Negotiating with the “Infidel”: Imperial Expansion and Cross-Confessional Diplomacy in the Early Modern Maghreb (1492–1516)

This article analyses the influence of confessional divides in the construction of a Mediterranean frontier between the Iberian Peninsula and the Maghreb at the very beginning of the early modern period. Questioning the influence that religious difference had on the geopolitics of the early modern Mediterranean could seem superfluous since historians have traditionally depicted the Mediterranean world as a space of confrontation between two confessional empires, the Ottoman and the Habsburg. Nevertheless, by focusing on a selection of diplomatic negotiations from the Western Mediterranean it appears that several actors envisioned a scenario where religious and political frontiers were far from coincide. This article will analyse the diplomatic negotiations promoted by different Muslim communities from the Maghreb to voluntarily enter under the rule of the Catholic Monarchs in the framework of the Spanish imperial expansion at the beginning of the sixteenth century. In studying these negotiations from an actor-based approach my aim is not to deny the religious or the political divide existing between the Christian and the Islamic shores. I will argue, however, that this frontier was constructed through the interaction of a wide array of agents such as local elites, royal officers, military men, religious actors, and rulers, with changing agendas towards religious difference.


Introduction
This article examines how religious divisions influenced the construction of a political frontier between the Iberian Peninsula and the Maghreb at the beginning of the early modern period. 1 I will analyse negotiations between Muslim actors from the Maghreb and the Catholic Monarchs in the period between 1492 (the conquest of Granada) and 1516 (the death of Ferdinand the Catholic), when the Hispanic Monarchy pushed imperial expansion in the Mediterranean. Many Muslim actors from the region chose to cede their sovereignty to the Catholic Monarchs and became their vassals, mainly in order to avoid the violence of military conquest.
Historians working on European overseas expansion have focused on these negotiations as a means by which the Spanish extended imperial power. 2 The important role that cross-confessional diplomacy played in expanding empires is well known. Nevertheless, historians have mostly focused on the opinions of lawyers, or on issues such as the validity of treaties with non-Christian rulers and the links between religious conversion and political subjugation. 3 By focusing on the many actors that were involved in this "diplomacy from below," 4 offering either individually or communally to become vassals of the Catholic Monarchs, we can better understand the limits within which imperial expansion was negotiated. In so doing, we will see that the diplomatic route gave these actors a more complex set of options when dealing with political subjugation and religious difference.
It might seem strange to ask how the religious divide influenced Mediterranean geopolitics in the sixteenth century. Historians have not yet agreed on the delineation of the Mediterranean frontier. Most often, they conceive of the Mediterranean as two different (and opposed) worlds, one Muslim and the other Christian. 5 This aligns perfectly with the traditional image of the early modern Mediterranean world ruled by two confessional empires, the Ottoman and the Habsburg, in which only specific allegiances between Christian and Muslim rulers could take place. 6 Was cross-confessional diplomacy possible in this world? Did actors at the time envision a different Mediterranean world? Rather than placing the phenomenon of cross-confessional diplomacy as a failed option in the history of the Mediterranean, we can look at it as the result of traditions of cross-confessional dialogue among medieval Iberian societies as well as a crucial precedent of some of the key practices characterising Iberian expansion in the Americas, Africa, and Asia. Moreover, showing the complex role that religious factors played in Mediterranean geopolitics, this contribution offers fresh insights into the Mediterranean as a frontier, as it illustrates how religious groups interacted with each other and how they envisioned this particular space. 7 The aim of this article is not, however, to question the existence of a clear religious divide in the early modern Mediterranean world. Rather, it seeks to shed more light on the role that religious difference played in shaping the frontier between the Iberian Peninsula and the Maghreb. Therefore, I will proceed with an in-depth analysis of the different actors and how they conceived of religious difference in negotiations between Muslims and Christians. Adopting an actor-based approach is of paramount importance for this study, as early modern frontiers were not simply defined by central powers. On the contrary, as we will see, the construction of this frontier involved multiple actors. 8 I will address the following questions: who were the actors engaged in cross-confessional diplomacy? What motivated them to negotiate? When and why were they able to bargain? And, how did these negotiations shape the Mediterranean frontier?

