Wednesday, March 28, 2007

The Story of the Mongols Whom We Call the Tartars by Friar Giovanni DiPlano Carpini

The Story of the Mongols Whom We Call the Tartars

Branden Publishing Company, Boston

Excerpt from Friar Giovanni Di Plano Carpini's Account of his Embassy to the Court of the Mongol Khan (1245-1247).

John of Plano Carpini was a Franciscan monk sent in 1245 by Pope Innocent IV to learn of Mongol political intentions and bring Christianity to the East. These two excerpts from the History are first, his description of "Tartar" customs, and next, his account of events at the great assembly near Karakorum at which Guyak was proclaimed Great Khan.

Chapter IV

These men, that is to say the Tartars, are more obedient to their masters than any other men in the world, be they religious or seculars; they show great respect to them nor do they lightly lie to them. They rarely or never contend with each other in word, and in action never. Fights, brawls, wounding, murder are never met with among them. Nor are robbers and thieves who steal on a large scale found there; consequently their dwellings and the carts in which they keep their valuables are not secured by bolts and bars. If any animals are lost, whoever comes across them either leaves them alone or takes them to men appointed for this purpose; the owners of the animals apply for them to these men and they get them back without any difficulty. They show considerable respect to each other and are very friendly together, and they willingly share their food with each other, although there is little enough of it. They are also long-suffering. When they are without food, eating nothing at all for one or two days, they do not easily show impatience, but they sing and make merry as if they had eaten well. On horseback they endure great cold and they also put up with excessive heat. Nor are they men fond of luxury; they are not envious of each other; there is practically no litigation among them. No one scorns another but helps him and promotes his good as far as circumstances permit.

Their women are chaste, nor does one hear any mention among them of any shameful behaviour on their part; some of them, however, in jest make use of vile and disgusting language. Discord among them seems to arise rarely or never, and although they may get very drunk, yet in their intoxication they never come to words or blows.

Now that the good characteristics of the Tartars have been described, it is time for something to be said about their bad. They are most arrogant to other people and look down on all, indeed they consider them as nought, be they of high rank or low born.

For at the Emperor's court we saw Jerozlaus, a man of noble birth, a mighty duke of Russia, also the son of the King and Queen of Georgia, and many important sultans; the chief also of the Solangi received no fitting honour from them, but the Tartars who were assigned to them, however base-born they were, went ahead of them and always had the first and highest place; indeed they were often obliged to sit behind their backs.

They consider it a great sin if any food or drink is allowed to be wasted in any way; consequently they do not allow bones to be given to dogs until the marrow has been extracted. They do not wash their clothes nor allow them to be washed, especially from the time when thunderstorms begin until the weather changes. They drink mare's milk in very great quantities if they have it; they also drink the milk of ewes, cows, goats and even camels. They do not have wine, ale or mead unless it is sent or given to them by other nations. In the winter, moreover, unless they are wealthy, they do not have mare's milk. They boil millet in water and make it so thin that they cannot eat it but have to drink it. Each one of them drinks one or two cups in the morning and they eat nothing more during the day; in the evening, however, they are all given a little meat, and they drink the meat broth. But in the summer, seeing they have plenty of mare's milk, they seldom eat meat, unless it happens to be given to them or they catch some animal or bird when hunting.

They also have a law or custom of putting to death any man and woman they find openly committing adultery; similarly if a virgin commit fornication with anyone, they kill both the man and the woman. If anyone is found in the act of plundering or stealing in the territory under their power, he is put to death without any mercy. Again, if anyone reveals their plans, especially when they intend going to war, he is given a hundred stripes on his back, as heavy as a peasant can give with a big stick. When any of the lower class offend in any way, they are not spared by their superiors, but are soundly beaten. There is no distinction between the son of a concubine and the son of a wife, but the father gives to each what he will; and if they are of a family of princes, then the son of a concubine is a prince just the same as the son of a legitimate wife. When a Tartar has many wives, each one has her own dwelling and her household, and the husband eats and drinks and sleeps one day with one, and the next with another. One, however, is chief among the others and with her he stays more often than with the others. In spite of their numbers, they never easily quarrel among themselves.

