The Horde

by Marie Favereau

  In 1237 Ögödei went a step further and established a preferential marriage system between the male and female descendants of Chinggis Khan and the Qonggirad. Imperial chief wives were thus supposed to be of Qonggirad origin, although Ögödei’s order was not rigorously followed. The decision probably reflected the fact that Börte, Chinggis Khan’s first wife, was the daughter of Dei Sechen, chief of the Qonggirad. A khan had many khatun because marriage was a political partnership, but only few of the women were chief wives, with their own extended households. Secondary wives and concubines often stayed with chief wives who controlled them. The chief wife could be highly influential; she might have her own court of secretaries, treasurers, and traders and sit at the quriltai.10

  The customs of the Horde, including patterns of succession and marriage, were influenced by those of the old Mongol Empire. The people of Jochi considered the sons of the Qonggirad khatuns the preferred candidates to the throne, and, in the long run, the Qonggirad remained the primary quda, marriage partners, of the Jochids. Eight of Jochi’s wives are mentioned by name in the old genealogies. Among them, Sarqadu-khatun of the Qonggirad was the eldest and was the mother of Jochi’s first son, Orda. One of his younger wives, the daughter of a Qonggirad noyan, gave birth to Batu. These princes would both rule. So would a third, Berke, who was the son of Jochi and Sultan-khatun, a daughter of the Khwarezmian shah. Berke would have to fight harder than his brothers for a place on the Jochid throne, owing to his mother’s lower rank.

  No record states exactly how many descendants Jochi had in the 1250s, but Rashīd al-Dīn writes that “he had nearly forty sons and innumerable grandsons.” The Jochids thus flourished quickly. In only two generations, their house had grown to include several hundred people. They were born of an alliance between the Qonggirad and the Borjigid through the male line of Jochi.11

  The Shift Men

  Thirty years after Chinggis Khan had given 4,000 men to Jochi, the same warriors and their descendants still lived with their families in the Qipchaq steppe. They were forever attached to the service of the Jochid khans. The Mongols called these loyal servants keshigten, the shift men. In the steppe world, there was no khan without a keshig.12

  Following the structure installed by Chinggis, Batu organized his shift men according to a cycle of twelve days—monkey, hen, dog, pig, mouse, ox, tiger, hare, dragon, snake, horse, and sheep days. Batu had four shifts and four keshig elders, one to lead each shift. Every three days one of Batu’s keshig elders took over and commanded the shift men on duty. The four shifts were the stem cells from which the Horde originated; they protected and served the khan by administrating and controlling his people.13

  The shifts were divided into night guards, day guards, and quiver-bearers. Batu might have had as many as 1,000 night guards, 2,000 quiver-bearers, and 8,000 day guards. Their numbers varied over time, and the Jochid khan had smaller shifts than did the great khan, but both keshigs were organized the same way. All Mongol keshigs, Jochid and otherwise, also included cooks and stewards as well as chancellery members such as secretaries, accountants, and translators. The keshig was neither a warband—a small nomadic force—nor an army. It was the khan’s government.

  The kebte’ül, the night guards, were vital, for they were in charge of the khan’s household. Their primary task was to run the supply system that provided the court with food, drinks, games, and music. While the horde was settled down, the night guards watched the ordo geren, the khan’s tents, and set everything in order. The khan had a giant palace-tent, which the Mongols called the golden tent. Around it was a security perimeter called the ordo: a precinct where shift men were stationed and where they ate. The precinct was an independent unit within the camp, with its own rules and administration. The ordo was mobile and followed the khan’s migration route. In general, Mongol camps were malleable structures, the layout of which could morph as needed to regulate the circulation of people and goods. In the case of the khan’s headquarters, the organization could be altered to permit selective access to particular tents and to court assemblies.14