Imperial expansion, political cultures and local elites
In the late fifteenth century, the Catholic Monarchs (Queen Isabella of Castile and Ferdinand the Catholic, King of Aragon) expanded their new empire in the Maghreb by conquering some of the most important strongholds on the Mediterranean coast. 9 The authorities of many Maghrebi coastal towns addressed the Spanish authorities, offering their surrender in order to avoid violent conquest (see figure 1). In contrast to what would happen in the Americas, there was no debate on the right of the Catholic Monarchs to conquer territories belonging to "infidels." As a consequence, the Spanish chroniclers and official sources ignored the negotiated character of Spanish expansion in North Africa. 10 The first negotiations took place following the conquest of the Muslim Kingdom of Granada (1492), when two Jews offered to deliver the city and the fortress of Mers-el Kébir, a strategic stronghold in the Kingdom of Tlemcen. 11 The following year, Hernando de Zafra, the royal secretary, who acted on the monarchs' behalf in Granada, received a delegation of notables from Tabaharique and Tiquente, two towns in the same kingdom. 12 The latter offered not only the surrender of both towns, but also Honein, the second port city of the kingdom, as well as Melilla. 13 Between the last years of the fifteenth century and those opening the sixteenth century, negotiations took place in the eastern region of the Maghreb; for instance, in Dellys, close to Algiers. 14 Fear of a Spanish intervention even reached the far east of the Maghreb. The Sheikh of Djerba, a strategic island of present-day Tunisia, chose to surrender to the Catholic Monarchs. 15 In 1497, people of rank in Tripoli proclaimed themselves subjects of the Hispanic Monarchs. 16 Negotiations also took place with the elite of Oran, and it seems that the conquest of the city in 1509 was made possible thanks largely to the collaboration of a group of inhabitants. 17 In 1510, after the conquest of Béjaïa, the elites of Algiers signed a treaty recognising their being subjects of Ferdinand the Catholic. 18 Finally, in 1511, the local elites and inhabitants of Mostaganem, a port city on the Algerian coast, did the same. 19 As we can see, submission to the Hispanic monarchs was clearly an option for Muslim people of rank of the Maghreb. 20 This list of negotiations should not lead us to think that religious divides played no part in regional geopolitics, for in the early modern Maghreb, a strong sense of hostility reigned between different religious groups. It fuelled violent responses when losing sovereignty to a foreign, non-Muslim king. According to the Muslim humanist al-Hasan al-Wazzan (known in Europe as Leo Africanus), popular riots against Christians and Jews were frequent. 21 In Oran, an anti-Christian riot broke out following the conquest of Mers el-Kébir. European merchants who had facilitated negotiations to deliver Oran were victims of popular rage: a mob assaulted and sacked the fonduk, the inn where they lived and stored their merchandise. 22 Gabriel Mas, one of these merchants, lost more than 2,000,000 maravedíes when the mob sacked his shop. He testified: "when Mers el-Kébir was conquered, all the Christians living in Oran where robbed, and two Christians and a woman were killed." 23 Maghrebi authorities were well aware of the risks that negotiations with Christian rulers implied. In 1506, the merchant Nicolao Cattaneo reported to Hernando de Zafra that the mizwār of Tlemcen (the highest military authority of the kingdom) avoided meeting him as the latter sought to maintain his reputation with the local population and feared that he would be accused of selling out the city. 24 The Mizwar's attitude is more than understandable when we consider that during the Muslim revolt of Granada (1501) some of the Muslim notables who surrendered and collaborated with the Catholic Monarchs were punished severely because it was felt that they had sold their communities to Christians. 25 From a legal point of view, according to Islamic law, Muslims were not allowed to conclude treaties with Christians. Inhabitants of territories that came under Christian rule had to emigrate because Muslims were not allowed to live under Christian rule. Muslim lawyers issued fatwas (Islamic legal rulings) saying that those who preferred to live under Christian rule would lost their membership in the Muslim community and that their lives and properties were forfeit. 26 However, from the Middle Ages, lawyers interpreted the law in different ways according to changing relations between Muslim and Christian powers. 27 Some fatwas were issued when the Portuguese took control of some of Morocco's port towns through negotiation, profiting from the internal crisis of royal power in the region. These fatwas can be understood as a clear response to the collaboration of Muslim inhabitants who had helped establish Christian rule and who refused to migrate to territories under Muslim control. 28 While some Muslim lawyers condemned the practice of submission to Christian rulers, others did not hesitate to take part in them. The main qadi (judge applying Islamic law) of Tlemcen participated in a number of negotiations between the king of Tlemcen and the Spanish authorities. 29 In some contexts, the religious divide did not allow for negotiations with Christians. Becoming subject to a crusader such as Cardinal Cisneros was not an option for Muslims living in Oran. Just before the departure of the fleet that conquered this city, Íñigo López de Mendoza, the Count of Tendilla, Captain General of Granada, wrote that the inhabitants of Oran "will prefer to be destroyed rather than to deliver themselves to him because of the Christianity. So everything has to be taken by force, and the conquest will not be so appetising, neither to those who do it, nor to those who suffer it." 30 It is quite meaningful that this experienced warrior in the struggle against the Muslims felt that negotiations were not possible when the expedition had been led by a crusader. Clearly, from Tendilla's viewpoint, negotiations were impossible when the religious rationale behind the conquest was so evident.
