace between them and the Normans? This diplomatic exchange, although it was ultimately nonviolent, contained undertones of potential conflict if the Hammadids did not act as the Norman count wished.

Roger’s decision to help the monks of Montecassino was also a calculated political move meant to show his loyalty to Christian monastic communities despite the tension between him and Pope Paschal II. As mentioned earlier, Pope Urban II had granted Roger I and his heirs unprecedented powers over clerics in their domains. The Normans were given oversight of churches, could veto the attendance of bishops at papal councils, and had the right to voice their approval or disapproval of papal legates to their lands. These expansive rights became the subject of scrutiny early in the reign of Roger II, as both Pope Paschal II and the Norman count jockeyed for control over clerics in Norman lands. In 1112, Roger placed the Latin archbishop of Reggio Calabria under his direct authority. Two years later, the archbishop of Cosenza complained to Pope Paschal II that Roger “had forced him … to return to a monastery.”⁶¹ These assertions of power proved unpopular with Paschal, who sent a sternly worded letter to Roger in 1117 in which he reminded the Norman lord to “not fight against the Church but … to assist it” and warned him to “not judge or suppress bishops but to venerate them as the representatives of God.”⁶² Roger II’s prolonged dispute with the papacy motivated the Norman count to opportunistically intervene on behalf of the captured Montecassino monks so that he could demonstrate his loyalty to Christian clerics even as he looked to exert his authority over them.

This posturing over control of clerics in Sicily was the first round in a decades-long fight between Roger II and Latin Christian spiritual leaders. Across the Strait of Sicily, little is known about Yahya ibn Tamim’s administration until his death under mysterious circumstances on the day of the Festival of Sacrifice in the spring of 509H (1116). The accounts of Yahya’s death are disputed in the chronicles of Ibn ʿIdhari and Ibn al-Athir, but both authors agree that his death had unexpected consequences for the relationship of the Normans and Fatimids decades down the line.⁶³ Ibn ʿIdhari recounts how three men dressed as alchemists mortally wounded Yahya and killed his vizier at the Zirid palace in Mahdia in 509H (1115–16).⁶⁴ While suffering from his wounds, Yahya banished his son al-Futuh to the fortress of Ziyad (located between Mahdia and Sfax) for his suspected role in the plot, then died soon after, on the day of the Festival of Sacrifice. Yahya’s successor ʿAli later exiled al-Futuh and al-Futuh’s family to Egypt.

Ibn al-Athir tells a similar story, set in the year 502H (1108–9), when three alchemists failed to assassinate Yahya but succeeded in killing his adviser, Abu al-Hasan.⁶⁵ Following Yahya’s narrow escape, he exiled his brother (not son) Abu al-Futuh ibn Tamim to the fortress of Ziyad and executed several conspirators, including several men with “the clothing of the people of al-Andalus.” After Yahya’s death in 509H (1115–16), his successor ʿAli sent his uncle Abu al-Futuh, Abu al-Futuh’s wife Ballara, and their son al-ʿAbbas ibn Abi al-Futuh to Egypt. As discussed in chapter 6, this move had profound and unforeseen consequences, for al-ʿAbbas ibn Abi al-Futuh eventually rose through the ranks of the Fatimid court and used his position to undermine the relationship between the Fatimids and Normans.⁶⁶

The overlapping details of the assassination plot against Yahya as provided by Ibn ʿIdhari and Ibn al-Athir highlight the precariousness of the Zirid emir’s administration. Yahya was wary of the machinations of his immediate family. Perhaps this was a well-founded sentiment—he himself had been involved in a revolt against his father’s rule in the 1090s.⁶⁷ While Yahya acted in the tradition of his father by sending his sons to govern cities, he was also distrustful enough of his brother or son (depending on the chronicler) to implicate him as a conspirator in the assassination attempt. In the absence of detailed records about the inner workings of the Zirid administration, anecdotes like these help inform our perception of how tenuous the Zirid emir’s power was over his governors, many of whom were members of his immediate family.

The Affair at Gabès, 511H (1117–18)

Yahya ibn Tamim asserted Zirid power in the central Mediterranean through expansive raiding campaigns until his death in 509H (1115–16). Under the rule of his son ʿAli, the Zirids continued these expeditions and went a step further by trying to monopolize regional maritime commerce against the wishes of the governor of Gabès. This strategy ultimately led to the first instance of conflict between the Zirids and Normans since the conquest of Sicily, which took the form of a naval standoff in 511H (1117–18) outside the city of Gabès. Although Arabic sources differ on the details of the campaign, they agree that this encounter led to a hardening of attitudes between the once-friendly dynasties. ʿAli in particular was so wary of Roger II that he formed an alliance with the Almoravids, a Berber dynasty based in the Maghreb, to invade Sicily. The fallout from this affair at Gabès and the establishment of the Zirid-Almoravid alliance was foundational to tensions between the Zirids and Normans that persisted for the next decade.

During the first several years of his reign, ʿAli sought to expand his influence through a series of campaigns in Ifriqiya. He brokered a treaty with the lords of Djerba in which they agreed to stop raiding local waters.⁶⁸ Soon after, ʿAli forced the governor of Tunis to submit to his rule and conquered Mount Wasilat (located between Tunis and Kairouan).⁶⁹ Then, in the spring of 511H (1117–18), he looked to assert Zirid economic supremacy on the Ifriqiyan coast by interfering in the commercial activities of the lord of Gabès. Ibn al-Athir and al-Tijani provide detailed but differing accounts of this encounter.⁷⁰ Both authors agree that the inciting incident was the construction of a merchant ship by the independent governor of Gabes, Rafiʿ ibn Makan al-Dahmani, against the wishes of the Zirid emir.⁷¹ Fearing further retaliation from the Zirids, Rafiʿ requested aid from Roger II in keeping his ship at sea. Roger obliged and sent his own fleet to Gabès. When ʿAli saw the Norman fleet traveling to Gabès, he retaliated by dispatching his own ships. The two fleets met in the waters off the coast of the city.

The result of this standoff is disputed. Ibn al-Athir relates how the Normans withdrew from this standoff in what was presumably some sort of negotiated truce (he provides no indication of violence). ʿAli then defeated Rafiʿ and his Arab allies in a series of battles outside of Gabès, Mahdia, and Kairouan. At the request of Arab tribes and other nobles in Ifriqiya, ʿAli and Rafiʿ made a peace treaty in which Rafiʿ was permitted to continue governing Gabès. Al-Tijani, however, recounts how the Norman fleet proceeded from the waters off Gabès to the coast of Ifriqiya, where Rafiʿ invited the soldiers into the city for a feast.⁷² It was at this moment that the Zirid squadron arrived and ambushed the banqueters in an act of aggression that ʿAli undertook in defiance of advice from his closest advisers. The result was a decisive victory and the deaths of many Normans. ʿAli followed this victory by defeating his rivals at Gabès, Mahdia, and Kairouan.⁷³

Although the accounts of Ibn al-Athir and al-Tijani differ on the details of this encounter, fortunately for us, these specifics are less important than the overarching consequences, on which the two authors agree: rising estrangement and animosity between the courts of Palermo and Mahdia. Ibn al-Athir relates that soon after the incident, Roger II addressed ʿAli in inappropriate terms and then sent him an insulting letter. ʿAli responded by building up his fleet and planning to invade Sicily with the help of the Almoravid dynasty. The expansion of the Zirid navy and the threat of a Zirid-Almoravid alliance was enough to cause Roger II to back down.⁷⁴ Like Ibn al-Athir, al-Tijani sees the Gabès affair as the reason for increased tensions between Roger II and ʿAli. He goes further, however, and details additional actions taken by the Zirid emir against the Normans. At an unspecified time after his victory at Gabès, ʿAli ordered property belonging to Roger II in Mahdia to be seized and for Norman agents in the city to be imprisoned.⁷⁵ When an envoy from Palermo arrived in Mahdia, ʿAli promptly released the goods and prisoners, which did not satisfy an angered Roger II. The Norman count sent a second envoy to the Zirid emir with a letter containing so many threats and insults that ʿAli did not respond to the message.