The men do not make anything at all, with the exception of arrows, and they also sometimes tend the flocks, but they hunt and practise archery, for they are all, big and little, excellent archers, and their children begin as soon as they are two or three years old to ride and manage horses and to gallop on them, and they are given bows to suit their stature and are taught to shoot; they are extremely agile and also intrepid.

Young girls and women ride and gallop on horseback with agility like the men. We even saw them carrying bows and arrows. Both the men and the women are able to endure long stretches of riding. They have very short stirrups; they look after their horses very well, indeed they take the very greatest care of all their possessions. Their women make everything, leather garments, tunics, shoes, leggings and everything made of leather; they also drive the carts and repair them, they load the camels, and in all their tasks they are very swift and energetic. All the women wear breeches and some of them shoot like the men.

Chapter IX
The Countries through which we passed, their position, the witnesses we came across, and the Court of the Emperor of the Tartars and his Princes

…and many a time we arrived so late that we did not eat that night but were given in the morning the food we should have eaten the previous evening. We went as fast as the horses could trot, for the horses were in no way spared since we had fresh ones several times a day, and those which fell out returned, as has already been described, and so we rode swiftly without a break.

On our arrival Cuyuc had us given a tent and provisions, such as it is the custom for the Tartars to give, but they treated us better than other envoys. Nevertheless we were not invited to visit him for he had not yet been elected, nor did he yet concern himself with the government. The translation of the Lord Pope's letter, however, and the things I had said had been sent to him by Bati. After we had stayed there for five or six days he sent us to his mother where the solemn court was assembling. By the time we got there a large pavilion had already been put up made of white velvet, and in my opinion it was so big that more than two thousand men could have got into it. Around it had been erected a wooden palisade, on which various designs were painted. On the second or third day we went with the Tartars who had been appointed to look after us and there all the chiefs were assembled and each one was riding with his followers among the hills and over the plains round about.

On the first day they were all clothed in white velvet, on the second in red-that day Cuyuc came to the tent-on the third day they were all in blue velvet and on the fourth in the finest brocade. In the palisade round the pavilion were two large gates, through one of which the Emperor alone had the right to enter and there were no guards placed at it although it was open, for no one dare enter or leave by it; through the other gate all those who were granted admittance entered and there were guards there with swords and bows and arrows. If anyone approached the tent beyond the fixed limits, he was beaten if caught; if he ran away he was shot at, but with arrows however which had no heads. The horses were, I suppose, two arrow-flights away. The chiefs went about everywhere armed and accompanied by a number of their men, but none, unless their group of ten was complete, could go as far as the horses; indeed those who attempted to do so were severely beaten. There were many of them who had, as far as I could judge, about twenty marks' worth of gold on their bits, breastplates, saddles and cruppers. The chiefs held their conference inside the tent and, so I believe, conducted the election. All the other people however were a long way away outside the afore-mentioned palisade. There they remained until almost mid-day and then they began to drink mare's milk and they drank until the evening, so much that it was amazing to see. We were invited inside and they gave us mead as we would not take mare's milk. They did this to show us great honour, but they kept on plying us with drinks to such an extent that we could not possibly stand it, not being used to it, so we gave them to understand that it was disagreeable to us and they left off pressing us.

Outside were Duke Jerozlaus of Susdal in Russia and several chiefs of the Kitayans and Solangi, also two sons of the King of Georgia, the ambassador of the Caliph of Baghdad, who was a Sultan, and more than ten other Sultans of the Saracens, so I believe and so we were told by the stewards. There were more than four thousand envoys there, counting those who were carrying tribute, those who were bringing gifts, the Sultans and other chiefs who were coming to submit to them, those summoned by the Tartars and the governors of territories. All these were put together outside the palisade and they were given drinks at the same time, but when we were outside with them we and Duke Jerozlaus were always given the best places. I think, if I remember rightly, that we had been there a good four weeks when, as I believe, the election took place; the result however was not made public at that time; the chief ground for my supposition was that whenever Cuyuc left the tent they sang before him and as long as he remained outside they dipped to him beautiful rods on the top of which was scarlet wool, which they did not do for any of the other chiefs. They call this court the Sira Orda.