Nevertheless, cross-confessional negotiations took place frequently. The most important driving force was fear of Spanish military intervention. In one of his reports, Hernando de Zafra wrote to the Catholic Monarchs in 1494: "That the whole Kingdom [of Tlemcen] is trembling with the keys in the hand because [these peoples] neither have force to defend themselves, nor talk about any other solution, except in coming to the service of Your Highness." 31 Zafra's words were too optimistic and we have to be cautious reading them. The political career of this royal secretary was closely associated with the conquest of Granada. As a consequence, he tried to convince the monarchs to embrace similar enterprises that, in addition to enlarging the monarchy, would have improved his career prospects. The expansion of the Ottoman Empire into Mamluk Egypt and the Maghreb in 1516-17 underscores the fact that most of these negotiations were motivated by the fear of violent conquest. The case of Algiers is a prime example. As we have seen before, in 1511, the local elite of this city signed a treaty to submit to the sovereignty of the Hispanic Monarchy. Nevertheless, when Barbarossa and his military forces appeared in the Maghreb five years later, they changed their political loyalty and put themselves under the sovereignty of the Ottoman Empire. 32 In fact, when Spanish military pressure lessened, these diplomatic contacts became less frequent, as surrender to the Ottoman rulers proved effective in preventing Spanish conquest.
Fear was an important factor in the expansion of the Hispanic Monarchy, but it was not the only element that motivated Maghrebi rulers to negotiate surrender. Divisions within Maghreb society also played an important role. 33 Due to the endless struggle between different local kings, some Maghrebi elites preferred the distant sovereignty of the Catholic Monarchs. According to Leo Africanus, the local men of rank of Annaba, a port town on the Algerian coast, killed the local governor who had been appointed by the king of Tunis, and threatened the monarch by communicating to him their willingness to deliver the town to the Christians. 34 The same principle also applied for local rulers. The count of Tendilla designed a plan to take hold of Vélez de la Gomera, the most dangerous pirate's lair on the Maghreb coast, by offering to its king, Abu 'Ali Mansur, military support in his fight against the king of Fez. 35 Fear in the face of aggression or gaining external help against local enemies fuelled many negotiations. However, it is worth noting that they were initiated by Christian and Muslim societies alike. The political culture of Maghreb societies allowed surrender to a foreign king. Furthermore, tribal groups, urban elites, and frontier societies enjoyed a great amount of liberty when dealing with a weak monarchical power. In fact, they were able to withdraw their allegiances and search for a new sovereign who could offer peace, social stability, and protection. 36 On the Christian side, the imperial project inherited from the Reconquista the option of ruling over diverse religious communities. 37 Cross-confessional diplomacy had been a common phenomenon in medieval Iberia, especially in concluding war and political disputes. Thus, the political culture of late medieval Iberia provided a framework in which to develop practices of cross-confessional diplomacy for the Maghreb. 38 This was equally true for the Muslim side of the Mediterranean frontier. 39 In 1516, the sheikh and local elites of Algiers sent a letter to the Hispanic monarchs asking for protection from raids by Christian corsairs. The correspondence shows that they had mastered the language and codes of Spanish administration. After having presented themselves as "the Sheikh and Moors of your city of Algiers, vassals of Your Highness," they denounced an attack by corsairs from Malaga who, after being welcomed and supplied in Algiers, had sacked some villages nearby. The attack was considered illegal as the inhabitants were "vassals of Your Highness, who pay your rents and rights, and are under the guard and protection of Your Highness." The sheikh made use of specific and efficient administrative channels within the Spanish bureaucracy. Ferdinand the Catholic wrote to Malaga's local authorities ordering the release of Muslims enslaved during this raid. As the local authorities ignored the order, the sheikh insisted on referring to the Spanish monarch's obligation to keep "the capitulation made with them, when they delivered themselves as vassals of Your Highness." 40 The king sent a letter to Malaga's authorities ordering them to satisfy the petition of his Muslim vassals, but, unfortunately, Ferdinand's death and Spain's ensuing internal crisis meant that their demands were ignored. In 1519, the local notables used the same argument (political subjection in exchange of protection) in a letter sent to Sultan Selim I, asking to enter under the suzerainty of the Ottoman Empire. 41 The parallels between the letters to the Spanish king and the ottoman sultan show not only the pragmatism of the Algerians, but also that for these local elites, the same principle ruled the negotiations, regardless of whether they wanted to enter a Christian or a Muslim empire.

Religious difference in an expansionist war
Military officials supported cross-confessional diplomacy as a means to expand the Hispanic Monarchy's rule in the Maghreb. The frontier between the Iberian Peninsula and the Maghreb had been subject to a long tradition of conquest warfare, 42 as the example of Malaga (1487) shows, and Muslim and Christian officers shared a common military culture. Muslim Malaga had resisted the attacks of the Castilian army during the last phase of the war against Granada. As a result, when the city capitulated after a long siege, the Castilians applied the laws of siege warfare: they enslaved Malaga's inhabitants with the exception of those who had negotiated the final surrender. The Maghrebi garrison that had refused surrender months before was executed. 43 The echoes of this event facilitated later negotiations in the Maghreb, because Muslim military men had learned from the consequences of ignoring honourable surrender.