Both Ibn al-Athir and al-Tijani write that this episode at Gabès caused increased tension and posturing between the Zirids and Normans. Roger II’s first interaction with the Zirids thus marks a dramatic departure from the more passive policies of his father. He was willing to insert himself directly into Ifriqiyan affairs by intervening on behalf of a local governor against the wishes of the Zirid emir. Likewise, ʿAli ibn Yahya was willing to break with the tradition of his father by engaging in a confrontation with the Normans and attempting to bully them by seizing their goods and agents in Mahdia. The expansionist desires of both dynasties in the 1110s were strong enough to risk the commercial partnership that had defined the Zirid-Norman relationship in past decades.

Central to the Gabès affair was economic and political power in Ifriqiya, and no clear victor emerged in this clash. Emir ʿAli sought to limit the commercial capabilities of the governor of Gabès so that the Zirids of Mahdia could monopolize trade in a section of the Ifriqiyan coast around their capital. Conversely, the plight of Rafiʿ of Gabès provided an opportunity for the Normans to expand their influence across the Strait of Sicily and (possibly) to secure additional commercial contracts with other Ifriqiyan governors. The centrality of economic considerations to the Gabès affair is substantiated in an entry from the chronicle of Ibn ʿIdhari for the year 512H (1118–19), which narrates how envoys from Sicily arrived in Mahdia demanding new contracts for trade.⁷⁶ Although Ibn ʿIdhari does not explore the specifics of this contract, it is likely that the Normans’ demands were tied to the conflict at Gabès that had happened a year prior.

The Gabès affair also shows ʿAli ibn Yahya’s relative power over local lords in Ifriqiya and, to a more limited extent, over the Normans of Sicily. The singular might of the Zirid army was enough to overcome the alliance of Rafiʿ of Gabès and various Arab tribes in Ifriqiya across several land confrontations. Furthermore, after ʿAli’s encounter with the Norman navy (whether violent or not), Roger II did not respond with military force against the Zirids. His preferred course of actions was to taunt the Zirid emirs with inappropriate language and to demand new trading contracts—a strategy that ultimately failed when ʿAli built up his own fleet. Roger was unable or unwilling to confront the Zirid navy in 511H (1117–18). He was also unable to bully the Zirids because of the strength of ʿAli’s fleet and (later) his diplomatic connection with the Almoravids. The picture to emerge from this episode, then, is one in which the Zirids wielded some level of regional power. They defeated a coalition of local lords in Ifriqiya, negotiated or defeated the Norman navy, and renovated their own navy as a show of power that forced Roger II to respect their autonomy.

Events happening around the same time as the affair at Gabès show that Ifriqiya was just one of several areas across the Mediterranean where Roger II sought to expand his influence and legitimize his dynasty. Around 1117 or 1118 (the exact date is unknown), the Norman count married Elvira Alfónsez, daughter of the late Alfonso VI of Castile and León.⁷⁷ At the time of Alfonso VI’s death in 1109, he had claimed the titles of “Emperor of the whole of Spain” and “Emperor of the two religions.”⁷⁸ Alfonso VI had also worked with the papacy to introduce Roman liturgical practices into his kingdom and undertaken successful campaigns against Muslim rulers in al-Andalus during the late eleventh century.⁷⁹ Although Elvira did not hold any lands and had little chance of obtaining them, this marriage still had clear political overtones. Elvira’s family ruled a religiously divided land on the Mediterranean and her father had waged campaigns against Muslim rulers, a situation reminiscent of that of Roger II. Elvira also had demonstrable connections to the esteemed counts of Burgundy in France.⁸⁰ Marrying the daughter of Alfonso VI bolstered Roger’s own standing by connecting him to one of the preeminent families of al-Andalus, and by extension, France.

Roger’s desire to strengthen Norman ties to the powers of the western Mediterranean might have been related to his simultaneous loss of power in the Latin East. Following Roger’s coming of age in 1112, his mother Adelaide married King Baldwin I of Jerusalem on the condition that, should the couple not produce an heir, Roger would inherit the kingdom.⁸¹ The problem with this marriage was that Baldwin was already married to an Armenian woman named Arda of Edessa, a fact that Baldwin’s vassals brought up when he, still heirless, fell ill in the winter of 1116. These vassals, with the help of the patriarch of Jerusalem (and possibly Pope Paschal II), forced Baldwin to end his marriage to Adelaide, who promptly returned to Sicily in 1117. Roger II was furious. William of Tyre relates that this affair fostered in Roger a “mortal hatred” of the Kingdom of Jerusalem, which caused him not to offer aid to the Crusader states during the later years of his reign.⁸² Nonetheless, as we will see, Roger retained an interest in the Crusader states, especially the Principality of Antioch, insofar as he could look to incorporate them into his own domain.

The encounter between the Zirids and Normans at Gabès in 511H (1117–18) should be contextualized within the larger ambitions of both ʿAli ibn Yahya and Roger II. ʿAli had spent the early years of his reign vigorously campaigning in the cities of Ifriqiya, looking to regain the domains that had been lost under his great-grandfather al-Muʿizz ibn Badis. The city of Gabès, which had once been controlled by the Zirid emirs, was one such city over which ʿAli sought to expand his reach. Roger II likewise looked to expand his authority on a much larger, Mediterranean-wide scale. He married into a powerful Iberian family, tried to worm his way into the line of succession of the Kingdom of Jerusalem via another marriage alliance, and sent a navy to the Ifriqiyan coast to intercede on behalf of a local governor. Although Ifriqiya was only one theater of this larger campaign of opportunistic assertiveness that Roger II undertook in the 1110s, it would assume greater importance a decade later due to the looming threat of an alliance between the Zirids and Almoravids.

The Zirid Victory at the Battle of al-Dimas, 517H (1123–24)

Tensions between the Zirids and Normans persisted into the early 1120s. After the Almoravids raided southern Italy with the backing of their Zirid allies, the Normans launched an assault on Ifriqiya in 517H (1123–24). This campaign ended in disaster for Roger II, as much of his army was massacred at the fortress of al-Dimas near Mahdia. For the new Zirid emir, al-Hasan ibn ʿAli, this campaign was a dramatic victory. Two panegyrics written in the aftermath of this victory depict the Zirids as a Muslim dynasty triumphant in the face of Christian aggressors, with the power to unite local tribes under the banner of jihad.⁸³

In the wake of the Zirid-Norman encounter at Gabès in 511H (1117–18), ʿAli ibn Yahya took precautions against Roger II’s rising assertiveness in Ifriqiya by preparing his navy and contacting the Almoravids in Marrakesh about working together to invade Sicily.⁸⁴ The timing was right for such an alliance. The Almoravids had dramatically expanded their lands under the reign of Yushuf ibn Tashfin (d. 1106) to include much of the Maghreb and al-Andalus. The initial decade or so of the reign of his son ʿAli (d. 1143) continued this prosperity to some degree (although the latter half of his reign was marked by the gradual erosion of Almoravid authority).⁸⁵ This prosperity, combined with a long-established tradition of maritime raiding and rhetoric of jihad in Almoravid circles, made the prospect of forming an expedition against the Christian lord of Sicily a welcome premise for the lords of Marrakesh.⁸⁶

Emir ʿAli ibn Yahya died in 515H (1121–22) before this campaign could come to fruition. The attack eventually materialized under ʿAli’s son al-Hasan, who succeeded him at the age of twelve.⁸⁷ An Almoravid fleet raided the city of Nicotera, located on the coast of Calabria in southern Italy, at the beginning of 516H (spring/summer 1122).⁸⁸ The details of the raid are not considered in any medieval sources, though its severity was enough to provoke Count Roger to act against both the Almoravids and their Zirid allies. He mobilized his navy and put an embargo on travel to Ifriqiya and “the lands of the Maghreb.”⁸⁹ Seeing these actions from Roger, for the first time during his reign al-Hasan reached out to the Fatimid imam-caliph al-Amir and pledged his allegiance to him.⁹⁰ In doing this, al-Hasan was acting in the tradition of his father and grandfather, both of whom had pledged allegiance to the Fatimids early in their reigns.⁹¹ Al-Hasan had motivations that went beyond ancestral tradition, however, for he also hoped the Fatimids could assist him against an impending Norman invasion.