  Khans were close to their shift men, some of whom would be known to the khans throughout their lives. An heir apparent—in fact, any important member of the golden lineage—began to recruit his own keshig in his youth. They rode together and engaged in manly activities and would exclude all others from their hunting parties. A new khan would inherit a part of his predecessor’s keshig, which he would combine with his own. When a khan died, his night guards normally stayed in his horde, which would be turned over to the command of his chief widow. These loyal followers worshipped their master’s soul for the rest of their lives. A horde survived its founder through the night guards, who became the founder’s memorializers.15

  Official positions were kept in families and transmitted from father to son or nephew, so keshig elders could bequeath their role to the next generation. A keshig elder often attempted to control his own succession and when assigned a faraway mission such as delivering a diplomatic letter or overseeing tax collection, he would delegate his shift responsibility to a son or to another subordinate who could replace him if he did not come home. But a new khan still nominated his own men, although he might favor existing shift men. Batu renewed at least half of Jochi’s keshig.16

  While the keshig collaborated with and served the Jochids, the two groups were genealogically distinct. The descendants of Jochi did not belong to the shift men, and the keshig elders did not typically marry princesses from the golden lineage. When intermarriage did occur, it did not result in familial political alliance. To preserve the balance of power, Batu kept the Qonggirad and other powerful in-laws away from influential keshig positions. In this respect, he followed Chinggis Khan’s rules of kinship control, which affirmed the idea that a family-run empire would be a stillborn empire. It was essential, therefore, that elite families have their separate tasks and expectations. It was also with power-balancing in mind that khans manipulated lineages, interfering with genealogy in order to promote those family members loyal to themselves while ensuring that rival family members would not accrue too much influence. For political reasons, khans could convert a junior into a senior and grant a teenager a keshig elder’s position.17

  Keshig elders were among the begs, the aristocracy. In the Horde the elders and other non-Jochid officials were also known as qarachu, the western steppe term for bo’ol. The qarachu begs were the elite of the subordinated peoples, serving the golden lineage and given charge of important management and military positions. Even in the sixteenth century, most of the qarachu begs claimed to descend from Jochi’s four original minggan, the regiments of elite steppe warriors Chinggis had gifted his son. As outsiders to the golden lineage, the qarachu begs were strictly prevented access to the supreme position, but the khans could never rule without them.18

  Each Jochid horde had its own keshig, whose members did not necessarily come from their chief’s family. Again, this structure served to prevent the emergence of power centers independent of the khan, for keshig members would be grouped through a kind of military structure, not by family or ethnicity. The shift men were recruited from anywhere in the empire, as a group or on an individual basis. Many came from the conquered peoples—Qipchaqs, Alan, Russians, Hungarians, and others seized the opportunity to become Mongol shift men.

  In addition the keshig included elite hostages from afar—future leaders, whom the khan held as a token of their noble relatives’ loyalty. Batu’s keshig, for instance, counted a large number of sons of kniazia.19 The hostage-taking practice was deeply alien to Westerners like John of Plano Carpini, a Franciscan friar who traveled the Mongol Empire and learned of the hostage-taking program. Plano Carpini perceived the practice as a means to destroy foreign aristocracy. “Of those whom they allowed to return they demand sons or brothers and they never afterwards give these their liberty,” he wrote. “This is how they have treated the son of Jerozlaus and a chief of the Alans and many others,” a referenc
e to Alexander Nevsky, son of Kniaz Ieroslav of Vladimir-Suzdal. In fact Plano Carpini was wrong. The Mongols did not keep their elite hostages for life. Rather, the Mongols trained their hostages both to lead and to obey, so that they could return home as vassals, ruling their homelands in the name of the khan—with his full support, including military assistance and proof of investiture. Taking hostages was an old steppe diplomatic institution that processed outsiders into the nomads’ social systems and built long-term political relationships. Chinggis’s success with the practice had confirmed that the loyalty of the vassals originated with their physical presence in the ruler’s court. Once more, the Jochid khans copied Chinggis’s imperial methods.20