This shared military culture was put into practice by military leaders such as Diego Fernández de Córdoba, the Alcaide de los Donceles. During the campaign against Mers-el Kébir, he only needed a couple of hours to successfully negotiate the surrender with the fortress's governor. It is worth noting that the Alcaide treated the Muslim garrison according to the same rules used in European siege warfare. 44 He not only negotiated military surrender, in fact, it was in part through his position in the region that diplomatic negotiations were used to expand the dominion of the Hispanic Monarchy into the Maghreb. In his correspondence with Ferdinand the Catholic, the Alcaide urged the king to take advantage of the negotiations with the Maghrebi Muslim elites, who were asking to enter the sovereignty of the Hispanic Monarchy. 45 Later on, the Alcaide took part in the negotiations with the Oranian urban elite to deliver that city. 46 The captain general of Granada, the aforementioned Count of Tendilla, promoted and organised several missions to the Maghreb to facilitate subjection. Prior to the conquest of Oran, Tendilla wrote to Bernardino Gómez de Sandoval, the Marquis of Denia, describing this expedition as "the biggest madness I have ever seen, because who goes to conquer, should have some hope of taking something by deal, and here it is missing." 47 Tendilla's letter must be read in terms of his usual criticism of his political enemy: Cardinal Cisneros, who directed the expedition against Oran, as we will see. 48 Nevertheless, with these words this experienced military official was also expressing his concerns about the role that cross-confessional diplomacy played in the development of military campaigns.
For military leaders, negotiating with Muslim rulers was considered useful in the war against them as an alternative strategy to warfare and conquest. When Tendilla referred to his contacts with the king of Velez (negotiations that he conducted as captain general of Granada) he wrote: "I want peace until it could be possible to make war." 49 During his expedition against Mers el-Kébir, Tendilla advised "to not care about agreements" because the Spanish army was, according to him, able to conquer this stronghold without difficulty. 50 Unfortunately, the Maghrebi military authorities have not provided us with sources reflecting how they viewed these negotiations. However, we can trace their participation by reading Iberian sources. According to the chronicler Lorenzo de Padilla, who worked as a spy in the Maghreb, the conquest of Mers el-Kébir was planned when the Muslim governor of the fortress addressed an agent of Ferdinand the Catholic offering him this strategic enclave. The sudden death of this governor at the beginning of the campaign thwarted the peaceful inclusion of Mers el-Kébir.
Nevertheless, as mentioned before, after a few hours of siege the Alcaide de los Donceles, Captain General of the Spanish armed force, was able to negotiate a surrender with the new governor. 51 In this case, the Muslim military officials were pushed to reach an agreement with their Christian enemies due to their military inferiority, although violence was not the only factor fuelling the negotiations between Muslim and Christian military officers.
Frequently the use of arms was combined with payments to Muslim military commanders. The first governor of Mers el-Kébir would have received a substantial sum of money in exchange for the delivery of the fortress. He died before he could receive his compensation, but the Alcaide de los Donceles gave Aben Dali, the new governor, 300 ducats as payment for delivering the fortress. 52 The case of Mers el-Kébir was far from unique, as many other Muslim military officials were persuaded to deliver their strongholds in exchange for large sums of money. In 1504, the Venetian Jerónimo Vianello, unsuccessfully negotiated an agreement to obtain the Muslim fortress in the islet of Algiers. On Tendilla's behalf, Vianello offered an annual salary of 1,500 ducats to Hamad Beton, governor of the fortress. Vianello's offer also included a royal licence allowing Beton and his successors to live as Muslims in the Catholic Monarchs' domains. 53 Christian and Muslim military leaders were of one mind that personal gain could facilitate surrender.
The list of Muslim military officers engaged in cross-confessional diplomacy not only included governors of fortresses under the threat of attack by Christian forces, but also high-ranking Muslim military officials. In 1506, Hernando de Zafra established contacts with the mizwār of the kingdom of Tlemcen. 54 In his letter, Zafra invited him to enter the service of King Ferdinand. Zafra included a veiled threat when he advised the mizwār to think carefully and to do "what really is convenient to you and to the lord king your master, and to all the Moors of that kingdom, because by doing the opposite you will cause the destruction of the king and your own." 55 We do not know the mizwār's response to Zafra's letter, but we have evidence that some years later this Muslim officer was engaged in negotiations with Ferdinand the Catholic. In fact, shortly after the conquest of Mers el-Kébir, the Alcaide de los Donceles established negotiations with the mizwār through his brother-in-law. 56 Muslim military officers such as governors of strategic fortresses and high ranking military officers like the mizwār were responsible for the defence of the kingdom. However, it was precisely these officials who engaged in negotiations with Christian actors to facilitate their subjection to the Catholic Monarchs.
We can conclude that in some cases neither Christian nor Muslim military leaders paid much attention to religious differences when negotiating surrendering or changing allegiance, because both sides ascribed to the same kind of logic when conducting these negotiations. Again, religious difference was not an insurmountable obstacle. Nevertheless, it must be emphasized that military men needed some motivation to negotiate. In other words, negotiations took place when it was convenient for strategic reasons, or when it was possible to gain from it financially. Negotiating the delivery of a stronghold or a city was permissible for Muslim officers such as Aben Dali, Hamad Beton, or the mizwār, but only when a painful military defeat had to be avoided. The case of Mers-el Kébir is a clear proof of this, as is that of Algiers. The conversations between Beton and Vianello took place against the backdrop of the preparation of the Spanish fleet in Malaga, but the negotiations ceased when it became clear that the attack was directed against another target. Something similar happened regarding the aforementioned talks between the Alcaide de los Donceles and the mizwār. Once again, the main factor fuelling agreements between Muslim and Christian military officers was the fear of armed violence.

