The godlen Horde and the Mamluks Текст научной статьи по специальности «История и археология»

УДК 930.23 DOI: 10.22378/2313-6197.2017-5-1.93-115

THE GODLEN HORDE AND THE MAMLUKS1

1 2

Marie Favereau '

1 Oxford University Oxford OX1 2RL, United Kingdom

2 La Sorbounne - Paris IV Paris 75005, France E-mail: marie.favereau@history.ox.ac.uk

Research objective: to discuss characteristics of the relations between the Mamluks and the Golden Horde.

Research materials: Mamluk (royal biographies, official histories, chancellery manuals, insa' literature), Italian, Ilkhanid and Timurid sources.

Results and novelty of the research: the Golden Horde and the Mamluk sultanate exchanged envoys during almost two centuries. The reasons why these two societies enjoyed relations over the long run were complex and changed over time. This study shows that the Mamluk-Jochid relationship cannot be considered as a single, long-lasting alliance, but as a succession of silent periods and active exchanges. The novelty of the research lies in the fact that it aims at clarifying the motivations of the successive alliances set between the sultans and the khans by reconstructing the chronology of these exchanges and carefully investigating key moments. The traditional views on the "Mamluk-Jochid diplomacy", described by historians as a loose entente, is here reconsidered and the internal situations of both the Golden Horde and the Mamluk sultanate are taken into account. This study gives an overview of the long relationship of the Golden Horde and the Mamluk sultanate and offers some answers to the following questions: Did the fact that Mamluks and Jochids shared the same enemies remain the essential cornerstone of their diplomatic relations? Can we speak of a "foreign policy" that went beyond the personal investment of a single ruler? As a result of our research we came to the conclusion that the long-term relationships of the Mamluk sultanate and the Golden Horde had a strong impact not only on both empires but also on the whole inter-regional balance.

Keywords: Mamluk sultanate, slave trade, diplomacy, Genoese, Black Sea, Ilkhanids.

For citation: Favereau M. The Golden Horde and the Mamluks. Golden Horde Review. 2017. Vol. 5, no. 1, pp. 93-115. DOI: 10.22378/2313-6197.2017-5-1.93-115

From the mid-thirteenth to the mid-fifteenth century, the Jochids and the Mamluk sultans of Egypt and Syria regularly exchanged letters, presents and goods; and people used to travel from one territory to another. Slaves, civil administrators, warriors, traders, women, poets and scholars went from the Golden Horde to the sultanate [29, p. 304-319]; Muslim literati, jurists, and traders left the sultanate for the khan's court. Diplomatic missions were an important channel for the transmission of luxury goods - in particular through the circulation of "gifts" ex-

1 This study is the English version of my article in Russian: Золотая Орда и мамлюки // Золотая Орда в мировой истории. Коллективная монография / Хакимов Р. С., Фаверо М., Трепавлов В.В., Миргалеев И.М., Хаутала Р. (ред.). Казань: Институт истории им. Ш.Марджани АН РТ, 2016. C. 334-353 [31].

pected and even demanded by rulers and elite members - but also of technical skills, knowledge and court fashion [21, p. 28, 32, 61-65]. Jochids and Mamluks established agreements most likely connected to written statements. The diplomatic letters themselves were sometimes considered as proof of agreement. These agreements were valid for a certain period, usually a ruler's lifetime. A new sovereign would renegotiate the terms of the alliance according to his own background and to the new political situation. Diplomatic clashes between Mamluks and Jochids and periods of hiatus, when the network was disrupted, occurred a few times. As we will see, the Mamluk-Jochid relationship cannot be considered as a single, long-lasting alliance, but as a succession of silent periods and active exchanges. Obviously, their complex ties went beyond simple cordial relations, and tensions between the two courts were palpable.

The reasons why these two societies enjoyed relations and exchanges over the long run were complex and changed over time. The historical reconstruction of the entire period raises many questions, and the details of the Mamluk-Jochid diplomacy are either yet to be uncovered or remained disputed2. Did the fact that Mamluks and Jochids shared common interests - the same enemies, for instance - remain the essential cornerstone of their diplomatic relations? Were they also seeking something else? What was the degree of mutual formal and organized contact? What was the level of formalization of their exchanges? In other words, can we speak of a "foreign policy" that went beyond the personal investment of a single ruler?

This study gives an overview of the long relationship of the Golden Horde and the Mamluk sultanate and offers some answers to these questions. The diplomatic interplay began precisely in 660 H. (between November 1261 and October 1262), year of the sending of the first letter from Cairo, and ended in the mid.-fifteenth century. By then, the Ottomans controlled the Black Sea, Constantinople and the strait of the Bosphorus. The khans lost their hands on the trade roads, and the means of communication between the Golden Horde and the Mamluks, via the Crimean peninsula, were cut. The Giray khans based in the Crimea, torn between the Grand-duchy of Lithuania and the Ottomans, were not able to pursue a diplomatic and economic alliance on their own with the Mamluks.

For a period of almost two centuries (c.1260-c.1440), leaving apart the dubious or ambiguous mentions, we can identify some eighty missions. Times of intense diplomatic activity witnessed continuous reciprocal traffic: the Mamluk envoys sent to the khan came back to the sultan accompanied by Jochid messengers who carried the answer of their ruler and vice versa. By reconstructing the chronology of these exchanges and carefully investigating key moments, we should be able to clarify the motivations of the successive alliances set up by the Jochids and the Mamluks. This study reveals how crucial these alliances were for both parties by looking at the three most important periods of exchange:

I. The beginnings (1261-1267): the first alliance launched by khan Berke and sultan Baybars.

II. The Pax Mongolica (c.1300 - c.1350): a period of time that witnessed the prohibition of the slave trade under khan Toqta and, later, when Uzbek and al-

2 Among the numerous studies that deal with the Mamluk-Jochid alliance, the most important contributions are Zakirov, Amitai, Broadbridge and Ciocîltan [1; 11; 24; 26].

Malik al-Nasir Muhammad b. Qalawun were in power, the collapse of the Ilkhanate, arch-enemy of both the Golden Horde and the sultanate.

III. The last alliance (1382-1399): the last attempt, by khan Toqtamish and sultan al-Zahir Barquq, to secure a diplomatic and economic relationship between their empires.

The sources

Most historical studies on the Golden Horde and the Mamluk sultanate tend to focus on the Mamluks' position. The khans' politics appear almost as secondary. This imbalance is generated by the sources themselves: even if the original letters were not preserved, some of them were recorded by the secretaries of the Mamluk chancellery and included in chronicles, chancellery manuals and works of insa' literature. The academic literature is based on these Arabic versions of the summarized letters. The fact that we do not have the original Jochid versions raises questions concerning the content of their copies as registered by the Mamluk secretaries. The Jochid missives could be written in Turkish, Persian, Arabic or Mongolian [30, p. 76-80]; they were then summarized in Arabic and included in official works. How can we know that these copies truly reflected the Jochids' messages? How should we deal with the strong bias of the Cairo-centric view [24, p. 4]? Can we still rely on these sources? I am inclined to think so.

Mamluk sources of the Golden Horde are precious. They are almost the only "pro-Jochid sources" we know. Because the first task of a diplomatic mission was to gather and register such information, secretaries and Mamluk historians really meant to record concrete and reliable details about the khans' environment: ruling figures, dynastic history, political achievements and plans, languages spoken at court, geographical descriptions, etc. The Mamluk sources provide, thus, concrete information about political and military issues. The careful reading of these texts, in which the names of the envoys, the list of gifts exchanged and the transcription of the oral report transmitted by the ambassadors are often mentioned, allows us to penetrate the formal conventions and to understand the practical side of the Mamluk-Jochid contacts.

This material is especially useful in reconstructing the chronology of the exchanges between Baybars and Berke: who opened the diplomatic friendship? What reasons were officially offered? When did it occur? To counterbalance the fact that there are no sources left from the Jochid side, we have to take the "events" into account: exchanges occurred and missions were sent. The simple fact that the Jochids sent ambassadors to the Mamluks means they cared.

For the second half of the thirteenth century, the major source is Ibn Abd al-Zahir (620/1223-692/1292), who enjoyed a privileged position as Baybars's personal secretary and biographer. The basics - in terms of facts and chronology - are given in his official biography of the sultan, al-Rawd al-zahir fi sirat al-malik al-Zahir, which may be complemented by the narratives of Ibn al-Dawadan, SafT b.Ali, Baybars al-Dawadar, al-Yunini, al-Nuwayn, al-Dahabi, Mufaddal Ibn Abi-l-Fada'il, Ibn Katir, Ibn al-Furat, al-Maqrizi and al-Ayni. Date disputes and confusion are common in the Mamluk sources, so it is necessary to compare all versions available to us.

For the period of the Pax Mongolica, on top of the major Mamluk historians, we can rely on important Persian sources; bearing in mind that they were patro-

nized by the Ilkhanids and often biased when it comes to the Jochids. To complete the Islamic sources, Italian and Latin sources are very useful. Venetians and Genoese played a crucial role in the Mamluk-Jochid exchanges and were well informed eye-witnesses [8; 26].

For Toqtamish and sultan al-Zahir Barquq times, we benefit a lot from crossing Mamluk and Genoese information. The Timurid sources are also important but should be taken with a grain of salt, as their authors often held very negative views of the Jochids [25; 3]. Regarding the last years of the Mamluk-Jochid relation, which ended in the first half of the fifteenth century, we mainly rely on Arabic sources3.