  The quriltai was another linchpin of Mongol political life, which the Jochids retained. The Jochids assembled in quriltais at least twice a year: at the lunar new year, which coincided with January or February, and during the sixth lunar month, which began toward the end of June. Thousands of people assembled under a huge tent, including male and female descendants and next-of-kin of Jochi; their in-laws; the keshig elders; and a number of shift men, especially secretaries, accountants, and stewards. Each attendee could bring a limited number of guests. The rest of the keshig watched and served the assembly. Attendees entered, sat, ate, and toasted according to strict precedence; women took their seats on the khan’s eastern side and men on his western side. The guests brought gifts to the khan, and he granted them cloth, silver, and gold. The assembly divided itself into aqa and ini, older and younger brothers. The aqa were more influential than the ini but could not rule without them, as both groups had to validate the khan’s decisions in order to make them law.21

  A quriltai was a closed-door meeting. The Mongols did not share internal politics with foreigners, who were denied access to crucial meetings and rituals. Given that foreigners provided most contemporaneous written accounts of Mongol life, it is hard to say exactly how the Jochid assembly functioned, and in what way its rules might have differed from those used in the East, which were established by Ögödei in 1229 and codified in 1234. What we do know is that attendees ruled on judicial cases, consequential marriages, key investitures, rewards, war, and diplomacy.22

  In the mid-thirteenth century, the new elites Chinggis Khan had shaped began to crystallize socially. In the Horde, these elites helped to establish the Jochids as a distinctive group, albeit one whose robust political institutions—quriltai, keshig, patterns of succession, and so on—mirrored those of the empire as a whole. As we will see later, the sons of Jochi maintained these institutions even as they achieved self-governance. The Jochids also modified the institutions they inherited, learning to apply Mongol political theory and practice under novel conditions and drawing on tradition to support their evolving regime.

  The Seasons of the Khan

  Batu’s residence in the lower Volga was in many ways ideal. It was safe, about five hundred miles from the Dnieper River, where the Lithuanians often conducted devastating raids, and from the Caucasus, where rebellious Alan and Circassian groups hid in the mountains. The lower Volga also provided excellent pastures and salt, as Sübötei and Jebe had discovered when the Westward first wintered there in 1222–1223. In spring, when insects proliferated in the humid delta, Batu moved his horde two hundred miles upriver, where the air was cool and healthy. And the whole area of the lower Volga was a crossroads of water and land routes. In winter the frozen rivers became a constellation of roads that converged in the lower valleys of the Dnieper, Don, and Volga. In summer riverboats connected the hordes and Russian villages and towns within a day or two’s travel.23

  Yet there would soon be limits to the seemingly unstoppable push of the Mongols, and the allotment of the territories fixed during the time of Orda and Batu would remain unchanged over the next generations. The routes of seasonal migrations also varied little, but the number of tents increased every year, and the areas of the winter and summer camps attracted ever more permanent settlers. The growing population was not a result of conquest, for the Jochids were largely at peace during the first decade of Batu’s and Orda’s rule. Rather, demographic increase was both the cause and the result of intensive pastoral activities. The growth was also supported by expanding trade opportunities, for, while at peace, the Jochids had no other way to obtain the luxury goods necessary for diplomacy and for sustaining the sharing system that bound elites to the khan and commoners to elites, as I describe in detail below. A combination of commerce and sophisticated herding techniques bolstered the Horde when it could not rely on the spoils of war.24

  A basic challenge was to achieve economic efficiency while also enabling the social interaction that was so crucial to Mongol politics. The hordes had to be mobile in order to allow sufficient grazing, lest overgrazing in one spot damage the steppe ecology. But, at the same time, they had to convene in order to carry out political meetings. Another pitfall was the difficulty of grazing at a scale sufficient to support the population. Doing so required heavy labor. To solve their economic challenges while enabling political life, the Mongols developed several tactics. First, war captives—when available—and outsiders without status were forced to contribute to the collective workload. Second, they multiplied satellite camps and markets that helped to supply the hordes. Finally, they turned seasonality into a political instrument, scheduling their political activities during times of year when their subsistence needs were most easily filled. They also accepted that sometimes political demands would result in economic losses.25