Religious agents in cross-border negotiations
After military leaders, religious actors were the most important intermediaries in cross-confessional diplomacy. For instance, Juan Rena, a Venetian priest, negotiated with Muslim rulers on behalf of Spanish and Muslim authorities. In 1504, he participated in the negotiations with the king of Velez, as well as in those concerning the delivery of Oran. 57 The involvement of Christian religious agents in cross-cultural diplomacy went beyond the participation of priests such as Rena. While dignitaries of the church have to be considered promoters of the wars against Muslims, 58 even the most zealous understood that diplomacy had a role to play in the art of warfare, as the case of Cardinal Fray Francisco Jiménez de Cisneros shows. Cisneros had a prominent political career, first as Queen Isabella's confessor and later as archbishop of Toledo, the most important religious dignitary of the Hispanic Monarchy. He employed his political influence as well as the church's economic resources to ensure the conquest of Oran, which he saw as a first step in his personal crusade to reach Jerusalem. 59 Despite the ambitious nature of his plan, Cisneros was pragmatic, and he considered the possibility of accomplishing his objectives by sword as well as through diplomatic negotiations. Cisneros used Franciscan friars to negotiate with Muslims in Oran, 60 as Tendilla's letter indicates: "That lord [Cisneros] is taking of everything, he has sent to recruit all the friars algarabiados [friars who speak Arabic] because he says that he needs them for this enterprise. Look carefully at the riddle; he had to become pope, I do not know why he is becoming sultan." 61 Despites Tendilla's ironic criticism, Cisneros's measure proved to be very efficient, as we can conclude just by looking at the activity of one of his agents' in the Maghreb: Fray Jorge de Benavides. 62 After the conquest of Oran, this Franciscan friar successfully negotiated with the local governors, as well as some notables of Mostaganem and Maçagranil for the delivery of both cities. 63 At more or less the same time, he was engaged in negotiations with the king of Tenes, who ruled over a little coastal territory. 64 It seems that this friar was quite convincing in these negotiations, as he managed to obtain the submission of the Muslim authorities. In a letter to the cardinal, another friar described Fray Jorge as the perfect instrument for Cisneros's crusade: The tools that His High Reverence has are two: that God has chosen and selected you for conducting war against these enemies of the faith, and second: having the interpreter that has to tell him the truth dressing his habit, that is the father Fray Jorge, who has credit in these parts of Africa, and even in the Sahara. And with him going in the front side, that he has to keep true, and the Moors trust him, you could achieve even the house of Jerusalem. 65 The words of this friar show the extent to which religious agents became involved in the Hispanic Monarchy's imperial expansion in the Maghreb. Negotiating with Muslims was of a paramount importance and for that reason it was necessary to use skilful agents such as Fray Jorge. 66 Again, we have less information about the attitudes of Muslim agents towards cross-cultural diplomacy due to the lack of sources from the Maghreb. Nevertheless, we have some insights into their negotiations with Christian authorities. Spiritual leaders such as qadis and marabouts (holy men or leaders of religious communities) were also important political authorities recognised as such by the inhabitants of rural and urban areas where central authorities had no real power. 67 The reasons why some Islamic religious leaders agreed to take part in these negotiations differed from case to case. A marabout called Cidi Alhacen obtained the liberty of his enslaved son thanks to his services to the Spanish crown in this diplomatic effort. Another marabout came to the service of Ferdinand the Catholic acting as a hostage in negotiations that concerned the subjection of some populations near Oran. 68 At the subjection of the city of Mostaganem, one of the treaty's signatories was a marabout called Cidi Abdille. 69 Why these religious leaders collaborated and assisted the Catholic Monarchs to increase their hold on the Maghreb seems clear. In the first case, the marabout wanted to protect his family, while in the other two cases the spiritual leaders sought to protect their communities. Nevertheless, the engagement of Muslim religious agents was not only a response to the violence deployed by the Spanish authorities. Sometimes the engagement of these actors was the outcome of their active role in the internal struggle for power in Maghrebi society, as we can conclude from the negotiations that delivered Oran. In his report on the Oranian notables, Nicolao Cattaneo, one of the Genoese merchants involved in this effort, referred to an important religious leader as follows: "[My brother] told me that he has in his hand a great faqīh, the most important of this kingdom, to whom Arabs, Berbers, and citizens obey more than to the king. I know his name because he is the one who made possible to conquest Tenez and he is a great enemy of the Mizwār." 