The role of the Fatimids as an intermediary between the Zirids and Normans during the early twelfth century is a curious one. In theory, the Zirids remained vassals to the Fatimids, a power dynamic made explicit through lavish ceremonies and the exchange of gifts that inaugurated the reign of a new Zirid emir. The Fatimid caliphate benefited from tribute payments provided by the Zirids and both dynasties benefited from commerce that circulated between their ports. At the same time, however, the Fatimids were unable to provide an essential service to their vassals: protection. The Fatimid imam-caliphs, occupied with campaigns in the Levant against the newly formed Crusader states and troubled by internal unrest, were unable to provide their Ifriqiyan governors with military assistance. The Zirids were left to fend for themselves, as they had done for over one hundred years. When al-Hasan ibn ʿAli provided his ceremonial oath of allegiance to his Fatimid overlord in 516H (1112–23), therefore, he likely hoped that this show of deference would lead the imam-caliph to step in as his diplomatic broker.

Besides the ties that nominally bound the Fatimids to the Zirids, the imam-caliph also had a vested interest in cultivating a positive relationship with the Normans in Sicily. Indeed, the courts of Cairo and Palermo “maintained close, friendly, and regular contact with each other” during the early twelfth century even though the Normans had conquered the island that had once been under Fatimid control.⁹² One of the reasons for this peaceful coexistence was economic. Trade routes that connected Sicily, Ifriqiya, and Egypt provided customs duties for the treasuries of both dynasties. Letters from the Cairo Geniza, although relatively scarce in the twelfth century, nonetheless show continued exchange across this triangular network of Mediterranean trade.⁹³ In addition to this commercial link, a personal connection between administrators in Cairo and Palermo facilitated cooperation between the two dynasties. It requires some degree of explanation.

In the wake of the First Crusade, a number of prominent Armenian Christian families emigrated from Asia Minor and the Levant. Some of them sought employment in Muslim administrations, as was the case with a man named Bahram, who rose through the ranks of the Fatimid government to become a vizier of the imam-caliph.⁹⁴ George of Antioch (who very well may have been Armenian Christian, though his lineage is contested) also found employment in the Muslim world in Ifriqiya.⁹⁵ While under the employ of the Zirids, George of Antioch undoubtedly had contact with their Fatimid overlords in Cairo, which included correspondence with high-ranking administrators like Bahram. When George defected to Sicily, his relationship with Bahram facilitated friendly exchanges between the Fatimids and Normans. Indeed, the chronicler al-Maqrizi mentions that Roger II “dispatched George [of Antioch] as an ambassador to Egypt” many times during his reign.⁹⁶ Diplomatic exchanges between George and Bahram were further complemented by cultural ones, through which the Normans adopted many parts of the Fatimid administration and appropriated elements of its rulership.⁹⁷

It is likely that the Normans had similar nonviolent exchanges with local Ifriqiyan lords like the Zirids and Hammadids. The evidence for this is spotty but evocative. Certain cursive styles of calligraphy, which were used in Sicily during the early to mid twelfth century, were common in contemporary Ifriqiya and the Maghreb but not in Fatimid Egypt.⁹⁸ Diplomatic correspondence between administrators like Abu al-Salt in Ifriqiya and Abu al-Dawʾ in Sicily likely facilitated the Norman adoption of Ifriqiyan writing styles. The Normans were potentially influenced by the architecture of the Hammadid (and then Hilalian) palace of Qalʿa, which contained one of the earliest examples of muqarnas honeycomb vaults in the southern Mediterranean, for some of their own palaces utilized similar vaulting.⁹⁹ The stylings of Muslim Sicilian poets like Ibn Bashrun at the court of Palermo also drew on contemporary literature found in much of the Muslims world, especially in Ifriqiya and the Maghreb.¹⁰⁰ It is thus more than “probable that the contacts between Palermo and the north of the African continent … were multiple,” despite the poor documentation of them.¹⁰¹

Against the backdrop of peaceful cultural exchange between the Zirids, Fatimids, and Normans, however, there was latent political tension. For the Fatimids, who benefited from peace in the central Mediterranean, it was in their best interests to keep the Zirids and Normans from open conflict during the 1120s. Therefore, when al-Hasan ibn ʿAli reached out to his overload in Cairo after the 1122 Almoravid raid on Norman lands, the Fatimid imam-caliph intervened with Count Roger and, according to the thirteenth-century Fatimid historian Ibn Muyassar, restored peace between the two dynasties.¹⁰² The details of this peace are not explored, however, nor are they mentioned in other sources. Indeed, if the imam-caliph al-Amir did manage to broker a truce between the Zirids and Normans at all, it did not last long.

In the summer of 517H (1123) Roger II dispatched a fleet to attack Mahdia.¹⁰³ Al-Hasan had anticipated this attack, though, and gathered a number of “people from the lands of the Arabs” to assist him in the impending battle, which provided him with the military support to contest the Norman military in open conflict for the first time since the affair at Gabès.¹⁰⁴ The Norman fleet, meanwhile, was commanded by Christodoulos and George of Antioch, who conquered the island of Pantelleria before landing on the small island of al-Hashi, about ten miles from Mahdia. The two commanders spent a day sailing around Mahdia, sizing up its fortifications, but when they returned to the island, they found that a Zirid raid had plundered their camp and killed many of their soldiers.

The next day, the remaining Normans set off for the fortress of al-Dimas, which was located near Mahdia on the Ifriqiyan coast.¹⁰⁵ The commanders of the expedition bribed the unnamed Arab commander of the castle and entered it without opposition.¹⁰⁶ This victory quickly soured, however, when the combined forces of al-Hasan and his Arab allies surrounded al-Dimas. The Normans who managed to escape to their ships, including commanders Christodoulos and George of Antioch, were forced to turn back to Sicily. Their comrades at al-Dimas resisted the Zirids and their allies for a week and attempted (unsuccessfully) to negotiate for their lives before they were slaughtered in a doomed sortie out of the fortress. On hearing of the fate of this expedition, a member of the Norman court (possibly Christodoulos) pulled at his beard out of frustration until it bled.¹⁰⁷

The Zirids and Jihad

The Zirid victory at the Battle of al-Dimas was a momentous triumph for the young emir al-Hasan ibn ʿAli. In commemoration of his victory, he sent out a letter to his Muslim neighbors in Ifriqiya. The letter, which survives intact in the travelogue of al-Tijani, is one of the few from the Zirid court of the twelfth century.¹⁰⁸ Written by the Zirid historian and panegyrist Abu al-Salt soon after the victory, this letter portrays the battle as Muslims united against an evil Christian enemy:

The lord of Sicily was obstinate in the tyranny of his trespassing. He persisted in his aggression and his injustice. The evil of [Roger’s] intent and the wickedness of his scheming compelled him to oppress the side of Islam. He thought that this plan would be easy [to accomplish], its desired goal so close. Thus he mobilized and gathered an army. He called upon [his soldiers] to fight and prepared reinforcements. And when [he] thought that his affairs were in order and that he had finished his plan—which was to be his annihilation!—his fleet set out toward Mahdia—God defend it!—with three hundred ships bearing on their decks thirty thousand soldiers and about one thousand cavalry.