I. The beginnings

Between November 1261 and October 1262 (660 H), sultan Baybars entrusted his first letter for khan Berke to Alan merchants. The sultan congratulated the khan on his recent conversion to Islam and exhorted him to fight Hulegu in the name of jihact. The war against a common enemy and the defense of Islam were the two reasons offered to justify the establishment of an alliance [37, p. 88-89]. Yet, if we look at the crucial events of 1261-1262, what predominates is not the conversion of Berke but the military conflicts between the khan and Hulegu5.

A common enemy

Indeed, Baybars's decision had nothing to do with the recent conversion of Berke - who had been a well-known Muslim for more than ten years - but with the fact that the khan launched a first attack against Hulegu's positions during the winter 1261-1262. In a few weeks Berke's army reached the south of Derbent and took the city of Shirvan within Hulegu's domain - their common border was situated at the Terek River, north of Shirvan. In August 1262, Hulegu left the city of Ala-tag, northeast of Lake Van, to travel to Shirvan, and stroke back.

On 13 October 1262, a group of two hundred Jochid warriors arrived at Damascus. This group - called in the Arabic sources wafidun or wafidiyya, "the refugees" - was fleeing Hulegu's troops. They showed a written order of the khan and asked the protection of the Mamluk sultan. Baybars welcomed them in Cairo on 10 November 1262 [37, p. 137-139; 9, p. 220; 39, VIII, s. 90-91; 56, p. 442-443; 67, I, p. 487, 496-497]. The first Jochid-Mamluk alliance was set.

Exactly at the same time, Hulegu expelled the Jochids from Shirvan (14 November 1262). On 8 December 1262, he arrived before Derbent, which was occupied by the khan's vanguard. He took the city back, crossed the Terek and chased the Jochid troops, plundering their encampments. Berke retaliated. On 13 January 1263, his army pushed Hulegu's troops beyond Shabran. The khan stopped at that point and went back to the Volga valley. For almost a year, there was no fighting [61, p. 392-399]. The conflict was still in its first stage, and both sides were making preparations to resume battle. It was precisely during this "truce" that Berke sent his first diplomatic mission to Baybars. His messengers arrived in Cairo on 21

3 Next to the range of Arabic sources edited by Tizengauzen, we can find mentions of the Mamluk-Jochid exchanges in a few additional sources such as the Qahwat al-insa' [44, letters 30-31].

4 Hulegu (d.1265) was the brother of the great-khans Mongke (1251-1259) and Qubilai (1260-1294). In 1251, he was put in charge of the new Chinggisid conquests to the west.

5 The first part of this study draws mainly from [31].

May 1263 along with Seljuq, Byzantine and Genoese envoys [56, p. 452-453; 67, I, p. 533-534, II, p. 194-195; 45, XIII, p. 238; 27, p. 97, 167; 58, p. 87-88]6. Why did these fives parties convene for?

Internal Legitimacy

If the Jochids decided to side with the Mamluks against other Mongols, it implied that the alliance involved more than military logistics. At that time, the Mamluk sultanate and the Golden Horde were young political formations. The rulers themselves were newcomers on the throne: Berke arrived in 1258 and Baybars in 1260. In this context of war against Hulegu, the khan and the sultan had to deal with internal legitimacy issues and they needed each other's support. Both rulers' situations have to be taken into consideration to understand what was at stake.

Baybars's situation

The Battle of Ayn Jalut (north of Jerusalem) took place on 3rd of September 1260. The Mamluks led by the sultan Qutuz opposed the Mongols led by Hulegu's commander Qitbugha. The Chinggisids were defeated, marking a crucial step in the regional balance of power that was until that point favourable to the Mongols. Their advance in the west was restrained for the first time since the fall of the major Islamic clusters in Central Asia, in Iran and in the Middle East7. It was a badly-needed Islamic victory, put under the banner of the Mamluks.

Most of these Mamluks came from Turkic nomadic families living in the western part of the Eurasian steppes, the region between the Volga and the Black Sea called Dasht-e Qipchaq. Military slavery had distinctive features that allowed the successful candidates to have access to high-status functions. A Mamluk at the peak of his career was an elite-member freed from slavery more than a slave-warrior. The Mamluks who took power in Egypt and Syria were mainly part of the household of the last great Ayyubid sultan al-Salih Ayyub (r.1240-1249). Baybars had been recently recruited as a Bahriyya member, and he was only one of the Mamluk emirs fighting at Ayn Jalut. After the battle, on his way to Cairo, the sultan Qutuz was murdered. Baybars, who claimed to be the head of the plot, took the place of the hero of Ayn Jalut by ascending the throne [34, p. 127-135; 32, p. 3148]. He had to show that he was not only the victorious combatant of the crusaders, as he proved himself to be at the battle of al-Mansura, but also the challenger of the Chinggisids.

In his letters to the khan, the new sultan called for jihad against Hulegu and the non-Muslim Mongols [37, p. 88-89, 139-140]. For that purpose, however, he needed more Mamluks. The regular routes used at the time of their predecessors, the last Ayyubids, crossed the sultanate of Rum. Key markets such as Siwas were located there, and Turkmen and Seljuq merchants made deals with the Qipchaq traders living on the other side of the Black Sea, in the Crimea [38, p. 113 (602/1205-1206); 54, p. 404-405; 36, II, p. 93, 112-122]. Yet when Baybars became sultan, in 1260, the Anatolian section of the trade circuit transporting military slaves from the Qipchaq steppes to Syria and Egypt was in the hands of Hulegu. Baybars had to find a way to bypass the Seljuq roads and the Turkmen intermediar-

6 Zakirov [1, p. 50] gave a wrong date based on Baybars al-Dawadar and MaqrTzT.

7 The Khwarizmshah, the Abbasids and the NizarT-Isma'TlT were already dismantled at that

time.

ies. He had to shape a new itinerary and to secure it by all military and diplomatic means.

Berke's situation

When Batu died at the end of 1255 or beginning of 1256, great khan Mongke wanted that Sartaq, the eldest son of Batu, succeeded him in accordance with the rules of succession that tended to prevail among the Chinggisids [48, p. 266-268]. After Sartaq died, a few months after his election, Mongke asked that Sartaq's son, Ulagci, became khan. Ulagci died at the end of 1257 or beginning of 1258. Only then could Berke ascend the throne. He was the aga (eldest) of the Jochids. But his mother was only a secondary wife of Jochi - a daughter of the Khwarizmshah - so his pedigree was not as prestigious as Batu, Sartaq and Ulagci8. Mongke seemed, thus, reluctant to see Berke becoming khan of the Jochid ulus. The fact that he was Muslim might have played a role as well. Berke had links with members of the Khwârezmian royal family, seen as unfaithful by the Mongols, and connections with the rulers of the Dehli sultanate who refused to pledge allegiance to the Chinggisids9.

Once on the throne, Berke had nevertheless to obey the great khan's orders and to follow the imperial agenda in the Middle East [10, p. 61-63, 104]. Jochid troops participated, thus, in the Mongol conquests in Iraq and Syria-Palestine and helped to take Baghdad. As a result, they had rights on the conquered places and could claim a share of the tributes. Yet, Hulegu, who was leading the Chinggisid conquests to the west, started to threaten the Jochid positions in western and central Asia. He chased the Jochids from Herat, Tabriz, and Baghdad; and even made some Jochid princes killed [35, p. 260-276]10. Moreover, he discharged the Seljuq sultan Izz al-dîn, while he maintained his brother Qïlïc Arslân IV on the Seljuq throne [24, p. 31-32, 54, 55]. Izz al-dîn was Berke's protégé and his banishment meant a definitive loss of control for the Jochids in eastern Anatolia. This was the detonator of the hostilities launched by the khan. During the winter 1261-1262, Berke declared war on Hulegu [59, p. 327-332; 61, p. 358-361, 390-401, 404-415]11. At the great khan's court, the underlying tensions among the Chinggisids had already turned into open conflicts and two candidates were fighting for the supreme office: Qubilai, supported by Hulegu, and Arigh Boke, supported by Berke.

What Berke wanted in exchange for Qipchaq slaves

Why would Berke be ready to let Baybars buy young men whom he could potentially need in his own armies? The expected compensation must have been high. In the first letter he sent to Baybars in 1263, the khan claimed that Hulegu did not respect the Chinggisid laws: "[He] transgressed Gengis Khan [yasa] and the San a of his people; he aimed only at killing men with hate". For Berke, protection and propagation of Islam were also at stake: " [...] I, together with my four brothers,

8 On Berke's origins, see [63; 30, p. 63]; on the precedence of the mother, see [48, p. 40].

9 On Berke's attempt to set up an alliance with the Neguderis (or Qara'unas), enemies of Hulegu, see [47, p. 239-244].

10 I would like to express my thanks to Dr Yihao Qiu who raised my attention to this information.

11 This is the most probable date despite a lot of debate on that point in the historiography. See [47, p. 233-234; 11, p. 79; 30, p. 64-67].

stood up and fought him from all sides for the sake of reviving the light of Islam and returning the abodes of the True Religion to their old state of prosperity and to the mention of the Name of God, the call to prayer, the reading of the Qur'an, prayer, and avenging the Imams and the Muslim community" [7, p. 101-113; 14, p. 167-172; 11, p. 82-83; 24, p. 53-54].