Leaving there we rode all together for three or four leagues to another place, where on a pleasant plain near a river among the mountains another tent had been set up, which is called by them the Golden Orda; it was here that Cuyuc was to be enthroned on the feast of the Assumption of Our Lady, but owing to the hail which fell, as I have already related, the ceremony was put off. This tent was supported by columns covered with gold plates and fastened to other wooden beams with nails of gold, and the roof above and the sides on the interior were of brocade, but outside they were of other materials. We were there until the feast of St. Bartholomew, on which day a vast crowd assembled. They stood facing south, so arranged that some of them were a stone's throw away from the others, and they kept moving forward, going further and further away, saying prayers and genuflecting towards the south. We however, not knowing whether they were uttering incantations or bending the knee to God or another, were unwilling to genuflect. After they had done this for a considerable time, they returned to the tent and placed Cuyuc on the imperial throne, and the chiefs knelt before him and after them all the people, with the exception of us who were not subject to them. Then they started drinking and, as is their custom, they drank without stopping until the evening. After that cooked meat was brought in carts without any salt and they gave one joint between four or five men. Inside however they gave meat with salted broth as sauce and they did this on all the days that they held a feast.

At that place we were summoned into the presence of the Emperor, and Chingay the protonotary wrote down our names and the names of those who had sent us, also the names of the chief of the Solangi and of others, and then calling out in a loud voice he recited them before the Emperor and all the chiefs. When this was finished each one of us genuflected four times on the left knee and they warned us not to touch the lower part of the threshold. After we had been most thoroughly searched for knives and they had found nothing at all, we entered by a door on the east side, for no one dare enter from the west with the sole exception of the Emperor or, if it is a chief's tent, the chief; those of lower rank do not pay much attention to such things. This was the first time since Cuyuc had been made Emperor that we had entered his tent in his presence. He also received all the envoys in that place, but very few entered his tent.

So many gifts were bestowed by the envoys there that it was marvellous to behold-gifts of silk, samite, velvet, brocade, girdles of silk threaded with gold, choice furs and other presents. The Emperor was also given a sunshade or little awning such as is carried over his head, and it was all decorated with precious stones. A certain governor of a province brought a number of camels for him, decked with brocade and with saddles on them having some kind of contrivance inside which men could sit, and there were, I should think, forty or fifty of them; he also brought many horses and mules covered with trappings or armour made of leather or of iron. We in our turn were asked if we wished to present any gifts, but we had by now used up practically everything, so had nothing to give him. There up on a hill a good distance away from the tents were stationed more than five hundred carts, which were all filled with gold and silver and silken garments, and these things were shared out among the Emperor and the chiefs. Each chief divided his share among his men, but according to his own good pleasure.

Leaving there we went to another place where a wonderful tent had been set up all of red velvet, and this had been given by the Kitayans; there also we were taken inside. Whenever we went in we were given mead and wine to drink, and cooked meat was offered us if we wished to have it. A lofty platform of boards had been erected, on which the Emperor's throne was placed. The throne, which was of ivory, was wonderfully carved and there was also gold on it, and precious stones, if I remember rightly, and pearls. Steps led up to it and it was rounded behind. Benches were also placed round the throne, and here the ladies sat in their seats on the left; nobody, however, sat on the right, but the chiefs were on benches in the middle and the rest of the people sat beyond them. Every day a great crowd of ladies came.

The three tents of which I have spoken were very large. The Emperor's wives however had other tents of white felt, which were quite big and beautiful. At that place they separated, the Emperor's mother going in one direction and the Emperor in another to administer justice. The mistress of the Emperor had been arrested; she had murdered his father with poison at the time when their army was in Hungary and as a result the army in these parts retreated. Judgment was passed on her along with a number of others and they were put to death. …

Monday, March 19, 2007

Old Mongolian Manuscript Maps Website

Old Maps of Mongolia - Source :
Website Name: Old Mongolian Manuscript Maps from the website: 'Perception of the Landscape by the Mongols'.