But its departure was inauspicious from the start and bound to misfortunes, as preordained for [Roger] was the destruction of equipment and the perdition of souls. For one of the first things with which God rewarded him as part of His most beautiful design, and which He brought forth from His providence, the truth of which is not effected without the deepest gratitude, was to send on them a wind that moved them toward destruction. It came upon them with the cold of the water and the heat of fire. Their destruction befell them, alternating between the piercing of spears and the flashing of blades.…

Then, we sought assistance by summoning the Arab tribes that lived around us. They drew near in band upon band. The arrival of the torrent, which was a very violent commotion and surged in waves, came. All of them came with intentions of pure jihad.¹⁰⁹

The rest of the letter complements the above passage. It gives a detailed description of the Muslim victory at al-Dimas and demonizes the Norman invaders. The letter ends by giving thanks to God, who has “has triumphed for the hand of Islam, elevated it, and granted it victory. He, who has destroyed, ruined, debased, and driven away idolatry.”¹¹⁰

The rhetoric of this letter is clear in its presentation of a Zirid dynasty triumphant over the Normans. First, it extolls the virtues of the Arabs who fought on behalf of al-Hasan. The Zirid emir, mindful of the threat posed by the Normans, sought to keep his coreligionists on his side by highlighting their prowess in battle and mobilizing rhetoric of jihad. This alignment of Muslim powers against the Christian Normans represents a substantial diplomatic change when compared to the affair at Gabès several years earlier. The alliance that had once existed between governor Rafiʿ of Gabès and Arab tribes in 511H (1117–18) against al-Hasan gave way to a new one between al-Hasan and Arab tribes in the summer of 517H (1123). This loyalty was not absolute, however, since an undisclosed sum of money was able to spur the defection of the Arab commander of al-Dimas. This encounter suggests the fluidity of the political landscape in twelfth-century Ifriqiya, which unfortunately is visible to us only in irregular intervals during times of conflict.

Reading between the lines of this panegyric, we can also see how Zirid power was contingent on the emir’s ability to form alliances with other Ifriqiyan powers. This victory was only possible with the help of Arab tribes, whom al-Hasan was eager to praise in his letter, so that he could keep them loyal to his cause. Christodolous and George of Antioch’s ability to bribe the Arab commander of al-Dimas shows the fickleness of the alliances between Ifriqiyan leaders—they could be undone by the exchange of gold. Al-Hasan was well aware of these diplomatic realities and used the rhetoric of jihad to try to bind his coreligionists into a more lasting alliance against the Christian Normans. The Zirids were thus a prominent regional force, but, as will soon be apparent, not one with the divinely manifest power described in this letter.

The importance of the Zirid victory at the Battle of al-Dimas is confirmed in another panegyric—this one written by the poet Ibn Hamdis, who used the occasion of this triumph to advocate for a broader jihad against the Normans in Sicily itself. The content of this poem is best understood in the context of Ibn Hamdis’s life and his personal relationship to Sicily, where he was born around the 1050s. Sometime after the Normans conquered Palermo in 1072, he fled to Seville and lived there until about 1091. While in al-Andalus he “fashioned a poetics of exile, a stylized language of the philosophical and emotional aesthetics of being away from one’s rightful place in the world” through the hybridization of conservative and progressive themes in Arabic literature.¹¹¹ The poetry of Ibn Hamdis thus recalled his homeland and at times used imagery relating to jihad to persuade Muslim lords to reconquer it.¹¹² He brought these themes with him to Ifriqiya and the Zirid court, where he fled after being forced out of Seville due to internal political tensions.¹¹³

The frustration that Ibn Hamdis felt being so geographically close to Sicily yet so far from the island as it had existed during his youth is apparent in his poetry. While under the patronage of the Zirids, Ibn Hamdis wrote a number of poems that advocated for the Zirid conquest of Sicily and made emotional appeals for jihad against the Normans. Around the year 1091, he composed an ode (no. 27) to Tamim ibn al-Muʿizz that laid out his desire for his patron to conquer the island. After lamenting the hardships of his exile and placing blame on infighting between Muslims in Sicily, he juxtaposes the grandeur of Tamim with his own desire to see Sicily freed from the infidel Normans:

My nights in the two Mahdiyyas are like the pearls [that I draw] from your nearness [and arrange] on top of the breastbone,

Nights that passed like pearls arranged in necklaces of the passing years

If I wanted to aim at the moon with a glance, I would see Tamīm in the heavens of glorious deeds!

If my land was free, I would go to her, with a resolve that deems travelling an absolute necessity

But my land, how can I liberate it from the shackles wielded by the hands of the infidel usurpers?¹¹⁴

He continues later in the poem to appeal to the brave emirs of the Zirids to exact revenge on the Normans of Sicily, where so many other Muslims had failed:

When a people strays from guidance, they are upright, how can the fixed stars deviate from their course?

How many of them [are endowed] with true courage, when they turn to attack, they only think of going forward and not on the outcome.

His military expedition is split into two fronts, as one attacks the king in a chess game from each side

When they do not raid the lands of the Rum, they enter the bowels of swift ships on sturdy horses

They would die the death of glory in the thick of battle, while cowards die in the arms of buxom maidens

They stuffed cushions with the dust of holy war, that are placed under their shoulders for burial

They fell, like falling stars fall, into the pit of decay, leaving behind the black of darkness on the world.… ¹¹⁵

When Ibn Hamdis wrote this poem in the early 1090s, he clearly hoped to motivate the Zirid emirs to undertake a holy war against the Normans before they could solidify their power in Sicily. This did not come to pass under the reign of emirs Tamim, Yahya, or ʿAli.¹¹⁶ With the victory of al-Hasan at al-Dimas, however, Ibn Hamdis (now in his seventies) saw an opportunity to extort his patrons to wage a larger conflict against Norman dominion in his homeland by appealing to the teenager who sat atop the Zirid throne.¹¹⁷ Throughout this elegy (qaṣīda), Ibn Hamdis utilizes “both the literal and figurative discourses” of jihad:

God has willed that you have victory,

and that faith destroy what disbelief has erected;

And that He render the pagans disgraced after dispensing His justice on

them; and humiliation and subjugation follow them wherever they go.…

The army of God waged battle on them with the [force of the] wind;

such that no mortal could withstand such an attack.…

Arabs they were, who launched a holy war against the ʿAjam,¹¹⁸

[like] brave lions launching their war against a herd of swine.

Whenever the summons, “Oh you, our protectors,” are voice,

then they rise to the call, in their fold the most honorable noblemen …

Squadrons from every tribe came forth to perform jihad;

there is no excuse for whoever neglects it.

Through them, God has strengthened Muhammad’s religion,

and his divine providence bestowed protection on them.…

[Al-Hasan ibn ʿAli] has protected the [sacred] land of Islam

like a lion devouring with both claws but only receiving a scratch.

A young ruler who conducts himself like one much older;

God forbade that arrogance be connected to him.…¹¹⁹

The remainder of this poem expands on these themes—praising al-Hasan and the Arab tribes that brought him victory, extolling the virtues of jihad, and evoking the evils of the unbelieving Normans.¹²⁰ Although it would be foolish to accept without question the grandiose picture of the Zirid court and its emirs that Ibn Hamdis presents to his readers across his panegyrics, his persistent use of jihad rhetoric is evocative. Using the “well-honed weapon of classical Arabic rhetoric,” Ibn Hamdis intimately tied Zirid campaigns against the Normans (both those that happened in Ifriqiya and those that he hoped would happen in Sicily) to the idea of jihad.¹²¹ His poetry shows that some in the Zirid court sought to present conflict against the Normans as a holy war, and that Ibn Hamdis himself thought that this kind of religiously inspired appeal would find an audience with the Zirid emir and among his court.

In the absence of detailed records from the Zirid court during the twelfth century, Abu al-Salt’s letter and Ibn Hamdis’s poem following the Battle of al-Dimas provide a powerful glimpse into the mentality of the Zirid court as its relationship with the Normans soured. These writings demonstrate the coldness between the Zirid emirs and Roger II that Arabic chronicles mention in passing.¹²² Abu al-Salt’s invective against the lord of Sicily is framed as part of a broader assault on Islam itself, and the victory that the Zirids won over the Normans is thus construed along triumphant, religious lines. Ibn Hamdis, eager to return to his Sicilian homeland, evokes the idea of jihad to encourage al-Hasan ibn ʿAli to undertake future campaigns against the Normans. These letters, although inflated with praise for the lords of Mahdia by virtue of their authors and audience, nonetheless show that in the 1120s, the Zirids saw themselves as triumphant executors of jihad in the face of their Christian foe.