Berke offered Baybars to send troops towards the Euphrates to stop Hulegu's military operations and trap him between their two armies. They would share the conquered lands12. At that time, the Abbasid and Seljuq legacies, the greatest of the dar al-islam, were at stake. Negotiations concerning the allotment of the Seljuq territory were in process. The khan asked the sultan to support Izz al-dTn, who was in exile at Constantinople. According to Ibn Abd al-Zahir, Izz al-dTn also promised Baybars half of his kingdom in case of victory13. However, claims on the Seljuq domain, as on all the territories conquered by Baiju when he was in the service of Batu in 1242-1243 [47, p. 216-219; 10, p. 49, 177]14, were openly made by the Jochids. The common idea was that the region of Rum was a Jochid conquest [15, p. 476]. Yet, direct tensions between Jochids and Mamluks over the Seljuq realm never arose because Hulegu and, later, the Ilkhanids controlled the lands.

Since Baghdad was in the hands of Hulegu, Cairo became the new seat of the caliphate and the caliph was under the control of Baybars. Ibn Abd al-Zahir suggested that the khan was asking Baybars for the caliph's recognition of his status as sultan. Indeed, Berke was keen to consolidate his position as a Muslim ruler. Having territorial claims did not imply that the khan had the means to fight his enemy and, compared to Hulegu, Berke was militarily and politically weak. Hulegu was the loyal deputy of the great khans Mongke and Qubilai and could legitimately fight back; besides he had at his disposal more troops than did the khan15. Without external support, the Jochids could not hope to win against Hulegu. Berke's alliance with Baybars aimed at strengthening the position of the Jochids in the newly conquered regions and in the wider Islamic world.

The khan was setting up an alliance with the Qipchaq sultans (the main rebels to the Chinggisid cause) in Cairo and Delhi. His court was welcoming the nebula of administrators and officers who fled the court of the Khwarizmshah (Berke preferred to rely on faithful TurkT than on Uyghur secretaries sent by Qubilai). As the Golden Horde was becoming the main Islamic power of the area, it was crucial for the Jochids to have access to Mecca and Medina. Their friendship with the Mamluks would allow them to send their people to the Hedjaz to make hajj and to

12 On the first mission sent by Berke, see [37, p. 171; 58, p. 87]. For a divergent version, possibly based on Ibn Saddad, see [56, p. 452-453; 67, I, p. 534, II, p. 195; 39, VIII, p. 97]. As for Baybars al-Dawadar [20, p. 83], he mixed the texts of the two first letters sent by Berke.

It is highly improbable that the Jochids were ready to accept so easily the Mamluks' claim on the Abbasid caliphate, and perhaps some arrangements to share this crucial heritage were under negotiation as well. The Mamluk sources are obviously silent on a possible claim of Berke being associated with the caliphate. Yet a sign of such an association might be that the name of Berke was said during the Friday prayer in Medina, Mecca and Jerusalem [37, p. 174]; and previously in the regions of Ajam, Ma wara an-nahr and Khorasan [49, II, p. 1285, 1292-1293].

14 Baiju, appointed chief commander of the Chinggisid troops in western Asia in 1242, was then under Batu's authority. He passed later under the orders of Mongke.

15 According to Marco Polo [59, p. 329-330], Hulegu had 300,000 men against 350,000 for the Jochid khan. Allsen rightly questioned the plausibility of these totals. It is hard to believe that Berke had more men than Hulegu, who was in charge of the conquests [10, p. 203-207].

be recognized as part of the Muslim elite and significant members of the dar al-

islam16.

The Agreement

In 1262-1263, the khan Berke made a formal agreement with sultan Baybars and the byzantine emperor Michael VIII, who had recently retaken Constantinople from the Latins. At that time, the Mamluks were fighting the Mongols led by Hulegu in Syria and east-Anatolia. They had the same enemy as the Golden Horde had; and the same problem: they need to bypass the Anatolian markets where slaves were sent from Crimea. Thus, they formalized an agreement with Michael VIII stipulating that merchants and ambassadors the sultan or the khan sent were allowed to cross Byzantium and to travel through the straits of the Bosphorus. The configuration of a direct route between Alexandria and the Lower Volga via Constantinople and the Crimea - a travel of two months, two months and a half - was the main result of the negotiations that took place in Cairo between Baybars, Berke, Michael VIII, sultan Izz ad-Din and the Genoese17.

The setting up of a viable alternative to the Anatolian land route and the control of the slave-warriors market were a great accomplishment for all parties involved. Fifty days at least were needed to travel from Cairo to Saqsin in 1262 [37, p. 215; 39, VIII, p. 99; 67, I, p. 540; 58, p. 105; 56, p. 456-458]. The duration of the trip remained basically the same until the 1360s as did the itinerary from Cairo to the khan's orda on the lower Volga. According to Ibn Battuta's account (c.1334) it took indeed twenty days to travel between Sudak and Saray, and according to Pegolotti (c.1340), the distance between Tana, the gateway to the steppes, and Saray could be covered by ox-drawn carriage in twenty-six days, including one day of navigation on the Volga, or ten to twelve days for the same trip on horseback. If we follow this itinerary, it is obvious that the hub of the exchanges was the Black Sea. Whoever controlled the strait of the Bosphorus and whoever controlled the south of Crimea commanded the two pivotal access points to the "Mamluk-Jochid road". The fight for the control of these access points and the attempts to open competing routes would be one of the major issues at stake in this region until the end of the fourteenth century.

The short-term results of the first Mamluk-Jochid alliance were, therefore, substantial. True, the alliance was only partially successful. The plan outlined in the letters to recapture the Seljuq domain to the benefit of Baybars and Berke failed as did the resolution to kill Hulegu, who died a natural death. Nevertheless, Baybars and Berke succeeded in combining their military (human) resources through the slaves' recruitment. The alliance put a stop to Hulegu's conquests into the western Islamic lands. It was a major factor of stability during Baybars's reign and allowed him to recruit Mamluks for his armies, preventing the emerging sultanate from disintegrating. The consequences were equally positive for the Golden

16 For instance, under khan Toda-Mongke (r.1280-1287), two faqih came to the Mamluk sultan with a letter in Mongol script asking his permission to make hajj. The khan wanted also two battle flags or standards, one with the sultan's name and the other with the caliph's name [42, p. 354].

17 According to the treaty of Nympheum signed on 13th March 1261, the Genoese were to provide a fleet to the Byzantines and enjoyed commercial privileges in the Black Sea. For more details on the content of the agreement, see [28, p. 36-38 (no. 1890); 17, p. 42-45].

Horde. The cultural bonds between the two courts contributed to a first synthesis of Islamic and Chinggisid traditions. By allying with Baybars, Berke took independent and personal political decisions, which contributed to building up the foundations of the Golden Horde. Jochids and Mamluks were both looking for a legal basis to justify their expansionist agenda. Their agreement was framed by their convergence of interests. It was a good deal between two powers in equal need of each other.

II. The Pax Mongolica

After the Mongol conquests, a commercial boom transformed the Eurasian continent, connecting the Mediterranean Sea to India and China. The economic exchanges intensified integrating Asia, the Middle East and Europe - a global phenomenon, historians call Pax Mongolica (the Mongol Peace) because this was a world-shaping phenomenon on pair with Pax Romana. The post-conquest stability of the Mongol dominions and relatively peaceful relationships between the descendants of Chinggis Khan laid the ground for such commercial boom. The ways Mongols stimulated the exchanges resulted in a new form of long-distance trade and the agreements established between the Golden Horde, the Mamluks, the Byzantines, the Italians, and others, led to the transformation of the trade networks. A new economic order emerged, which cannot be seen as the mere revival of the continental silk road of the ancient world.

The Golden Horde was not a transit area that was merely crossed, but the melting point to which the great commercial circuits converged. If the route was open from Venice to China, and the goods exchanged and traded all the way long, only few merchants, like Marco Polo, did the whole trip. They usually stopped halfway: in the Golden Horde's lands. At the level of the lower Volga basin, the core area of the Golden Horde, two major routes were passable: the eastern one through Central Asia towards north India and China, and the western one through the Black Sea steppes and the Crimean peninsula towards the Mamluk sultanate and the Mediterranean world.

In this new economic context, the Mamluk-Jochid relation took a different shape. At the dawn of the fourteenth century, their exchanges were exposed to inter-regional competition and tensions were palpable through their diplomatic correspondence. Clashes between the two courts occurred a few times. They never led to direct military conflicts because the Golden Horde and the sultanate had no common borders but, as we will see, the consequences were significant for their go-betweens and intermediaries.

Toqta and al-Malik al-Nasir Muhammad b. Qalawun: the slave trade prohibited (c.1300-1312)

The early fourteenth century was a period of alliance at the level of the entire Mongol empire. At the end of the year 1304, a peace agreement was concluded between the Yuan, the Chagatayids, the Jochids and the Ilkhanids; the formal unity of the Chinggisid Empire was apparently restored. The Yuan kept the overarching title of 'Great Khan' and enjoyed a nominal primacy over the three other parts.

Yet, the same year, khan Toqta sent his first embassy to sultan al-Nasir Muhammad. In his letter, the khan said that he asked the Ilkhan to surrender the lands from Khorasan to Tabriz, and had threatened war otherwise. He sent also to

200 slave girls and 400 mamluks (most of whom died during the journey). In exchange, the khan asked al-Nasir Muhammad to support him against their Ilkhanid rivals18.