70 Reading Cattaneo's words we can conclude first, that this "great faqīh" (an expert in Islamic law) enjoyed a widespread trust among the different groups in this local society, and second, that his engagement in the negotiations with the Christians to deliver Oran was due to his conflict with the royal authorities. Hence, negotiating submission to a non-Muslim ruler was for some Muslim religious leaders not only a way to protect the community; it could also serve to reinforce their position vis-à-vis other authorities with whom they competed. 71 In order to get a better understanding of the role played by religious agents in diplomatic negotiations, we have to take into consideration the third religious community present: the Jews. 72 The role of Jews as mediators in negotiations between Christian and Muslims during the early modern period is known to some extent. 73 However, little attention has been paid to their role in the expansion of the Hispanic Monarchy. It is worth noting that the first negotiation to deliver a Muslim city to the Catholic Monarchs was conducted by David Segura and Abulafia, two prominent members of a group of Sephardic Jews who had been expelled from Spain. 74 Even if they did not succeed in achieving the delivery of the city, this case demonstrates that leading actors of the Jewish communities were involved in cross-confessional diplomacy. According to their contract with the Catholic Monarchs, Segura and Abulafia would have received 10,000 ducats if they managed to obtain the delivery of Mers el-Kébir. Nevertheless, The most important concession would have been the kings' offer of protection to the entire Jewish community in the region and the freedom to practice their faith in Mers el-Kébir. 75 Personal and collective benefices continued to sustain contacts between the Hispanic Monarchy's authorities and leaders of the Jewish community, and thus encouraged the latter's engagement in the Maghreb's cross-confessional diplomacy. 76 To better understand these actor's participation in the negotiations, we have to take into consideration that Jewish communities were often looking for a place to settle in the framework of the Maghrebi social fabric. 77 This explains why Jewish actors helped establish Christian rule in this area. In doing so they secured a place for their communities on the frontier of the Hispanic Monarchy. Thus, in Oran, rabbi Symuel Zatorra 78 took part in the negotiations that delivered Oran to Christian authorities. 79 The local Jewish community had suffered from the violence of the Spanish conquest; later, however, they were allowed to live in Oran as the Spanish authorities considered their presence necessary to maintain the city. 80 This case was rather exceptional throughout the Spanish Empire.
Christian, Muslim, and Jewish leaders clearly had different interests in negotiating with the "other": Christians were intent on conducting a crusade to reach Jerusalem; the Oranian faqīh aimed to fight the influence of his political enemy; and the marabouts only wanted to protect their followers. Zatorra and other leaders of the Jewish communities who engaged in cross-confessional diplomacy in the Maghreb region were looking for a place for their communities to settle within the Spanish Empire. Nonetheless, the three faiths agreed on the necessity of negotiating with "unbelievers" whenever it benefited their communities.

The rulers' viewpoint
Local elites, military leaders and religious agents played a vital role in the construction of the frontier between the Iberian Peninsula and the Maghreb. Nevertheless, they were not the only actors in this process. The attitudes of the monarch and royal authorities towards negotiations over the transmission of sovereignty over Muslim populations need to be considered. When the king of Fez was notified of the negotiations to deliver Melilla to the Catholic Monarchs, he reacted by killing families of the negotiations, thus sending those engaged in these conversations a violent reminder of his power. 81 His reaction was a punishment for what he rightly considered a direct attack on his authority. The men who negotiated the agreement belonged to a faction of the local elite opposing the king, and three months before starting negotiations with the Catholic Monarchs, the town's Muslim elite had expelled the officer who had represented royal authority. 82 Other Muslim monarchs such as the king of Tlemcen adopted a more pragmatic attitude. Having lost the port city of Mers el-Kébir and with the elites of Oran negotiating the surrender of their city to the Christian authorities, this Muslim ruler decided to establish formal negotiations to become a "king vassal" of Ferdinand the Catholic. In his letter, the Muslim king addressed Ferdinand the Catholic as "king of kings" and expressed "his complete willingness to have, between you and me, [a] strong alliance and a formal homage" as a sign of submission. 83 The king of Dudu, a local ruler who controlled the frontier between the kingdoms of Tlemcen and Fez, eadopted a similarly pragmatic approach. When some of the coastal population of this region started to negotiate their delivery, the king declared his willingness to become King Ferdinand's vassal. 84 As we can see, the attitude of some Muslim rulers towards cross-confessional diplomacy often depended on their own ability to impose their authority over their subjects and to resist the expansion of the Catholic Monarchs.