From the establishment of trading contracts at the end of the eleventh century through the Zirid victory at al-Dimas, the relationship between the Zirids and Normans underwent dramatic changes. Initially, the profitability of exchange with the Zirids was such that the Normans under Roger I had declined to participate in expeditions against them. The early twelfth century, however, saw a fracture in this relationship due to the Zirids’ attempts at asserting their strength in Ifriqiya as well as the Normans’ growing political aspirations across the Mediterranean. These initial exchanges favored the Zirid emirs. They repelled a Norman fleet in 511H (1117–18) when Roger II sided with the governor of Gabès against them, and they followed this with a series of victories against a coalition of Arab tribes. Several years later, the Zirids forged an alliance with the Almoravids that led to a raid against Norman territory and a victory at the Battle of al-Dimas, which members of the Zirid court presented as a righteous jihad against unbelievers. The Zirids had the upper hand against the Normans in this clash for supremacy in the central Mediterranean, though this power dynamic would soon unravel in the 1130s.

________________

1. 1. Al-Maqrizi, Kitab al-Muqaffa al-Kabir, 3:18–20. Translations of this entry are found in Adalgisa de Simone, “Il mezzogiorno normanno-svevo visto dall’Islam africano,” in Musca, Mezzogiorno normanno-svevo, 276–85; Johns, Arabic Administration in Norman Sicily, 80–84.

2. 2. Johns speculates that George’s appointment likely came in 482H (1089–90) following an attack on Sousse by Malik ibn ʿAlawi. Johns, Arabic Administration in Norman Sicily, 81.

3. 3. Zirid ships raided unspecified lands in central Italy in 411H (1020–21), but their plunder was seized on their way back to Mahdia by Pisan and Genoese ships. A Pisan raid on the city of Bône in 1034 further implies additional raids undertaken by Zirid (or Hammadid) forces that motivated the Pisans to strike back against them. Idris, Berbérie orientale, 1:125–26, 167.

4. 4. Malaterra, De rebus gestis Rogerii Calabriae, 5:66; Malaterra, Deeds of Count Roger of Calabria, 147–48.

5. 5. Medieval sources disagree on the year of this expedition, although Cowdrey’s analysis of the Latin and Arabic shows that 1087 is most likely the year when it took place. H. E. J. Cowdrey, “The Mahdia Campaign of 1087,” English Historical Review 92, no. 362 (January 1977): 2–8. See also William Heywood, A History of Pisa: Eleventh and Twelfth Centuries (Cambridge: Cambridge University Press, 1921), 34–35; Idris, Berbérie orientale, 1:285–87.

6. 6. The entirety of this poem is found in the original Latin in Cowdrey, “Mahdia Campaign of 1087,” 24–29. English translations and analyses of the poem are found in Matt King, “Perceptions of Islam in the Carmen in Victoriam Pisanorum,” Hortulus 11, no. 2 (Spring 2015), http://bit.ly/2qkXk57; Alasdair C. Grant, “Pisan Perspectives: The Carmen in Victoriam and Holy War, c.1000–1150,” English Historical Review 131, no. 552 (October 2016): 983–1009. See also Charles Dalli, “The Siculo-African Peace and Roger I’s Annexation of Malta in 1091,” in De Triremibus: Festschrift in Honour of Joseph Muscat, ed. Toni Cortis and Timothy Gambin (San Gwann, Malta: Publishers Enterprises Group, 2005), 268–69.

7. 7. The Carmen refers to Tamim as “Timinus.”

8. 8. Verses 7–9 in Cowdrey, “Mahdia Campaign of 1087,” 24.

9. 9. Ibn al-Athir echoes this sentiment. He writes that Tamim’s maritime raids against Christian lands provided the motivation for this attack. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 481H, 10:77.

10. 10. Contemporary poetry from the Zirid court of Tamim reinforces this evidence. The poet Abu Musa, for example, composed eight works about Christian men and women for the Zirid emir, indicating an “unusual authorial predilection” for Christians that was probably tied to Tamim’s raiding and desire for control of Sicily. Nathaniel A. Miller, “Muslim Poets under a Christian King: An Intertextual Reevaluation of Sicilian Arabic Literature under Roger II (1112–54) (Part II),” Mediterranean Studies 28, no. 1 (2020): 69.

11. 11. Cowdrey, “Mahdia Campaign of 1087,” 21–23.

12. 12. Zawila, the suburb of Mahdia, is rendered as “Sibilia” in the Latin text.

13. 13. Verses 39 and 40 in Cowdrey, “Mahdia Campaign of 1087,” 26.

14. 14. This poem was written by Abu al-Hasan ʿAli ibn Muhammad ibn al-Hadad and preserved in al-Tijani, Al-Rihla, 332. See also Heywood, History of Pisa, 38–39.

15. 15. Malaterra, De rebus gestis Rogerii Calabriae, 5:87. Translation is from Malaterra, Deeds of Count Roger of Calabria, 179.

16. 16. Al-Tijani, Al-Rihla, 331–32; Malaterra, De rebus gestis Rogerii Calabriae, 5:87; Malaterra, Deeds of Count Roger of Calabria, 179. Variations in the medieval sources are discussed in Idris, Berbérie orientale, 1:288–90.

17. 17. Karen Rose Mathews, “Other Peoples’ Dishes: Islamic Bacini on Eleventh-Century Churches in Pisa,” Gesta 53, no. 1 (2014): 5–23.

18. 18. An 1126 treaty between Pisa and Amalfi mentions a son of Tamim serving as a town crier in Pisa. It is also possible, as Bonaini suggests, that the son of Tamim fled to Pisa after the death of his father. Francesco Bonaini, “Due carte Pisano-Amalfitane dei secoli xii e xiv,” in Archivio Storico Italiano ser. 3, vol. 8, no. 1 (1868): 3–8; Grant, “Pisan Perspectives,” 1008.

19. 19. This attack is not corroborated in other sources. Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:333. See also Houben, Ruler between East and West, 33.

20. 20. These campaigns are summarized in Idris, Berbérie orientale, 1:291–302.

21. 21. It is unclear who governed Malta when the Normans first conquered it, though the discovery of a Fatimid quarter dinar minted in Mahdia in 1080 indicates that it was under Fatimid control, possibly via their Zirid vassals. Martin R. Zammit, “Ḍuriba bi-Mālṭa ‘Minted in Malta’: Deciphering the Kufic Legend on the Fāṭimid Quarter Dinar,” Melita Classica 3 (2016): 209–15.

22. 22. Metcalfe, Muslims and Christians, 33–39.

23. 23. Birk, Norman Kings of Sicily, 33–80. Ibn al-Athir even reports that the Normans “honored Muslims” after their conquests, Ibn al-Athir, Al-Kamil fi al-Tarikh, year 484H, 10:92. See also Nef, Conquérir et gouverner, 508–9.

24. 24. Overviews of the early years of Norman governance in Sicily can be found in Loud, Age of Robert Guiscard, 173–85; Johns, Arabic Administration in Norman Sicily, 31–62; Metcalfe, Muslims and Christians, 30–54; Nef, Conquérir et gouverner, 237–301.

25. 25. Metcalfe, Muslims of Medieval Italy, 114–24.

26. 26. Loud, Age of Robert Guiscard, 179.

27. 27. Antonio De Amico and Raffaele Starrabba, eds., I diplomi della cattedrale di Messina (Palermo: Tipografia Michele Amente, 1888), 1–2.

28. 28. These privileges are known as “apostolic legation.” Loud emphasizes that, although unprecedented, this papal bull affirmed what was already happening in practice, since Count Roger had substantial authority over the monasteries he founded and funded. Loud, Age of Robert Guiscard, 231–33.

29. 29. Dalli, “Siculo-African Peace,” 269–74.

30. 30. These “agricultural revenues” are literally rendered as “money from the price of crop yields” or māl min thaman al-galāt. I suspect that this statement, however, reflects the state of Zirid-Norman trade more in the middle of the twelfth century than at the end of the eleventh century. Ibn al-Athir is the only chronicler to write of this event, and he did so on the basis of Ibn Shaddad, who was likely not even alive for this supposed encounter. Ibn al-Athir, ever the fan of foreshadowing and showing collusion between Franks, might have been transposing Roger II’s exploitation of Ifriqiyan need for Sicilian grain onto Roger I.