At that time, khan Toqta was at the peak of his power. He had eliminated Nogay (d.1299), the powerful beglerbeg who was ruling over Crimea and southeastern Europe. He had pacified the local tensions in the Dnieper and Dniester areas and in the Crimean peninsula. He wanted to control Transcaucasia. In May 1301, he sent his envoys to Ilkhan Gazan [62, p. 79]. We do not know the exact mission of this embassy nor the content of the letters brought by the Jochid messengers but, a few months later, some Jochid troops attacked the Ilkhanid empire along the Terek River. The military operations failed to change the balance of power in favour of the Golden Horde. They ended on a status-quo agreement between the khan and the Ilkhan and the Transcaucasian road was re-opened [26, p. 168-169].

The only way to really break the power of the Ilkhan was to open several military fronts at the same time and to attack the Ilkhanate not only from north Caucasus but also from Syria. The Mamluks were, therefore, crucial allies for the Jochids. But when Toqta sent his envoys to al-Nasir Muhammad, the sultan refused to comply arguing that "Allah had called Ghazan to him and that his brother Kharbende (Oljaytu) was already asking for peace" [26, p. 169-170]19. In fact, an Ilkhanid embassy had already informed the sultan that Oljaytu allowed Mamluk merchants into Persia. For the sultan, the peace with the Ilkhan had huge economic advantages and allowed him to secure the transit of slave-warriors via Tabriz. In 1306/07, Toqta sent again the same letter to al-Nasir Muhammad who again refused to support any military operation against the Ilkhan. A few months after, the khan ordered to expel the Genoese from the Golden Horde and to confiscate their goods. Caffa was besieged for eight months and the Genoese had to abandon their fortified settlement in May 1308.

Since the formation of the Golden Horde, the Jochids had stopped the merchant activities only once20. The khans' policies combined state control (treaties, currency issue, taxes, roads supervision) and liberal exchange (fluidity in partnership, alliances based on common interest and not on ethnic or religious affiliation, low taxation regime). They did not interrupt the exchanges except in case of warlike situation. Only very serious reasons could have led to Toqta's decision. Historians often presented the khan's actions as brutal and unexpected and they struggled to provide convincing interpretations of what happened21. We should consider the explanations given in the sources at the time when the first Mamluk-Jochid crisis took place.

18 In 704 /1304-1305: [57, p. 144; 52, p. 424; 20, p. 381; 39, IX, p. 128; 16, IV, p. 345]; in 706/1306-1307: [19, p. 180; 20, p. 388; 16, IV, p. 421-422; 52, p. 424]; in 710/1310-1311: [19, p. 220]; in 711/1311-1312: [57, p. 144]; in 714/1314-1315: [52, p. 424].

19 We have to bear in mind that at this time, al-Nasir Muhammad was only a puppet. The emirs were ruling in his name (Salar, Baybars al-Jashnakir, and others). I am grateful to Professor Amitai for this significant information.

20 At the end of the thirteenth century, when Nogay fought with the Genoese, it had nothing to do with the Mamluks, tax issues and political struggles within the Horde were at stake [26, p. 161-162; 55, p. 184].

21 Ciociltan offers an overview of the historiography and its limits: [26, p. 163-173, and esp. p. 166, note 93].

According to both Mamluk and Christian sources, Toqta took his revenge because the Genoese "stole" Tatar and Turk Qipchak children to sell them on the slave markets. The Dominicans argued further that the Genoese were acting too independently and were not respectful of the imperial laws [26, p. 165-167]. Indeed, kidnapping was illegal and merchants had to pay taxes on selling and purchasing. Yet, these taxes were very low and we know cases of pure robbery involving Genoese, or others, that were not punished by the sword. At last, if Toqta was against the slave trade or against selling local children - why was this happening in 1307 and not before? We should remember that in 1304 he had sent 200 slave-girls and 400 mamluks to the sultan. Therefore, selling local children cannot be considered the reason for the tension between the Genoese and the Golden Horde.

Some historians argued that Toqta was a "shamanist" and not a Muslim -which would explain the misunderstandings and disagreements he had with the Mamluk sultan [24, p. 131]. We can counter-argue that the Mamluks had fine diplomatic relations with khan Mongke-Temür who was not Muslims. Besides, depending on the sources, Toqta was a sky-worshipper, a Christian (baptized by the Franciscans under the name of Iohannes and buried in a Franciscan monastery near Saray) and a Muslim, bearing the name of Ghiyat ad-din, as can be seen on series of coins that were minted under his name in Saray and Ukek. Thus, it is hard to believe that religion had anything to do with the crisis between the khan and the merchants.

Obviously, the Genoese paid for al-Nasir Muhammad's refusal to organize a joint attack against the Ilkhanids. The timing of the khan's decision is a revealing indication: right after the sultan's third refusal, Toqta expelled the Genoese (November 1307). The Genoese were not only key intermediaries between the sultanate and the Golden Horde, they were also responsible of the most strategic slave-market places, at Caffa especially. In the wake of his expulsion of the Genoese, the khan stopped all diplomatic exchanges with the Mamluks, despite the conciliation attempt of the sultan, who wrote to Toqta, in September-October 1308, that "having heard of military clashes in the Caucasus between Ilkhanids and Jochids, he had prepared a military unit to help the khan but since he heard that the fights had stopped, he dismissed the unit" [18, p. 96]. Toqta was not fooled by the sultan's story and he remained closed to further diplomatic discussions [26, p. 171-173].

Mamluks and Jochids had no common border, they could not fight on the battlefield but they could fight on the market arena. Striking the slave merchants was striking the sultanate. For the first time, the Jochids tried to stop the slave trade; but the trade could not be stopped because there were other slave traders than the Genoese (Venetians, Greeks, local merchants, Alans, Turks etc.) and alternative networks22. As the Byzantine historian Nicophorus Gregoras (c. 1292 - c. 1360) noticed: every year one or two ships would cross the Black Sea to bring slave-warriors to serve in the Mamluk army and nothing could be done to end this trade which had become so significant [13, p. 364-366; 33, p. 101-102].

Even the most powerful empires cannot control their surrounding world. In this specific case, the use of embargo had two effects. On the one hand, the Genoese and Venetians merchants living in the Golden Horde sought for a new legal and

22 As far as I know, there is no indication in the sources that the slave trade from the Qipchak steppes went down at that time.

institutional framework that would provide better protection to their business [8]. On another hand, the Mamluks looked for alternative solutions to secure the supply of military-slaves, which meant to cooperate further with the Ilkhanids, masters of the Tabriz road. But, even if there was a short detente, the war had not ended; the Ilkhanids were still the enemies of the Mamluks and the Jochids. In 1310/11, Toqta sent his envoys back to Cairo with the Mamluk messengers who had came to announce al-Nasir Muhammad's third enthronement to the khan [18, p. 98]. The Mamluks and the Jochids still needed to exchange and communicate.

Uzbek and al-Malik al-Nasir Muhammad b. Qalawun: the end of a common enemy (1312-1341)

Uzbek and al-Nasir Muhammad entered into relations soon after the khan was elected. Uzbek's first embassy arrived in Cairo in March-April 1314. His letter informed the sultan of his throne accession and conversion to Islam. al-Nasir Muhammad sent his congratulations of his accession to the throne and a long list of gifts. In December 1315, he made a special request to the khan's envoys: he asked for a Jochid bride. After three years of negotiations, Tulunbay-Khatun (the khan's niece) was ready to be married to the sultan. In exchange, the khan forced the Mamluk ambassadors to borrow 27 000 dinars from his merchants to pay the dowry and to finance the wedding festivities. Tulunbay and her huge retinue (up to 3000 people, depending on the sources) departed from the lower Volga in 1319 and arrived in Egypt in early May 1320 [24, p. 132-133].

Among the most important persons of the princess escort, we should mention the shaykh Nu man KhwarazmT, the Genoese merchant Segurano Salvaygo, nicknamed Sakran, and Qusun, a future mamluk emir23. Tulunbay-Khatun was sumptuously received by the sultan and, in May, a marriage contract was arranged for a dowry of 30 000 dinars. This was not an extravagant bride-price if we consider that a year after, the sultan sent a gift package that was worth 40 000 dinars to thank the Ilkhan Abu Said for his series of gifts equivalent to 25 000 dinars [21, p. 67].

For Uzbek khan, this wedding alliance was a reassessment of the Mamluk-Jochid alliance and it implied military support against the Ilkhanids. Thus, as soon as 1320-1321, the khan asked al-Nasir Muhammad to join him on the battlefield against their common enemy. Yet, the sultan refused to comply and even warned the Ilkhan of Uzbek's plan. At that time, peace negotiations between the Mamluks and the Ilkhanids were in full swing and, for al-Nasir Muhammad, it was worth a diplomatic clash with the khan [24, p. 134].

In 1322, when the Mamluk messenger carrying this negative answer went back from the Golden Horde, he complained of Uzbek's treatment. The khan received him only once. During the meeting, he did not inquire of al-Nasir Muhammad's health. Besides, the sultan's envoy was prevented from purchasing slaves and slave-girls [24, p. 134-135].

The khan had two major reasons to complain as it was expressed in his letters to the sultan: first, al-Nasir Muhammad sent no army to Iraq while he, Uzbek, had faced the enemy for a month and, so far, has received no news of any advance from the Mamluk side. Second, the shaykh Numan, who had very high status in the Golden Horde, was ill-treated by the sultan and his entourage. The shaykh wished

23 On Segurano Salvaygo, see: [50]; on Qusun, see: [66].

to visit Jerusalem and Hebron (al-Khalil) and to build in Jerusalem a pious endowment. The sultan had not let him to do so. The khan was particularly displeased with this issue, as he had provided the shaykh with the necessary funds [24, p. 135]. It meant no respect for someone close to him and considered as a holy figure in the Golden Horde. Moreover, the khans had always allowed the sultans, and people from their entourage, to patronize the construction of religious buildings in the Volga Valley and in the Crimea. For instance, the gifts sent by Qalawun to the Horde in 1287 contained goods worth 2000 dinars for the mosque being built in Solkhat and masons were sent with their material and paint to carve the name and titles of Sultan Qalawun on this mosque [42, p. 356]. Between sultans of the same standing, reciprocity in the exchanges was expected. al-Nasir Muhammad acted as if he was despising the khan or as if they did not have the same status.