Mongolia Map 1932

Mongolian Map 1902

Mongolian Map 1864

Old Mongolian Manuscript Maps from the website: 'Perception of the Landscape by the Mongols'. Website:

Sunday, March 18, 2007

Manchu Language Dying Out

Meng Shujing with her grandson, Shi Junguang, and great-grandson, Shi Yaobin, in their hometown of Sanjiazi.
China's Manchu speakers struggle to save language

By David Lague
Published: International Herald Tribune,March 13, 2007

SANJIAZI, China: Seated cross- legged in her farmhouse on the kang, a brick sleeping platform warmed by a fire below, Meng Shujing lifted her chin and sang a lullaby in Manchu, softly but clearly.

After several verses, the 82-year-old widow stopped, her eyes shining.

"Baby, please fall asleep quickly," she said, translating a few lines of the song into Chinese. "Once you fall asleep, Mama can go to work. I need to set the fire, cook and feed the pigs."

After 5 children, 14 grandchildren and 5 great-grandchildren, Meng has the confidence that comes from long experience. "If you sing like this, a baby gets sleepy right away," she said.

She also knows that most experts believe the day is approaching when no child will doze off to the sound of these comforting words.

Ms. Meng is one of 18 residents of this isolated village in northeastern China, all older than 80, who, according to Chinese linguists and historians, are the last native speakers of Manchu.

Descendants of seminomadic tribesmen who conquered China in the 17th century, they are the last living link to a language that for more than two and a half centuries was the official voice of the Qing Dynasty, the final imperial house to rule from Beijing and one of the richest and most powerful empires the world has known.

With the passing of these villagers, Manchu will also die, experts say. All that will be left will be millions of documents and files in Chinese and foreign archives, along with inscriptions on monuments and important buildings in China, unintelligible to all but a handful of specialists.

"I think it is inevitable," said Zhao Jinchun, an ethnic Manchu born in Sanjiazi who taught at the village primary school for more than two decades before becoming a government official in the city of Qiqihar, 50 kilometers, or 30 miles, to the south. "It is just a matter of time. The Manchu language will face the same fate as some other ethnic minority languages in China and be overwhelmed by the Chinese language and culture."

(While most experts agree that Manchu is doomed, Xibo, a closely related language, is likely to survive a little longer. Xibo is spoken by about 30,000 descendants of members of an ethnic group allied to the Manchus who in the 18th century were sent to the newly conquered western region of Xinjiang. But it too is under relentless pressure from Chinese.)

The disappearance of Manchu will be part of a mass extinction that some experts forecast will lead to the loss of half of the world's 6,800 languages by the end of the century. But few of these threatened languages have risen to prominence and then declined as rapidly as Manchu.

Within decades of establishing their dynasty in 1644, the Qing rulers had brought all of what was then Chinese territory under control. They then embarked on a campaign of expansion that roughly doubled the size of their empire to include Xinjiang, Tibet, Mongolia and Taiwan. However, the dynasty's fall in 1911 meant that the Manchus were relegated to the ranks of the more than 50 other ethnic minorities in China, their numbers dwarfed by the dominant Han, who today account for 93 percent of the country's 1.3 billion people, according to official statistics.

Indistinguishable by appearance, the Manchus have melded into the general population. There are now about 10 million Chinese citizens who describe themselves as ethnic Manchus. Most live in what are now the northeastern provinces of Liaoning, Jilin and Heilongjiang, although there are also substantial numbers in Beijing and other northern cities.

For generations, the vast majority have spoken Chinese as their first language. Manchu survived only in small, isolated pockets like Sanjiazi, where, until a few decades ago, nearly all the residents were ethnic Manchus. Most are descended from the three main families that made up a military garrison established here in 1683 on the orders of the Qing emperor, Kangxi, to deter Russian territorial ambitions, according to Zhao.

The traditional Manchu-style wood- and-adobe farmhouses have largely been replaced by Chinese-style brick homes, the local residents say. The village now looks just like any other settlement in this region as a biting wind whips snow across the bare ground between the houses and the piles of dried cornstalks, stacked high to feed cattle and pigs through the winter.

Traditional shamanistic rites, along with ethnic dress and customs, have also been mostly abandoned, although some wedding and funeral ceremonies retain elements of Manchu rituals, Zhao said. But, villagers still observe one Manchu taboo that sets them apart from others in China's far northeast."We don't eat dog meat," Zhao said. "And we would never wear a hat made from dog fur." The prohibition, tradition has it, honors a dog credited with having saved the life of Nurhachi, the founder of the Manchu state, who lived from 1559 to 1626.