31. 31. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 491H, 10:126. Translation is from Ibn al-Athir, The Chronicle of Ibn Al-Athir for the Crusading Period from al-Kāmil fī’l-Ta’rīkh, trans. D.S. Richards, (Aldershot: Ashgate, 2006), 1:13.

32. 32. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 491H, 10:126. Ibn al-Athir, who frequently foreshadows events relating to Ifriqiya, makes a similar reference to the Normans’ eventual conquest of Ifriqiya when he describes the conquest of Sicily. He writes that the Normans conquered the islands between Mahdia and Sicily and stretched their conquests to include the coast of Ifriqiya. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 484H, 10:92.

33. 33. This theme is explored in chapter 6. D. S. Richards, “Ibn Al-Athīr and the Later Parts of the Kāmil: A Study of Aims and Methods,” in Medieval Historical Writing in the Christian and Islamic Worlds, ed. D. S. Richards (London: School of Oriental and African Studies, 1982), 76–108; Cobb, Race for Paradise, 39–40; Françoise Micheau, “Ibn Al-Athīr,” in Medieval Muslim Historians and the Franks in the Levant, ed. Alex Mallett (Leiden: Brill, 2014), 52–83.

34. 34. Loud, Age of Robert Guiscard, 179.

35. 35. Jeremy Johns, “I titoli arabi dei signori normanni di Sicilia,” Bollettino di Numismatica 6–7 (1986): 13.

36. 36. The extent of civil unrest in Sicily during the regency of Adelaide is disputed. See Hubert Houben, “Adelaide ‘del Vasto’ nella storia del regno normanno di Sicilia,” in Mezzogiorno normanno-svevo: Moansteri e castella, ebrei e musselmani, ed. Hubert Houben (Naples: Liguori, 1996), 81–115; Houben, Ruler between East and West, 24–29; Graham Loud, “Norman Sicily in the Twelfth Century,” in Luscombe and Riley-Smith, New Cambridge Medieval History, 4.2:446.

37. 37. A paucity of administrative records makes it difficult to catalog the bureaucratic changes that accompanied Norman ascent in Sicily. Alex Metcalfe, “Language and the Written Record: Loss, Survival and Revival in Early Norman Sicily,” in Multilingual and Multigraphic Manuscripts and Documents of East and West, ed. Giuseppe Mandalà (Piscataway, NJ: Gorgias Press, 2018), 3–31.

38. 38. Vera von Falkenhausen, “Zur Regentschaft der Gräfin Adelasia del Vasto in Kalabrien und Sizilien (1101–1112),” in Studies in Honour of Cyril Mango, ed. Ihor Ševčenko and Irmgard Hutter (Stuttgart: B. G. Teubner, 1998), 87–115.

39. 39. Johns, Arabic Administration in Norman Sicily, 78. On the development of Palermo and the Normans’ attempts to appropriate spaces in the city to bolster their reign, see Vera von Falkenhausen, “I ceti dirigenti prenormanni al tempo della constituzione degli stati normanni nell’Italia meridionale e in Sicilia,” in Forme di potere e struttura sociale in Italia nel medioevo, ed. Gabriella Rossetti (Bologna: Il Mulino, 1977), 321–77; Theresa Jäckh, “Space and Place in Norman Palermo,” in Jäckh and Kirsch, Urban Dynamics and Transcultural Communication, 67–95.

40. 40. Alexander of Telese, Alexandri Telesini abbatis, 112:7; Graham Loud, Roger II and the Creation of the Kingdom of Sicily (Manchester: Manchester University Press, 2012), 65. Another anecdote about Roger’s childhood is recorded in Snorri Sturlason’s Heimskringla, in which King Sigurd of Norway stops in Sicily en route to the Holy Land. While there, “King Sigurd took the duke by the hand, led him up to the high-seat, and gave him the name of ‘king’ and the right of being king over the realm of Sicily; before, there had been jarls over that realm.” Snorri Sturlason, Heimskringla, or The Lives of the Norse Kings, ed. Erling Monsen and Albert Smith (New York: Dover, 1990), 104.

41. 41. Christodoulos’s title was amiratus, which can be translated as “emir” or “admiral.” This position did not have any fixed duties; instead, the title was an honorific that was used sparingly. Léon-Robert Ménager, Amiratus–Ameras: L’émirat et les origines de l’amirauté (XIe–XIIIe siècles) (Paris: S.E.V.P.E.N., 1960), 20–26; Johns, Arabic Administration in Norman Sicily, 71–74; Adalgisa de Simone, “Note sui titoli arabi di Giorgio di Antiochia,” in Giorgio di Antiochia: L’arte della politica in Sicilia nel XII secolo tra Bisanzio e l’Islam, ed. Mario Re and Christina Rognoni (Palermo: Istituto Siciliano di Studi Bizantini e Neoellenici Bruno Lavagnini, 2009), 283–308.

42. 42. Roger II probably knew some Arabic, Latin, and “some form of Romance dialect” used in the Norman court. The extent to which he knew Greek is unknown, although it is likely he had much less exposure to it than Arabic or Latin. Metcalfe, Muslims and Christians, 103. On the contribution of Greek administrators in the court of Palermo, see Vera von Falkenhausen, “The Greek Presence in Norman Sicily: The Contribution of Archival Material,” in Loud and Metcalfe, Society of Norman Italy, 253–87; Vera von Falkenhausen, “The Graeco-Byzantine Heritage in the Norman Kingdom of Sicily,” in Burkhardt and Foerster, Norman Tradition and Transcultural Heritage, 57–78.

43. 43. Alexander of Telese, Alexandri Telesini abbatis, 112:82. A translation of this text can be found in Johns, Arabic Administration in Norman Sicily, 78.

44. 44. Giovanni Oman, Vassilios Christides, and Clifford E. Bosworth, “Ṭarābulus al-3ẖarb,” EI2; Brett, “City-State in Medieval Ifriqiya,” 81–84.

45. 45. The relationship of Yahya’s predecessor, Tamim, with the Fatimids is relatively unexplored in medieval sources. Johns argues that during the reign of Tamim, “there is only one recorded contact between the two courts, and that of no great significance.” Johns, “Malik Ifrīqiya,” 94. See also Idris, Berbérie orientale, 1:255; Ghāda al-Ḥijjāwī al-Qaddūmī, trans., Book of Gifts and Rarities (Kitāb al-Hadāyā Wa al-Tuḥaf) (Cambridge, MA: Harvard University Press, 1996).

46. 46. Annales Cavenses, MGH, SS, 3:191. This report is collaborated in Ibn Khaldun, Kitab al-ʿIbar, 6:213 and Ibn ʿIdhari, Al-Bayan al-Mughrib, I:336–37, the latter of whom reports that the fleet of Mahdia returned from the lands of the Rum in the year 507H (1113–14) with many captives.

47. 47. Ibn Khaldun, Kitab al-ʿIbar, 6:213. I translate this text with the amendment given for footnote 2 in this edition of Ibn Khaldun’s text, in which the word “Ifriqiya,” which does not make sense in this context, is replaced with al-Franja. Although Louis de Mas Latrie takes this word to mean the lands of southern France, the vagueness of the word al-Franja means that these raids could have been carried out virtually anywhere in the Christian Mediterranean. Mas Latrie, Traités de paix et de commerce, 34.

48. 48. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 503H, 10:224.

49. 49. Ibn Hamdis, Canzoniere di Ibn Ḥamdîs, 405–7. See also Idris, Berbérie orientale, 1:308–9.

50. 50. Ibn al-Athir, year 501H, 10:211. The accounts of Ibn al-Athir and Ibn Khaldun disagree about whom Yahya appointed as governor of Sfax. Ibn al-Athir writes that Yahya appointed his son Abu al-Futuh in 503H (1109–10). Ibn Khaldun, meanwhile, writes that Yahya appointed his son and future successor ʿAli in 508H (1114–15). Ibn al-Athir, Al-Kamil fi al-Tarikh, year 503H, 10:224; Ibn Khaldun, Kitab al-ʿIbar, 6:213.