To retaliate, the khan killed Segurano Salvaygo, a rich Genoese merchant who was close to the sultan and used to purchase slaves in the Golden Horde. When al-Nasir Muhammad complained, the khan said that this merchant was killed by one of the kings of al-Jaza'ir (the islands); openly lying to the sultan's envoys [50, p. 86]. A few years after, al-Nasir Muhammad took his revenge by making a fool of the khan's messengers. In 1327-1328, al-Nasir Muhammad had divorced Tulunbay and married her to one his Mongol commanders [64, p. 482]. When the khan complained that she should be sent back to the Golden Horde, the sultan answered the messengers she had died and asked a judge to produce the legal proof for it [24, p. 136].

The divorce, the shaykh's bad-treatment and no military support were the major reasons explaining the diplomatic tensions between the Golden Horde and the sultanate. The inter-regional context has evolved since the time of Baybars and Berke. Al-Nasir Muhammad was in a strong position to negotiate with the khan: the Bosphorus route was not the sole access to the sultanate for the slave traders; the Anatolian and the Syrian land routes were then passable. In 1321, the Mamluk troops occupied Ayas (Laiazzo) in Cilician Armenia - one of the greatest commercial hubs of the area. The sultan had also a peace treaty with the Ilkhan who offered very advantageous conditions for trading in the Ilkhanate [12]. Yet, as the main source of Mamluks was still the core territory of the Golden Horde, Uzbek khan and al-Nasir Muhammad continued to exchange embassies and gifts, even during this 'cold war'.

Between the reigns of Toqta and Uzbek, there was a clear evolution in the status of the Genoese. Under Toqta, the clash with the Mamluks led to a massive expulsion of the Genoese merchants. The entire community living in the Golden Horde was held responsible for it. Under Uzbek, the Genoese, as an intermediary group between the Jochids and the Mamluks, could not be held responsible anymore for any diplomatic struggles with the Mamluks. In fact, when Uzbek arrived on the throne, the khan agreed on the reconstruction of Caffa, which started in 1316 on the basis of a new agreement:

"Anno domini MCCCXVI redificata fuit civitas Caffa per dominum Antoniun gallum et dominum Nicolaum de pagana sindicos comunis Janue per gratiam sibi concessam per Usbech imperatorem tartarorum" [60, p. 500-501].

The detailed content of this agreement was not preserved24, but for the same period we can rely on the contract concluded between Uzbek khan and the Venetians. Dated in 1333, it covered the crucial questions of the liability (collective and individual) and protected status of the trading community [8]. It is highly probable that the same rights were granted to the Genoese and their most important settlement in Caffa.

If Segurano Salvaygo paid for the diplomatic clash between the khan and the sultan, it was not because he was Genoese. Indeed, the Italian community was left in peace and the khan did not confiscate their goods. Neither was it for the divorce with Tulunbay, that Uzbek learnt later, nor because of the military defection of the Mamluks against the Ilkhanids. Segurano Salvaygo, a close friend of the sultan, paid with his life for the shaykh Numan's ill-treatment. The sultan has scorned the image of the khan as a powerful Islamic ruler, influential beyond the Golden Horde. The khan had to reply, for his internal and external legitimacy was at stake. Like the messages the diplomatic gifts suggested, actions towards rulers' representatives (ambassadors, close relations) required appropriate reactions. It was a matter of reputation for the rulers of the time.

III. Toqtamish khan and al-Zahir Barquq: the last alliance

The reigns of Toqtamish khan (c.1377-c.1397) and al-Zahir Barquq (1382-1390)25 offer coincidently interesting similarities. They both ruled during approximately two decades; Barquq died in 1399, which is also the date of Toqtamish's political death after his defeat on the Vorskla. Both reigns marked the end of an era and the beginning of new times. In Egypt, the connections with the Qipchak world became looser, as well as the importance of the nomadic culture in court and political life. al-Zahir Barquq started to integrate into the Mamluk elite an increasing number of Circassian people. It is worth noting that he was the last sultan to perform court rituals associated with kumys [51, p. 173]. He was also the last sultan to exchange extensively with the Jochids.

Notable studies have been focused on the relations between the Mamluks and the Timurids [24, p. 168-197] yet, they rarely highlight the role that the khan and his entourage, especially his deputies in the Crimea, played in the geo-political games of the area. In this last part, we will see what was concretely at stake at the end of the fourteenth century between the Mamluk sultanate and the Golden Horde and why this period was a turning point in the exchanges between both powers.

Timur, the new common enemy

The chronology of the missions exchanged between al-Zahir Barquq and Toqtamish is quite easy to reconstruct (especially if we compare with previous sets of diplomatic exchange between the khans and the sultans). We know about two missions: the first one in 1384-1386, the second one in 1394-1395. Both seemed to have been initiated by Toqtamish.

According to Ibn Duqmaq, in 786/Feb. 1384 - Feb. 1385, the khan's envoys arrived in Cairo. MaqnzT gave the more precise date of 11 Dhu al-hijja 786/ 24 January 1385. Ibn Hajar al-'AsqalanT added that the chief ambassador was Hasan,

24 In the case of the Genoese, if we except the chronicles, the oldest agreements with the Golden Horde that were preserved are from the 1380s, fortunately they included older contracts.

25 The sultan's rule was interrupted for almost a year due to a revolt in 1389-1390.

son of the former governor of Crimea, Ramadan. The embassy itself was sent by al-haqim bi-l-Qrim or Sahib al-Qrim on behalf of the khan. The envoys brought gifts: seven hawks and seven types of textile made of cotton from diverse origins, and slaves [40, p. 322; 52, p. 427-428; 43, p. 450].

A following embassy, this time sent directly by Toqtamish, arrived in Muharram 787/12 February - 14 March 1385. The content of the letter is not preserved in the Mamluk sources, we only know that Toqtamish was asking al-Zahir Barquq to reestablish relations as in olden days [24, p. 172]26. The reason for this diplomatic meeting was probably the war with Timur. Indeed, the first campaign launched by Toqtamish in Transcaucasia started soon after. In 1384-1386, Timur had made a military incursion as far as Azerbaidjan where he had left a governor over Sultaniyyah and Tabriz. In 1386, the khan's troops attacked Derbent and Tabriz. Timur replied by sending troops led by one of his sons. The fighting stopped on a status quo. At this point, the military strategy of Toqtamish was apparently to multiply the military fronts to weaken Timur's positions [2, p. 99]. The sultan would attack Timur's troops stationed between Derbent and Tabriz, and they would be caught between the Jochid and the Mamluk armies.

In 1388, Toqtamish launched his first campaign towards the East. His armies reached the banks of the Amu Daria River but failed to conquer the city of Bukhara. In 1389, Toqtamish's troops clashed with Timur's troops near the Syr Daria River. Again, both sides stopped on a status quo. This time, Timur decided to retaliate. In January 1391, after three months of preparation, he launched a campaign deep inside the heartlands of the Golden Horde. When the Timurid and Jochid armies met on the battlefield, Timur's troops were exhausted after six months of travelling in the steppes. Yet Toqtamish was defeated. Having lost the battle, he had to escape to the Middle Volga region [2, p. 124-125, 138], where he activated his alliance with the Grand Duchy of Poland-Lithuania by sending his envoys to his ally Vladislav II Jagello (1377-1434), who was the head of the Grand-Duchy together with his cousin Vitovt (1392-1430). Toqtamish was defeated but the raids conducted by the Timurid troops in the aftermath of this campaign were limited. And indeed only three years after, in spring 1394, Toqtamish was ready to resume the fight.

Before launching his new military expedition, the khan actively communicated with Poland-Lithuania and the Mamluks. On the 26 March 1394/ 23 Jumada I 796, his envoys arrived in Damascus to discuss the terms of a viable alliance against Timur [42, p. 356; 52, p. 428]. Soon after, in 797/1394-1395, the khan's troops entered the Caucasus. The region was then again under the control of one of Timur's son. In April 1395, Timur stroke back and the khan lost a crucial battle on the shores of the Terek River. According to Tuluman Ali Shah, the Mamluk ambassador who went back to Egypt in September 1395, the khan lost the battle because one of the most influents emirs of the Golden Horde betrayed him and joined Timur's camp [42, p. 356-357]. We know from other sources that this emir was Aktau [4, p. 180]27.

26 According to al-'Asqalani, in Safar 787 (March 14 - April 11, 1385), Toqtamish's envoys came again. They probably came for the same purpose: to convince the sultan to attack Timur [43, p. 450].

27 My deepest thanks to Ilnur Mirgaleev for providing me with this information and, more generally, for his relevant comments on the conflict between Toqtamish and Timur.

After the khan was defeated, he flew away to a wooded area in Bular [6, p. 178; 2, p. 138]. The Lower Volga region and the area between the Don and the Volga were plundered by Timur's troops. The Mamluk ambassador Tuluman AlT Shah left the region just before Timur's men arrived. With his companions, he went to the Crimea, first to Solkhat, then to Caffa, where they looked for a boat to cross the Black sea and go back to Egypt. But the Genoese kept him as a hostage and he had to pay 50 000 dirham to leave and safely reached Samsun. According to MaqnzT, the Genoese were on Timur's side [52, p. 428].