Some rulers from the Maghreb considered the allegiance with the Catholic Monarchs an appropriate way to reinforce their position against more powerful Muslim rulers; thus they profited from negotiations between Christian authorities and other local notables. In the kingdom of Fez, a tribal chief entitled "sheikh of sheikhs" who controlled the coastal area around Melilla, addressed Hernando de Zafra and offered the subjection of his dominions. 85 Some years later the royal secretary conducted similar conversations with Sheikh Mahamet Aben Zuley, a prominent tribal leader from the kingdom of Tlemcen, to entice him into the service of King Ferdinand. 86 Needless to say, the Catholic Monarchs have something to tell us about the limits of this cross-confessional diplomacy. They justified their expansionist ambitions beyond the Mediterranean Sea as a holy war against the enemies of the Christian faith. 87 Nevertheless, from the beginning they showed much willingness to use diplomacy as a means of expanding their authority into the Maghreb. This becomes evident from the generous concessions made to their diplomatic brokers. Ferdinand the Catholic showed a real interest in the matter, as he ordered his agents to keep him informed about every single aspect regarding negotiations. 88 However, the attitude of the Catholic Monarchs to these negotiations evolved with time. After negotiations failed in Melilla and with other coastal populations, King Ferdinand became sceptical of the merits of this diplomatic activity. This becomes evident from the correspondence between him and Hernando de Zafra, who was much more enthusiastic about possible outcomes than the king. 89 At times Zafra bitterly complained about his sovereign's lack of interest even as he asked the Catholic Monarchs for instructions on how to deal with offers and envoys arriving from the Maghreb. 90 In the following years, Ferdinand the Catholic turned a cold shoulder towards cross-confessional diplomacy in the Maghreb. When the king of Tlemcen tried to negotiate his own submission to Ferdinand, the latter offered him only a brief truce in exchange for the payment of tribute and the freedom of all the Christian captives. 91 Ferdinand's scepticism was even more evident during the negotiations after the conquest of Mers el-Kébir, when the aforementioned king of Tlemcen offered to hand over "all the coastal places (…) and all the coast." In exchange, he would remain as Ferdinand's vassal ruling only over the capital and a little city called "Remezet, which is on the coast, where I am, because in this city are buried my father and my grandfather." 92 This humiliating and dramatic offer reflects the weak position of the Muslim ruler. But it is also interesting considering that it would have allowed Ferdinand to effectively gain control an important part of the Maghreb's coast. Nevertheless, the Christian monarch ignored the embassy that came to negotiate with him on behalf of the Muslim king. 93 Ferdinand's reticence towards agreements including the transfer of sovereignty over extended Muslim communities was even more explicit after the conquest of Béjaïa (1510) in the Algerian Kabylia. 94 In a letter to his general Pedro Navarro, the monarch shared his plans about the future government of this area, as well as the guidelines for negotiating with the local king. The only place under Ferdinand's direct rule would be the port city of Béjaïa. The rest of the territory, mainly the hinterland, would remain under the rule of the dethroned "king of Béjaïa" Mulley Abdallah, a local ruler who recovered the throne with the support of Ferdinand the Catholic. 95 The final remark of the king's letter to his general captain highlights Ferdinand's strategy to expand his influence over this area: "[Béjaïa] will be more safe and without war, and the other cities of the coast will be better conserved if he [the King of Béjaïa] has all that kingdom except the coasts (…). Furthermore, the said king will be better conserved with the Moors having his own state, and without being the things mixed." 96 In writing this Ferdinand was rejecting the possibility of exercising direct rule over an extended Muslim community in the Maghreb. The Christian monarch had good reasons to act in this way. He had experienced the problems of ruling an Islamic population in Granada, where the Muslim revolt of 1500/1501 had shown the fragility of his dominion in a land where Christians were a minority. 97 By rejecting the possibility of ruling over extensive Muslim territories, King Ferdinand was also showing his awareness of the difficulty that a Muslim ruler would face in trying to maintain his authority as a mere intermediary of a Christian king. A treaty was signed in order to pacify this area after the conquest of Béjaïa, putting into practice Ferdinand's ideas about not imposing Christian rule over a territory inhabited by Muslims. As a result, only Béjaïa and other coastal populations came under the sovereignty of the Hispanic Monarchy. 98 As we can see, from Ferdinand's view point, these negotiations were possible, but he knew that such a practice had its limits.
We can conclude here that some rulers such as the tribal leaders paid little attention to the religious factor when looking for support in their struggle against their local enemies. Hernando de Zafra was doing the same in his search for local supporters to carry out the expansion of the Hispanic Monarchy into the Maghreb. Undoubtedly these actors prioritised their political objectives over their religious divisions. The case of the king of Tlemcen is also meaningful, as he did not hesitate to negotiate his submission to a Christian ruler in order to save his threatened crown. Negotiating with a non-Muslim monarch and his agents was also admissible for the rulers in this region if it reinforced their position. Nevertheless, the religious difference was also a factor in the geopolitical plans of these rulers. Most of the Muslim rulers who offered to become subject to the Spanish crown did so only because they were compelled by circumstance. Furthermore, Ferdinand the Catholic rejected the idea of becoming the ruler of the entire Maghreb, due to the impossibility of ruling such a large Muslim territory. We should not forget the anti-Christian riots referred to above, or the problems of the Muslim notables who negotiated with the Christian authorities. Interestingly, Ferdinand the Catholic and the Muslim inhabitants who contested these practices agreed on one thing: religious beliefs were crucial factor to a ruler's legitimacy. Once again, the political culture from one side and the other matched.