51. 51. Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:337.

52. 52. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 509H, 10:240. Translation is from Richards, Chronicle of Ibn al-Athir, 1:175.

53. 53. Johns, Arabic Administration in Norman Sicily, 81. The timeline of these events is unclear; they may have happened in the midst of a revolt that Yahya waged against his father during the 1090s. Al-Tijani, Al-Rihla, 51–52; Idris, Berbérie orientale, 1:294–96.

54. 54. Simone, “Mezzogiorno normanno-svevo,” 276–77; Nef, Conquérir et gouverner, 311–14. Details on George of Antioch’s family and influence in the Mediterranean can be found in Catlos, Infidel Kings and Unholy Warriors, 64–68.

55. 55. Nef, Conquérir et gouverner, 592–93.

56. 56. Johns argues that “the two courts were clearly in almost constant contact” during the early twelfth century. Johns, Arabic Administration in Norman Sicily, 89–90.

57. 57. Chronica monasterii Casinensis, MGH, SS, 34:516.

58. 58. Chronica monasterii Casinensis, MGH, SS, 34:516.

59. 59. The Chronicle of Montecassino calls al-ʿAziz the “king of the city of Calama.” This title is derived from the original Hammadid capital of Qalʿa, which, unbeknownst to the author of this chronicle, had not been governed by the Hammadids since the arrival of the Banu Hilal. During the early twelfth century, the Hammadids governed from the coastal city of Bougie.

60. 60. Chalandon, Histoire de la domination normande, 1:369–70; Houben, Ruler between East and West, 32–35.

61. 61. Houben, Ruler between East and West, 34.

62. 62. Philipp Jaffé, ed., Regesta pontificum romanorum: Ab condita ecclesia ad annum post Christum natum MCXCVIII (Berlin: Veit et Socius, 1851), 516.

63. 63. Idris, Berbérie orientale, 1:310–15.

64. 64. Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:338.

65. 65. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 502H, 10:221.

66. 66. Ibn Khallikan, Wafayat al-Aʿyan wa-Anbaʾ Abnaʾ al-Zaman, 6:215–16; Paul M. Cobb, Usama Ibn Munqidh: Warrior Poet of the Age of Crusades (Oxford: Oneworld Publications, 2005), 38–42.

67. 67. Al-Tijani, Al-Rihla, 74–76.

68. 68. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 509H, 10:239–40.

69. 69. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 510H, 10:243–44.

70. 70. Al-Tijani, Al-Rihla, 98–100; Ibn al-Athir, Al-Kamil fi al-Tarikh, year 511H, 10:247. See also Idris, Berbérie orientale, 1:319–24.

71. 71. This narrative is also found in less detail in Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:339–40; Ibn Khaldun, Kitab al-ʿIbar, 6:212–13; al-Nuwayri, Nihayat al-Arab fi Funun al-Adab, 2:164–65.

72. 72. The Zirids and Normans deployed light war galleys for their fleets (called a dromon in certain sources). These ships featured a sail (sometimes more than one) alongside oars for propulsion. The primary goal of engagement for these ships was not to sink opposing ships, but to board and capture them. Pryor refers to them as “really huge rowing shells” because of their ability to hold large numbers of sailors relative to their size. Pryor, Geography, Technology, and War, 65. See also John Pryor and Elizabeth Jeffreys, The Age of the Dromon: The Byzantine Navy c. 500–1204 (Leiden: Brill, 2006); Stanton, Norman Naval Operations, 67–127.

73. 73. A panegyric from Muhammad ibn Bashir (and preserved in al-Tijani) relates the preparations that ʿAli made in building up his imposing forces against the “Rum.” Al-Tijani, Al-Rihla, 99–100. Ibn Khaldun writes that ʿAli defeated the Christian forces of Roger II in battle, though he does not provide additional details on this encounter. Perhaps this lends additional credence to the account of al-Tijani? Ibn Khaldun, Kitab al-ʿIbar, 6:221–22.

74. 74. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 511H, 10:247–48.

75. 75. Al-Tijani, Al-Rihla, 333–35.

76. 76. Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:340.

77. 77. On the marriages of Roger II, see Hayes, Family, Faith, and Empire, 33–74.

78. 78. Bernard Reilly, The Contest of Christian and Muslim Spain, 1031–1157 (Hoboken, NJ: Wiley-Blackwell, 1996), 74–98.

79. 79. Simon Barton, “Spain in the Eleventh Century,” in Luscombe and Riley-Smith, New Cambridge Medieval History 4.2:179–81. See also Bernard Reilly, The Kingdom of León-Castilla under King Alfonso VI, 1065–1109 (Princeton, NJ: Princeton University Press, 1988).

80. 80. Dawn Marie Hayes shows how the Normans would have benefited “significantly from associating themselves with the prestige of Alfonso and his predecessors.” Hayes, Family, Faith, and Empire, 42.

81. 81. Houben, based on the evidence that Patriarch Arnulf of Jerusalem was in Rome in 1116, speculates that Pope Paschal II’s potential involvement in this affair might have caused relations between him and Roger to sour. William of Tyre, PL, 201:507–8; Houben, Ruler between East and West, 29, 35.

82. 82. William of Tyre, PL, 201:519–20.

83. 83. Al-Tijani, Al-Rihla, 337–39.

84. 84. Al-Tijani, Al-Rihla, 98–100; Ibn al-Athir, Al-Kamil fi al-Tarikh, year 511H, 10:247–48.

85. 85. Bennison, Almoravid and Almohad Empires, 54–57.

86. 86. Hicham El Aallaoui, “Les échanges diplomatiques entre Islam et monde latin (milieu XIe–milieu XIIe siècle): La transition entre l’époque des taifas et la dynastie Almoravide,” Oriente Moderno 88, no. 2 (2008): 249–70. See also Ronald Messier, The Almoravids and the Meanings of Jihad (Santa Barbara, CA: Praeger, 2010).

87. 87. In the early years of his reign, al-Hasan relied on the help of older advisers and generals to govern, first a eunuch named Sandal and then a general named Abu ʿAziz Muqaffaq. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 515H, 10:275.

88. 88. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 517H, 10:285–86. A letter from the Cairo Geniza written by Abraham ibn Habib at an unknown date mentions an Almoravid raid consisting of a fleet of seventeen ships, which could be a reference to this 1122 raid or to another that occurred in 1127 (to be considered shortly). Goitein, Mediterranean Society, 1978, 1:308; Simonsohn, Between Scylla and Charybdis, xlii.

89. 89. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 517H, 10:285.

90. 90. Ibn Muyassar relates that al-Hasan sent an envoy to Cairo and pledged his allegiance to the imam-caliph in Jumada I 517H (summer 1123). It is likely, therefore, that this envoy arrived while Roger was building up his fleet but before he attacked the Zirids. Ibn Muyassar, Al-Muntaqa min Akhbar Misr, ed. Ayman Fu’ad Sayyid (Cairo: Institut Français d’Archéologie Orientale, 1981), 93; Johns, “Malik Ifrīqiya,” 94–95.

91. 91. Yahya and ʿAli received Fatimid envoys in 505H (1111–12) and 511H (1117–18). Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:336, 339.

92. 92. Johns, Arabic Administration in Norman Sicily, 267. See also Dominique Valérian, “Les ports d’Ifriqiya et les stratégies maghrébines des califes fatimides dans le Maghreb central,” ed. Annliese Nef and Patrice Cressier, Revue des mondes musulmans et de la Méditerranée: Les Fatimides et la Méditerranée centrale Xe–XIIe siècles 139 (2016): 93–106.

93. 93. See, for example, Simonsohn, Jews in Sicily, documents 171, 176, 177. Goitein, Letters of Medieval Jewish Traders, 239–43.