In fact, the Genoese - as the Byzantines - played all sides and switched from an alliance to another. We know, for instance, that at the same period they played the Ottomans against Timur and the other way around. Timur's military intervention has shaken the political organization of the peninsula and the Genoese were becoming increasingly autonomous.

Were the Genoese for or against the khan?

The first agreement between Toqtamish and the Genoese dated 1381. It recognizes the Genoese authority over eighteen places and villages in the southern Crimea. In exchange, the Genoese would support the khan, and not welcome his enemies in their towns. At that time, the khan needed to set up a strong alliance with the Genoese in order to have free hands to impose himself in the Rus' lands. If we compare the content of this agreement with the yarliks granted by Uzbek, Janibek and Berdibeg to the Venetians, we see that the khan made striking concessions to the Genoese: not only the eighteen settlements, but also the right to travel wherever they wanted in the Golden Horde [26, p. 225-229].

Right after they reached to this agreement, the Genoese started a large-scale renovation of the fortifications of the citadel in Caffa. They also increased their food supplies, acquired weapons and military equipment from Genoa and Pera. In 1385 started what we call the Solkhat war (which lasted until 1387) between local Crimean populations, the Jochids in Solkhat, and the Genoese [22].

Twice in less than ten years, Toqtamish launched military expeditions in the Caucasus. He targeted Derbent and beyond - Shirvan and Tabriz; yet even more important strategically, this was the only way to open a protected land route to Syria. The khan needed to keep the exchanges with the Mamluk world fluid for military and economic reasons. This direct land access to the sultanate would also put him in a stronger position to renegotiate his agreement with the Genoese who tended to dominate the Jochid connection with the Mediterranean world as the misfortune of the Mamluk ambassador Tuluman AlT Shah has shown. But the khan was never able to secure a Transcaucasian land road between the Golden Horde and the Syro-Palestinian coast.

The Crimea: the last bridge between Egypt and the Golden Horde

The succession of events that led to the Jochid embassies in Cairo in 13841386 and 1394-1395 is rather well documented. Yet the political agendas of Toqtamish and al-Zahir Barquq remained complex. We have to look under the stones to understand what was at stake. Timur's role and importance should not be exagarated. The exchanges between the Mamluk sultanate and the Golden Horde did not have for sole purpose the war against him. As we saw, Toqtamish did not only answer Timur's attacks, he had his own agenda.

The notion of an age-old friendship between the Mamluks and the Jochids, expressed in the Mamluk sources and in the recorded letters, is part of the language of diplomacy. The vocabulary used in the diplomatic missives, as their format, were very conservative partly because the secretaries used to take old letters as writing models [30, p. 89-91]. Yet, the geo-political situation was not the same as in the times of Berke and Baybars. Tradition was not the real motivation of the Jochids and the Mamluks, nor the drinving force of their exchanges; they did not send missions every year, but only when they needed it.

When Toqtamish sent his first embassy to the Mamluk sultan, it has been ten years that the diplomatic contact had stopped28, which does not mean that the commercial exchanges between the Golden Horde and the sultanate were cut. Even if the itineraries were the same, we have to deal with two different dynamics, trade and diplomacy, which sometimes converged but not always. As we already mentioned, the embassy of January 1385 was led by Hasan b. Ramadan, the son of the former beg of Solkhat. At the diplomatic reception, he offered falcons, cloth and military slaves, and the letter he was carrying for the sultan was read in front of the audience. The heads of the chancellery, then, discovered that the embassy was not coming directly from Toqtamish but from the governor of the Crimea. This generated embarrassment and confusion because the protocol had to be adapted to the status of the sender [23, p. 114-115]. This scene was recorded by al-Qalqashandi as an example of the kind of mistakes the secretaries in charge of the diplomatic protocol could make. Yet, the fact that al-haqim bi-l-Qrim sent a letter and envoys to the Mamluk sultan was not an innovation at all. We know that the diwan al-insha' received and kept letters from the highest Jochid officials and civil administrators. Ibn Fadl Allah al-'Umari and Ibn Nazir al-Jaysh recorded their correspondence with al-haqim bi-l-Qrim, with al-haqim bi-Azaq, with the umara' al-ulus, with the na'ib al-qan (or beglerbeg) and with the wazir (chief of the diwan) [41, p. 44-48; 46, p. 339-340].

The interesting point is not so much that the diplomatic reception was disrupted when the heads of the chancellery discovered the misunderstanding, but the fact that diplomatic exchanges between the khans' deputes in the Crimea and the Mamluks existed since, at least, the first half of the fourteenth century - not because the begs of Solkhat were acting independently of the khans but because the peninsula was a highly strategic place, a bridge between Egypt and the Golden Horde.

In the last quarter of the fourteenth century, the Mamluks were keen to maintain their commercial network and connections with the Black Sea area. The Crimea was the hub of the slave trade network; here stood the slave markets and the shipping facilities to the Bosphorus and the Mediterranean Sea. The Jochids needed the money from the sails and the Mamluk sultans the young slave warriors. The diplomatic gifts carried by the first embassy during Toqtamish' rule were a metaphore for exchange economy and fair trade: hawks and slaves were emblematic luxury commodities from the Golden Horde. The Jochid elite acquired textiles made of coton, silk, linen, from the Anatolian market and the Middle East. In the Crimean peninsula, cloths were exchanged for slaves since Seljuk times.

28 The last embassy had been sent in 776/1374-1375, when the Egyptian secretary Ibn Nazir al-Jaysh wrote a letter for the sultan to the khan 'Muhammad' (probably Urus); this embassy carried also gifts for the khan [1, p. 94].

For the sultanate, the importation of future Mamluks remained a crucial issue. al-Zahir Barquq personally acquired during his rule of sixteen years, some 5000 Mamluks. He once was a slave himself. When he was bought in 1363 by the emir Yalbugha al-Uman, he had approximately twenty-two years and was living in Qrim, as a slave, since several years [51, p. 64, 69]. The Crimea and especially Caffa remained the biggest slave market with Tabriz. And it was more advantageous to control the market places where the slaves were brought from everywhere, than their region of birth (the Dasht-e Qipchaq and the Caucasus where the Circassians came from).

In the first half of the fifteenth century, the Ottomans progressively controlled the Black Sea until they, finally, took control of the straits of the Bosphorus and Constantinople. Then, the connection between the Golden Horde and the Mamluks was cut for good.

Concluding remarks

The long-term relationships of the Mamluk sultanate and the Golden Horde had a strong impact not only on both empires but also on the whole inter-regional balance. The Horde and the sultanate were part of a bigger world - so, to understand what was at stake between them, we need to look at the broader context: the Mongol empire; the Genoese; the Venetians; the papacy and the sultanates of the dar a-islam.

Regarding the diplomatic conventions, we can say that basic topoi were elaborated during the founding period of Berke and Baybars. The expressions and images used in further letters show how important this first alliance was. The secretaries from both chancelleries based their rhetoric on the text of the letters recorded in the insa' literature, manuals and chronicles, and made frequent reference to the previous alliances. This first sequence set the terms for the birth of a diplomatic tradition and its "language". Yet we should not forget that Mamluks and Jochids had concrete and contextual reasons to engage in exchange and that only when the economic and political motivations failed did the alliance lose its raison d'etre.

Baybars and Berke, and their close entourages, were the instigators of a foreign policy that was continued and adapted by their successors. Jochids and Mamluks shared political and economic aims, they formalized and protected the exchanges on the basis of treaties and foregone itineraries, and they tried to coordinate their strategies to confront a changing political environment. They developed forms of communication between societies with different languages and traditions. They did believe that diplomacy could change things and used it as such.

There is no true alliance without tension. In this study, I tried to identify the reasons why the Mamluks and the Jochids, though officially allied, struggled at times. The Pax Mongolica was a period of tense economic competition. The evolution in the statuses of the merchants and of the ambassadors from Venice and Genoa is a significant consequence of such competition. Besides, notwithstanding their 'friendship', both the sultanate and the Golden Horde were empires in expansion with their own internal issues.