Conclusion
Religious divides were not an insurmountable barrier in the geopolitics of the Western Mediterranean when negotiating political submission to a ruler of a different faith. Nevertheless, we should not dismiss the importance of the religious factor in the creation of geopolitical frontiers during the early modern period. For many Muslim leaders, submission to a Christian king was an option only under certain conditions such as the need to protect the local community in the face of an imminent military threat. On the Christian side, the case of Ferdinand the Catholic underlines the important and complex role that religious difference played in the construction of the frontier between the Iberian Peninsula and the Maghreb. From the very beginning, Ferdinand was willing to accept the submission of some coastal cities. Nevertheless, he rejected the possibility of ruling over an extended territory inhabited by Muslims, and preferred instead to expand his influence over the Maghreb by establishing allegiances with Muslim vassal kings. It is difficult to imagine a clearer way of showing the extent to which religious difference mattered in the geopolitics of this region.
Despite the obvious limitations and shortcomings of the cross-confessional negotiations analysed above, the actors that engaged in them took part in the construction of the Mediterranean frontier between the Iberian Peninsula and the Maghreb. Their principal contribution was to define the practices that ruled the complex spiritual geopolitics of the region. The local elites of the Maghreb cities who offered their subjection to the Catholic Monarchs took part in shaping a shared political culture in which it was possible to negotiate political subjection to a non-Muslim ruler in exchange for protection. The Christian military officials who combined armed force and diplomacy in their struggle against the Muslims conducted a war on the frontier in which negotiating the surrender and delivery of the enemy was not only possible but advisable. Their Muslim counterparts who negotiated the delivery of the fortresses under their command in exchange for personal gain also contributed to the art of warfare and diplomacy on the frontier. Nevertheless, we should not forget that the Iberians' willingness to negotiate with the Maghrebi obeyed tactical reason. Something similar happened regarding the religious agents engaged in this cross-confessional diplomacy, as their participation in these negotiations showed that for them it was possible to participate in these negotiations if it helped them to accomplish their political objectives. Moreover, the leaders of Jewish communities also helped to construct a frontier in which this religious minority had a space, due to the services that they offered to the new authorities. In addition to this, Muslim rulers who negotiated their vassalage to Ferdinand showed that for them it was possible to rely on the support of a Christian monarch, though perhaps only in order to save their crowns and to reinforce their authority vis-à-vis local enemies. Finally, when Ferdinand rejected the possibility of ruling over the whole kingdoms of Tlemcen and Béjaïa, he asserted the principle that a Christian ruler like himself needed to rely on allied Muslim rulers in order to extend his influence over Muslim territories like the Maghreb. As we can see, each of these actors decided when and under which conditions it was permissible to engage in these negotiations.
Widening our focus, we can connect the practices and actors analysed here with some crucial patterns of the Iberian expansion beyond the Mediterranean. The clear distinction between dealing with Muslims and extending sovereignty over them can be traced in different scenarios in Africa, Asia, and the Americas. This is crucial because this clear distinction lies at the foundation of the expansionist strategy frequently adopted by Iberian authorities all around the world and which consisted of establishing alliances with local rulers or making pacts with different elites rather than extending their direct sovereignty over non-Christian communities. Furthermore, the plural character of this diplomacy from below as implemented by military officers, religious agents, local notables, and others also appears to be a clear precedent for the frontier diplomacy practiced in faraway spaces. 99 In brief, actors from both shores of the Mediterranean contributed to create a model that was later exported to other parts of the world.
The actors analysed here played an important role in shaping the Mediterranean frontier, and it is clear that such cross-confessional diplomacy continued to bring together Muslim and Christian characters acting according to similar principles. In fact, similar negotiations took place even after the rise of the Ottomans as the hegemonic power in the region. For instance, in 1520 the viceroy of Sicily obtained the vassalage of the sheikh of Djerba, on the Algerian coast. 100 The following year two Jews working in collaboration with a Muslim governor offered to deliver Vélez de la Gomera to the Spanish authorities, asking, in return, for almost identical conditions to those granted to Zatorra and the Jews in Oran. 101 After conquering Tunis in 1535, Charles V signed a treaty of vassalage with the deposed Muslim ruler of this kingdom. 102 Some years later, some obscure negotiations took place trying to attract Barbarossa, the admiral of the Ottoman navy in the Mediterranean, to the service of Emperor Charles V. 103 The Berber tribal leaders continued to switch alliances with the emperor's agents in Oran, as well as with the Ottomans up to the mid sixteenth century. 104 Even at the beginning of the seventeenth century, a local prince gave Philip III Larache, an important harbour town on the Atlantic coast of the Maghreb, in exchange for his support in an internal struggle for the Moroccan throne. 105 All these examples show that many actors continued to negotiate Mediterranean geopolitics according to the same principles that ruled the negotiations analysed here. Many of the actors who took part in this cross-confessional diplomacy from below did not succeed in achieving their objectives. Nevertheless, they contributed to the creation of a frontier where it was possible to engage in diplomatic negotiations in spite of religious difference. That is, this was a frontier where the religious and political divides did not overlap. Furthermore, they took part in the definition of a shared political culture comprising the conditions under which it was possible, and impossible, to negotiate with "infidels."