94. 94. The rise of Bahram in Fatimid Egypt caused friction between the predominately Muslim populace and Christians living under Fatimid rule. The career of Bahram is considered in detail in Seta B. Dadoyan, The Fatimid Armenians: Cultural and Political Interaction in the Near East (Leiden: Brill, 1997), 90–102.

95. 95. The familial networks through which George and Bahram were connected are explored in M. Canard, “Une lettre du calife fatimite al-Hafiz (524–544/1130–1149) à Roger II,” Atti del Convegno Internationale di Studi Ruggeriani 1 (1955): 125–46; Adalgisa de Simone, “Ruggero II e l’Africa islamica,” in Il mezzogiorno normanno-svevo e le crociate: Atti delle quattordicesime giornate normanno-sveve, Bari, 17–20 ottobre 2000 (Bari: Edizioni Dedalo, 2002), 275–85; Johns, Arabic Administration in Norman Sicily, 258–67; Catlos, Infidel Kings and Unholy Warriors, 206–12.

96. 96. Johns, Arabic Administration in Norman Sicily, 81.

97. 97. One striking example of this is the Normans’ use of a ceremonial parasol to shade their kings. It was rare for the Norman kings to be seen in public, and the use of this ostentatious parasol, which was gifted by the Fatimids sometime in the twelfth century, was one such adoption of Egyptian rulership. Ibn Hammad, Histoire des rois `obaïdites: Les califes Fatimides, ed. M. Vonderheyden (Paris: P. Geuthner, 1927), 14–15. See also Britt, “Roger II of Sicily,” 28–29; Jeremy Johns, “I re normanni e i califfi fatimiti: Nuove prospettive su vecchi materiali,” in Del nuovo sulla Sicilia musulmana: Gornata de studio (Rome: Accademia Nazionale dei Lincei, 1995), 9–50; Houben, Ruler between East and West, 113–34; Johns, Arabic Administration in Norman Sicily, 265–66.

98. 98. Jeremy Johns, “Le iscrizioni e le epigrafi in arabo: Una rilettura,” in Nobiles officinae: Perle, filigrane e trame di seta dal Palazzo Reale di Palermo, ed. Maria Andaloro (Catania, Italy: Giuseppe Maimone, 2006), 60–63.

99. 99. Lucien Golvin, “Les plafonds à Muqarnas de la Qal’a des Banû Hammâd et leur influence possible sur l’art de la Sicile à la période normande,” Revue des Mondes Musulmans et de la Méditerranée 17 (1974): 63–69.

100. 100. Miller argues that the “biographical profile of the poets present in Sicily under Roger [II] and, more importantly, their texts themselves, reveal a cadre of administrators in dialogue with other Arabophone Muslims around the Mediterranean, particularly in Mahdiya and the rest of North Africa.” Miller, “Muslim Poets,” 199.

101. 101. Nef, Conquérir et gouverner, 169.

102. 102. Ibn Muyassar, Al-Muntaqa min Akhbar Misr, 93; Johns, “Malik Ifrīqiya,” 94–95.

103. 103. The attack took place in late July 1123. Idris’s narrative notes minor variations between Arabic texts. I prefer the narrative of al-Tijani based on its (ostensibly) direct quotations from Abu al-Salt. Idris, Berbérie orientale, 1:334–42; Johns, Arabic Administration in Norman Sicily, 85–86. On this battle generally, see Simone, “Ruggero II e l’Africa islamica,” 107–13.

104. 104. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 517H, 10:285–86. See also Idris, Berbérie orientale, 1:335; William Granara, “Ibn Hamdis’s al-Dimas Qasida: Memorial to a Fallen Homeland,” in Poetry and History: The Value of Poetry in Reconstructing Arab History, ed. Ramzi Baalbaki, Saleh Said Agha, and Tarif Khalidi (Beirut: American University of Beirut Press, 2011), 252. Johns identifies this island as a small archipelago today called Le Sorelle. Johns, Arabic Administration in Norman Sicily, 85.

105. 105. Idris thinks that this fortress was located on a promontory between Mahdia and Monastir at the ruins of the ancient site of Thapsus. Johns supports this conclusion. Idris, Berbérie orientale, 1:336; Johns, Arabic Administration in Norman Sicily, 85.

106. 106. Al-Tijani, Al-Rihla, 335–36. Ibn al-Athir relates that the Normans took al-Dimas, but he does not mention this bribery. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 517H, 10:285–86.

107. 107. Johns writes that it “is highly possible” that the person who pulled on his beard was Christodoulos. Johns, Arabic Administration in Norman Sicily, 86.

108. 108. Al-Tijani, Al-Rihla, 337–39. Idris wrote a loose translation of this letter in a short 1951 article. Al-Tijani, Voyage du scheikh et-Tidjani; H.R. Idris, “Analyse et Traduction de Deux Textes de l’époque Ziride,” in 70ème Congrès de l’A.F.A.S. (Tunis: Nicolas Bascone & Sauveur Muscat, 1951), 209–16. An English translation of the letter can be found in Brett, “Armies of Ifriqiya,” 111–13.

109. 109. I am grateful to Paul Cobb for his help in translating this portion of al-Tijani’s letter. D. S. Richards states that Ibn Shaddad wrote this victory letter, but the chronicle of Ibn ʿIdhari specifically mentions that Abu al-Salt wrote it. Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:343; Richards, Chronicle of Ibn al-Athir, 1:246. See also Idris, Berbérie orientale, 1:335.

110. 110. Al-Tijani, Al-Rihla, 339.

111. 111. William Granara, “Remaking Muslim Sicily: Ibn Ḥamdīs and the Poetics of Exile,” Edebiyāt 9 (1998): 167. See also Simone, “Mezzogiorno normanno-svevo,” 272–73.

112. 112. Granara calls this rhetorical strategy a “poetics of jihad.” William Granara, Narrating Muslim Sicily: War and Peace in the Medieval Mediterranean World (New York: I. B. Tauris, 2019), 99–141.

113. 113. At the Zirid court, Ibn Hamdis lived in the intellectual network of Ifriqiyan scholars like Abu al-Salt as well as Sicilian administrators like Abu al-Dawʾ. Miller, “Muslim Poets,” 195–96.

114. 114. Translation is from Nicola Carpentieri, “At War with the Age: Ring Composition in Ibn Ḥamdīs No. 27,” in Islamic Sicily: Philological and Literary Essays, ed. Mirella Cassarino (Rome: Istituto per l’Oriente C. A. Nallino, 2015), 43.

115. 115. Carpentieri, “At War with the Age,” 44.

116. 116. Ibn Hamdis composed twelve panegyrics to Yahya and twenty-seven to ʿAli. These poems “consciously manipulate poetic license, or indulgence, to document and eulogize Muslim Sicily’s decline and the rise of Norman supremacy over the central Mediterranean” and in Ifriqiya “undoubtedly found a new and exciting resonance and political immediacy among the highly engaged and concerned audience that was not present in Seville.” Granara, “Ibn Hamdis’s al-Dimas Qasida,” 250.

117. 117. It was not only Ibn Hamdis who evoked this kind of rhetoric. After the Zirid defeat of the Normans outside of Gabès, the poet Muhammad ibn Bashir referred to the Normans as al-jāhilīn (the ignorant). Al-Tijani, Al-Rihla, 99–100.

118. 118. ʿAjam is a pejorative term meaning someone whose native tongue is not Arabic and who did not have the cultured upbringing of a Muslim, Arabic-speaking society.

119. 119. Translation is from Granara, “Ibn Hamdis’s al-Dimas Qasida,” 253–56.

120. 120. Granara argues that the degree of detail present in this poem was evidence of Ibn Hamdis’s awareness of the “reality of Muslim weakness” compared to Roger II, which led him to poeticize and mythologize the victory at al-Dimas. Given the narrative presented throughout this chapter, I find this interpretation unconvincing. Granara, “Ibn Hamdis’s al-Dimas Qasida,” 257.

121. 121. Granara, Narrating Muslim Sicily, 138.

122. 122. Ibn al-Athir, Al-Kamil fi al-Tarikh, year 511H, 10:247–48; Ibn ʿIdhari, Al-Bayan al-Mughrib, 1:342–43.