REFERENCES

1. Zakirov S. Diplomaticheskie otnosheniya Zolotoy Ordy s Egiptom — XIII—XIV vv. [The Golden Horde Diplomatic Relations with Egypt: 13th-14th centuries]. Moscow, Nauka Publ., 1966. 161 p. (In Russian)

2. Mirgaleev I.M. Politicheskaya istoriya Zolotoy Ordy perioda pravleniya Toktamysh-khana [Political History of the Golden Horde during the Reign of Khan Toqtamish]. Kazan, Alma-Lit Publ., 2003. 164 p. (In Russian)

3. Mirgaleev I.M. Materialy po istorii voyn Zolotoy Ordy s imperiey Timura [Materials on the History of the Golden Horde Wars with the Empire of Timur]. Kazan, Sh.Marjani Institute of History of Tatarstan Academy of Sciences Publ., 2007. 107 p. (In Russian)

4. Mirgaleev I.M. Bitvy Toktamysh-khana s Aksak Timurom [Battles of Khan Toqtamish with Aksak Timur]. Voennoe delo Zolotoy Ordy: problemy i perspektivy izucheniya. Materialy Kruglogo stola, provedennogo v ramkakh Mezhdunarodnogo Zolotoordynskogo Foruma. Kazan', 29—30 marta 2011 g. [The Golden Horde Warfare: Problems and Prospects of the Study. Materials of the Round table held in the framework of the International Golden Horde Forum. Kazan, March 29-30, 2011]. Otv. red. i sostavitel' I.M.Mirgaleev. Kazan, Sh. Marjani Institute of History of Tatarstan Academy of Sciences, 2011, pp. 170-182. (In Russian)

5. Tiesenhausen V.G. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. 1: Izvlecheniya iz" sochineniy arabskikh" [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from the Arabic Writings]. St. Petersburg, 1884. 564 p. (In Russian, Arabic)

6. Tiesenhausen V.G. Sbornik materialov,"otnosyashchikhsya k istorii Zolotoi Ordy. 2: Izvlecheniya iz persidskikh sochineniy [Collection of Materials Relating to the Golden Horde History. Volume 2: Extracts from the Persian Writings]. Moscow-Leningrad, Izd. Akademii Nauk SSSR, 1941. 308 p. (In Russian, Persian)

7. Favereau M. Pervoe pis'mo khana Berke sultanu Beibarsu po mamlyukskim istochnikam (661/1263 g.) [The First Letter of Khan Berke to Sultan Baybars according to Mamluk Sources (661/1263)]. Zolotoordynskaya tsivilizatsiya [Golden Horde Civilization]. Vypusk 4. Kazan, Sh.Marjani Institute of History of Tatarstan Academy of Sciences Publ., 2011, pp. 101-113. (In Russian)

8. Favereau M. Venetsianskie istochniki po istorii Zolotoy Ordy: novye perspektivy izucheniya [Venetian Sources on the Golden Horde History: New Perspectives of Studying]. Zolotoordynskoe obozrenie [Golden Horde Review]. No. 1. Kazan, Sh.Marjani Institute of History of Tatarstan Academy of Sciences Publ., 2016 (in press). (In Russian)

9. Abu Samah. Tarajim rijal al-qarnayn al-sadis wa-l-sabi' al-ma ruf bi-l-dayl ala al-Rawdatayn. Muhammad Zahid b. al-Hasan al-Kawtari (ed.). Cairo, 1947. 285 p. (In Arabic)

10. Allsen Th. Mongol Imperialism. The Policies of the Grand Qan Mongke in China, Russia, and the Islamic Lands, 1251—1259. Berkeley, University of California Press, 1987. xvii + 278 p.

11. Amitai-Preiss R. Mongols and Mamluks. The Mamluk-Ilkhanid War 1260—1281. Cambridge, Cambridge University Press, 1995. xv + 272 p.

12. Amitai R. Resolution of the Mamluk-Mongol War. Mongols, Turks and Others: Eurasian Nomads and the Sedentary World. Amitai R., Biran M. (eds.). Leiden and Boston, Brill, 2005, pp. 359-390.

13. Amitai R. Diplomacy and the Slave Trade in the Eastern Mediterranean: A Re-examination of the Mamluk-Byzantine-Genoese Triangle in the Late Thirteenth Century in Light of the Existing Early Correspondence. Oriente Moderno. Vol. LXXXVIII/1. Roma, Istituto per l'Oriente, 2008, pp. 349-368.

14. Ayalon D. The Great Yasa of Chingiz Khan. A Re-examination. Studia Islamica. Vol. 34. Paris, Maisonneuve et Larose Ed., 1971, pp. 151-180.

15. Al-AynI. Iqd al-juman fî tawarîh ahl al-zaman. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. 1 [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 475-477. (In Arabic)

16. Al-AynI. 'Iqd al-jumän f tawärih ahl al-zaman. Muhammad Muhammad Amin (ed.). Cairo, Dar al-kutub, 1987-1992. (In Arabic)

17. Balard M. La Romanie génoise (XIIe - début du XVe siècle). 2 Vols. Roma, École française de Rome, 1978. 1008 p. (In French)

18. Baybars al-Dawadar. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. T. 1 [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 76123. (In Arabic)

19. Baybars al-Dawadar. Kitäb al-Tuhfat al-mulukiya fi-l-dawlat al-turkiya (638-711). Abd al-HamId Salih Hamadan (ed.). Cairo, Dar al-Misriyya al-Lubnaniyya, 1987. (In Arabic)

20. Baybars al-Dawadar. Zubdat al-fikra f ta'rih al-hijra. Richards D.S. (ed.). Beyrouth-Berlin, Orient-Institut der DMG- Das Arabische Buch, 1998. 488 p. (In Arabic)

21. Behrens-Abouseif D. Practising Diplomacy in the Mamluk Sultanate: Gifts and Material Culture in the Medieval Islamic World. London, I.B. Tauris, 2014. 298 p.

22. Bocharov S., Sitdikov A. The Solkhat's War and Its Reflection in the Fortification of Caffa. Classica et Christiana. Vol. 9/2. Iasi, Ed. Univ. Alexandra Ioan Cuza, 2014, pp. 413-426.

23. Broadbridge A. Diplomatic Conventions in the Mamluk Sultanate. Annales Islamologiques. Vol. 41. Le Caire, Institut français d'archéologie orientale, 2007, pp. 97118.

24. Broadbridge A. Kingship and Ideology in the Islamic and Mongol Worlds. Cambridge, Cambridge University Publ., 2008. xiv + 232 p.

25. Charmoy F.B. Expédition de Timoûr-i-lènk contre Toqtamiche, Khân de l'Ouloûs de Djoûtchy en 793 H., 1391 de J.C: textes arabes, persans et turks. Publiés et traduits avec des notes et éclaircissements [par] François-Bernard Charmoy. Mémoires de l'Académie impériale des sciences de St-Pétersbourg. 6e série, Sciences politiques, histoire et philologie. T. 3, 5 (1836, 1835) [Reprint. Amsterdam, Philo Press, 1975]. (In French)

26. Ciocîltan V. The Mongols and the Black Sea Trade in the Thirteenth and Fourteenth Centuries. Leiden-Boston, Brill, 2012. 321 p.

27. Al-Dahabî. Kitäb duwal al-isläm. Vol. 1. Abdallah b. Ibrahim al-Ansarî (ed.). Qatar, Idara ihya' al-turat al-islamI, 1977. (In Arabic)

28. Dölger F. Regesten der Kaiserurkunden des Oströmischen Reiches von 565-1453. Vol. 3: (1204-1282). München, Oldenbourg, 1932. (In German)

29. Eckmann J. The Mamluk Qipchaq Literature. Central Asiatic Journal. Vol. 8. Wiesbaden, Harrasowitz, 1963, pp. 304-319.

30. Favereau M. Comment le sultan mamlouk s'adressait au khan de la Horde d'Or. Formulaire des lettres et règles d'usage d'après trois manuels de chancellerie (1262 -v.1430). Annales Islamologiques. Vol. 41. Le Caire, Institut français d'archéologie orientale, 2007, pp. 64-67. (In French)

31. Favereau M. The Golden Horde and the Mamluks: The Birth of a Diplomatic Setup (1261-67). Mamluk Cairo, a Crossroad for Embassies. Bauden F., Dekkiche M. (eds.). Leiden-Boston, Brill, 2017, pp. 282-301.

32. Flinterman W. Killing and Kinging: Altaic Notions of Kingship and the Legitimation of al-Zahir Baybars' Usurpation of the Mamluk Sultanate, 1249-1260. Leidschrift. Vol. 27/1. Leiden, Subfaculteit Geschiedenis, 2012, pp. 31-48.

33. Gregoras N. Nicephori Gregorae Byzantina historia graece et latine. Vol. I. L. Schopen (ed.). Bonn, E.Weber, 1829. (In Greek, Latin)

34. Haarmann U. Regicide and 'The Law of the Turks'. Intellectual Studies on Islam: Essays Written in Honor of Martin B. Dickson. Mazzaoui M., Moreen V. (eds.). Salt Lake City, University of Utah Press, 1990, pp. 127-135.

35. Haravï, Sayf Ibn Muhammad. Târïkh Nâmah-i Harât. Ghulam Riza Tabâtabâ'ï Majd (ed.). Tehran, 2004. (In Persian)

36. Heyd W. Histoire du commerce du Levant au moyen-âge. Leipzig, O. Harrassowitz, 1885-1886 [reprint. 2 vols. Amsterdam, Adolf M. Hakkert, 1967]. (In French)

37. Ibn Abd al-Zahir. al-Rawd al-zâhir f sïrat al-malik al-Zâhir. Abd al-Azïz al-Huwaytir (ed.). al-Riyad, Mu'assasat Fu'âd, 1976. 519 p. (In Arabic)

38. Ibn al-Atïr. The Chronicle of Ibn al-Athïr for the Crusading Period from al-Kâmil fïl-tarïkh. Part 3: The years 589-629/1193-1231: the Ayyubids after Saladin and the Mongol Menace. Transl. D.S. Richards. Aldershot-Burlington, Ashgate, 2008.