  Why so? Because when it was necessary to hold a political gathering, the Mongols would leave a part of their herds behind and force the rest to travel faster. The people of the hordes, who lived far away from one another, consumed a lot of energy to attend yearly trade fairs and festivals of political significance. These included quriltai, colossal gatherings organized at least twice a year. These interactions were as crucial to Mongol life as healthy herding, leading inevitably to friction between politics and herding. When a chief wanted to delay a quriltai, he would argue that his animals needed to fatten, preventing him from traveling. This was not a metaphor.

  The new strategy for sustaining population growth began with Great Khan Ögödei and his successors. Their basic pastoral needs grew to such an extent that local resources were becoming exhausted, so Ögödei built a formidable infrastructure to supply Qaraqorum and his people’s hordes. Ögödei achieved efficiency far beyond of earlier pastoral production. But there was more to Ögödei’s program than efficiency: he turned the subsistence strategy of his forefathers into a power strategy, in which the movements of his hordes followed not only the demands of grazing but also of politics, even when mobility was costly to the herds. Ögödei and his descendants developed a system in which they could tolerate economic losses as they focused on politics—gathering to perform rituals, talk, eat, and drink.26

  Like the great khans in the east, the Jochids developed efficiencies, markets, and political practices that enabled them to supply a vastly larger population than those of traditional nomads while still convening the dispersed hordes. Batu’s supply camps were perhaps modeled after Ögödei’s tergen yam, the relay supply system for heavy loads, which delivered to Qaraqorum five hundred wagons of food and drink every day. The Jochid logistics were less spectacular and more mobile, as Batu never accumulated the resources and manpower of Ögödei. Still, Batu did everything necessary to implement the new style of pastoral economy and politics. He expanded trade activities, allowed a huge number of workers in his horde, and organized drinking festivals to which the horde’s people were able to travel in spite of the costs of migration. Batu’s shift men were conscious of the dangers of overgrazing and responded to them by multiplying the number of satellite camps.

  The fattening of horses and camels during periods of calving and milking was crucial to the pastoral economy. During these months, usually from May to September, the herds needed to rest. When the mares were milking, they did not march with the khan�
�s horde. The Mongols used this five-month-long season not only to relax—these were essentially peaceful stretches—but also to organize extensive political meetings and take governing decisions. It was no accident that the Mongols planned enthronements and great quriltai during the drinking festival they held in summer.

  Batu’s satellite camps provided the court with horses, sheep, cattle, and enormous quantities of dairy and meat products. All of it was the property of the golden lineage. In the western steppe, the satellite camps marched separately and usually preceded the khan’s camp so that supplies could be organized at an agreed-upon stopping point before the khan arrived. One satellite camp was reserved for keepers of birds of prey; khans collected falcons, for they were symbols of rule and powerful weapons in the hunt for birds, rabbits, marmots, foxes, wolves, and small Saiga antelope. The Mongols hunted not only for meat but also for skins and furs. Shift men separated falcon farming, herding, and other supply camps from the horde; they also supervised the khan’s hunt. Keeping distance between satellite camps and hordes was a valuable sanitation practice, and the dispersal of animals and people helped to prevent overgrazing around the khan’s horde.27

  Besides supply camps, the Jochid hordes had mobile markets. These were likely under the supervision of night guards. Markets stood at the extreme ends of a given camp; in the khan’s camp, one needed a horse to travel from the orda, the ruler’s precinct, to the market. The Franciscan missionary William of Rubruck noted that “a market always follows Batu’s orda, but it was so far away from us that we could not go to it.” In the steppe, families knew how to sustain themselves and had no need to visit the markets every day. Poor visitors at the margins of the social system relied exclusively on the generosity of others to access food and drink for survival.28