39. Ibn al-Dawâdârï. Kanz al-Durar wa-jâmi ' al-Gurar. Vol. 8: al-Durra al-Dakiyya fï ahbâr al-dawlat al-turkiyya. Haarmann U. (ed.). Cairo-Wiesbaden, 1971. Vol. 9. Roemer H.R. (ed.). Cairo, 1960. (In Arabic)

40. Ibn Duqmaq. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. T. 1 [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 315-330. (In Arabic)

41. Ibn Fadl Allah al-Umarï. al-Ta rïf fï al-mustalah al-sarïf Cairo, 1312/ 18941895. (In Arabic)

42. Ibn al-Furat. Ta'rïh al-duwal wa-l-muluk. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. T. 1 [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 351-357. (In Arabic)

43. Ibn Hajar al- Asqalanï. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. Vol. 1 [Collection of Materials Relating to the Golden Horde History. Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 449-454. (In Arabic)

44. Ibn Higga. Qahwat al-insâ '. Vesely R. (ed.). Beyrouth, Bibliotheca Islamica 36, 2005. (In Arabic)

45. Ibn Katïr. al-Bidâya wa-l-nihâya fï ta'rïh. Vol. 13. Cairo, Matba at al-Saada, 1932. (In Arabic)

46. Ibn Nazir al-Gays. Kitab tatqïf al-tarïf bï-l-mustalah al-sarïf. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. Vol. 1 [Collection of Materials Relating to the Golden Horde History. Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 331-350. (In Arabic)

47. Jackson P. The Dissolution of the Mongol Empire. Central Asiatic Journal. Vol. 32. Wiesbaden, Harrasowitz, 1978, pp. 186-244.

48. Juvaynï, Ala' al-Dïn Ata Malik. Genghis Khan: The History of the World Conqueror. Transl. J.A. Boyle. New introduction and bibliography D.O. Morgan. ManchesterParis, 1997. lxvii + 763 p.

49. Jûzjânï, Minhaj-i Siraj. Tabaqât-i Nâsirï: A General History of the Muhammadan Dynasties of Asia, including Hindustan. Vol. 2. Transl. Major H.G. Raverty. London, 1881 [reprint. New Dehli, 1970].

50. Kedar B. Segurano-Sakran Salvaygo: un mercante Genovese al servizio dei Sultani Mamalucchi, c. 1303-1322. Fatti e idée di storia economica nei secoli XII-XX. Studi dedicati a Franco Borlandi. Bologna, Il Mulino, 1976, pp. 75-91. [Reprint. in The Franks in the Levant, 11th to 14th Centuries. Aldershot, 1993. Art. XXI]. (In Italian)

51. Loiseau J. Les Mamelouks (XIIF-XVF siècle). Une expérience du pouvoir dans l'islam medieval. Paris, Seuil, 2014. 434 p. (In French)

52. Al-Maqrïzï. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. T. 1 [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from

the Arabic Writings]. Tizengauzen V.G. (ed.). St. Petersburg, 1884, pp. 417-442. (In Arabic)

53. Al-Maqrïzï. Kitäb al-suluk li-ma rifat duwal al-muluk. Muhammad Mustafa Ziyada (ed.). Vol. 1/2. Cairo, Maktabat Lagnat al-ta'lïf wa-l-targima wa-l-nasr, 1936. (In Arabic)

54. Martin J. The Land of Darkness and the Golden Horde. Cahiers du Monde russe et soviétique. Vol. 19. Paris, École des hautes études en sciences sociales, 1978, pp. 401-421.

55. Mufaddal Ibn Abï-l-Fadâ'il. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. T. 1 [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 176-199.

56. Mufaddal Ibn Abï-l-Fada'il. Al-Nahj al-sadïd wa-l-durr al-farïd fîma bad Ta'rïh Ibn al-'Amïd. Histoire des sultans Mamlouks. Blochet E. (ed. & transl.). Patrologia Orientalis 12. Paris, Firmin-Didot, 1919, pp. 343-550. (In Arabic, French)

57. Al-Nuwayn. Sbornik materialov", otnosyashchikhsya k" istorii Zolotoy Ordy. T. 1 [Collection of Materials Relating to the Golden Horde History. Volume 1: Extracts from the Arabic Writings]. Tizengauzen V.G. (ed.). Saint Petersburg, 1884, pp. 128-171.

58. Al-Nuwayrï. Nihäyat al-Arab f funun al-adab. Vol. 30. Muhammad Abd al-Hadï Sa'ïra (ed.). Cairo, al-Mu'assasat al-Misrïyat al-'Ämma lil-Ta'lïf wa-l-Tarjama wa-al-Tiba'a wa-al-Nasr, 1990. (In Arabic)

59. Polo, Marco. La description du monde. Hambis L. (ed.). Paris, librairie C. Klinck-sieck, 1955. xvii + 433 p. (In French)

60. Promis V. Continuazione della Cronaca di Jacopo da Varagine dal 1297 al 1332. Atti della Società ligure di storia patria. T. X. Genova, Società ligure di storia patria, 1874, pp. 493-512. (In Latin)

61. Raschid-Eldin [Rasïd al-Dïn]. Histoire des Mongols de la Perse. Quatremère É. (ed. & transl.). Vol.1. Paris, Collection orientale de l'Imprimerie nationale, 1836. (In Persian, French)

62. Rasïd al-Dïn. Sbornik materialov"' otnosyashchikhsya k istorii Zolotoi Ordy. T. 2 [Collection of Materials Relating to the Golden Horde History. Volume 2: Extracts from the Persian Writings]. Tizengauzen V.G. (ed.). Moscow, Leningrad: USSR Academy of Sciences Publ.,1941, pp. 27-79. (In Persian, Russian)

63. Richard J. La conversion de Berke et les débuts de l'islamisation de la Horde d'Or. Revue des études islamiques. Vol. 35. Paris, Librairie Orientaliste Paul Geuthner, 1967, pp. 173-184. (In French)

64. Al-Safadï. A'yän al- 'Asr wa-a'wän al-nasr. Vol. 1. 'A. Abû Zayd (ed.). Beyrouth-Damascus, 1998. (In Arabic)

65. Safi' b.'Alï. Husn al-manäqib al-sirriyya al-Muntaza a mïn al-sïrat al-Zähiriyya. Abd al-'Azïz b. Abd Allah al-Huwaytir (ed.). 2nd ed. Al-Riyad, atabi' al-QUwat al-

Musallahat al-Sa udïya, 1989. (In Arabic)

66. Van Steenbergen J. The Amir Qawsun, Statesman or Courtier? (720-741 AH/1320-1341 AD). Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras. Vermeulen U., Van Steenbergen J. (eds.). Leuven, Peeters, 2001, pp. 449-466.

67. Al-YUnïnï. Dayl Mir'ät al-Zamän. Vol. 1-2. Haydar'abad al-Dakn, Matba at Majlis Da'irat al-Ma'arif al-'Uthmanïya, 1954-1955. (In Arabic)

About the author: Marie Favereau - Ph.D. (History), Research Associate, Faculty of History, Oxford University (George Str., Oxford OX1 2RL, United Kingdom); La Sorbonne - Paris IV (1, Victor Cousin Str., Paris 75005, France). E-mail: marie.favereau@history.ox.ac.uk

Received January 12, 2017 Accepted for publication March 03, 2017

ЗОЛОТАЯ ОРДА И МАМЛЮКИ

Мари Фаверо 12

1 Оксфордский университет OX1 2RL, Оксфорд, Великобритания

2 Университет Сорбонна - Париж IV

75005, Париж, Франция

E-mail: marie.favereau@history. ox. ac. uk

Цель: рассмотреть особенности мамлюкско-золотоордынских взаимоотношений.

Материалы исследования: мамлюкские (биографии султанов, официальные истории, канцелярские руководства и литература инша'), итальянские, ильханидские и тимуридские источники.

Результаты и научная новизна: Золотая Орда и султанат мамлюков обменивались послами на протяжении почти двух столетий. Причины, по которым оба государства поддерживали отношения в течение длительного времени, были сложными и менялись со времен. Данное исследование показывает, что мамлюкско-джучидские взаимоотношения не могут рассматриваться как единый и долгосрочный альянс, но как чередование между периодами молчания и активных взаимообменов. Научная новизна исследования заключается в том, что оно нацелено на выяснение мотивации следовавших друг за другом альянсов, заключенных между султанами и ханами, путем восстановления хронологии этих обменов и тщательного анализа их ключевых моментов. Устоявшееся мнение о "мамлюкско-джучидской дипломатии", описываемой историками в качестве аморфной договоренности, подвергается здесь пересмотру, принимая во внимание внутреннюю ситуацию как в Золотой Орде, так и в султанате мамлюков. Настоящее исследование представляет обзор длительного взаимодействия Золотой Орды с султанатом мамлюков и предлагает ответы на следующие вопросы: оставался ли факт того, что мамлюки и Джучиды имели общих врагов, неотъемлемым краеугольным камнем их дипломатических отношений? Можем ли мы говорить о "внешней политике", а не ограниченной заинтересованности какого-либо отдельного правителя? В результате нашего исследования мы пришли к выводу, что долгосрочные взаимоотношения мамлюкского султаната и Золотой Орды имели сильное влияние не только на обе империи, но и на всю межрегиональную расстановку сил.

Ключевые слова: султанат мамлюков, работорговля, дипломатия, генуэзцы, Черное море, Ильханиды.

Для цитирования: Favereau M. The Golden Horde and the Mamluks // Золотоор-дынское обозрение. 2017. Т. 5, № 1. C. 93-115. DOI: 10.22378/2313-6197.2017-5-1.93115

Сведения об авторе: Мари Фаверо-Думенжу - Ph.D. (история), научный сотрудник, исторический факультет, Оксфордский университет (OX1 2RL, ул. Георгия, Оксфорд, Великобритания); университет Сорбонна - Париж IV (75005, ул. Виктор Кузин, 1, Париж, Франция). E-mail: marie.favereau@history.ox.ac.uk

Поступила 12.01. 2017 г.

Принята к публикации 03.03.2017 г.