Even now, about three-quarters of Sanjiazi's 1,054 residents are ethnic Manchus but the use of Chinese has increased dramatically in recent decades as roads and modern communications have increasingly exposed them to the outside world. Only villagers of Meng's generation now prefer to speak Manchu.

"We are still speaking it, we are still using it," said Meng, a cheerful woman with thick gray hair pulled back in a neat bun. "If the other person can't speak Manchu then I'll speak Chinese."

But Meng disputes the findings of visiting linguists that there are 18 villagers left who can still speak fluently. By her standards, only five or six of her neighbors are word-perfect in Manchu.

Zhao, 53, on the other hand, estimates that about 50 people in the village have a working grasp of the language.

"My generation can still communicate in Manchu," he said, although he acknowledged that most villagers speak Chinese almost all the time at home.

Meng supports efforts to keep the language alive. Her 30-year-old grandson, Shi Junguang, has studied hard to improve his Manchu and teaches speaking and writing to the 76 pupils, 7 to 12 years old, at the village school.

This is the only primary school in China that offers classes in Manchu, according to officials from the local ethnic affairs office. These lessons, which Shi shares with one other teacher, take up only a small proportion of classroom time but they are popular with students, say the school's staff and other residents in the village.

"Because they are Manchus, they are interested in these classes," Shi said.

He is also teaching basic conversation to his 5-year-old son, Shi Yaobin, and encourages him to speak with his great-grandmother. "It would be a great blow for us if we lose our language," he said.

But most experts say that with so few people left to speak it, attempts to preserve Manchu are futile.

"The spoken Manchu language is now a living fossil," said Zhao Aping, an ethnic Manchu and an expert on Manchu language and history at Heilongjiang University in the provincial capital, Harbin. "Although we are expending a lot of energy on preserving the language and culture, it is very difficult. The environment is not right."

While scholars agree it is now only a matter of time before Manchu falls silent, in Sanjiazi, Meng clings to hope.

"I don't have much time," she said. "I don't even know if I have tomorrow. But I will use the time to teach my grandchildren.

"It is our language, how can we let it die? We are Manchu people."

Sunday, March 4, 2007

Dietary Decadence and Dynastic Decline in the Mongol Empire

Mongol cup 13th - 14th century


by Professor John Masson Smith, Jr.

Published in the Journal of Asian History, 34/1 (2000)

[may contain uncorrected botched accents, etc]

University of California, Berkeley



Most Mongol rulers lived short lives. Those in the Middle East died, on average, at about age 38, and the successors of Qubilai in the Far East at 33 (adding in Qubilai raises the average since he lived, atypically, for 78 years; Chinggis lived into his 60s; for the rest, few passed 50). Comparison of the Mongol and Manchu (Ch'ing) dynasties shows the importance of longevity. In each of the Mongol realms of China, the Middle East and the Golden Horde, an average of eleven Mongols ruled for an average of about a century (107 years):Qubilai and nine successors ruled China for 110 years (1260-1370); the Golden Horde had twelve khans in 132 years (1227-1359); and nine Mongols held the Middle East for 80 years (1255-1335). Nine Manchus, with an average reign of 29 years, occupied the throne of China for over two and a half centuries (1644-1908).

The Middle Eastern Mongol dynasty had further problems: high infant mortality and infertility. Ann Lambton considers that "the possibility cannot be ruled out that once the Mongols settled in Persia, they ceased to be good breeders."1 I suggest that the Mongols' difficulties stemmed in large part from dietary inadequacies and improprieties.

The diet of pre-imperial Mongols was simple, calorically-sufficient--and poorly balanced. Then as now (or until very recently) the average Mongol family possessed a herd consisting largely of sheep, with some goats, and a few each of bovines and camels.2 Then, however, families kept more horses (ponies, actually) to maintain a military capability. For decent subsistence, a family required 100 sheep or the equivalent; for its military role, at least five (gelding) ponies; besides these, perhaps three more ponies and some oxen and camels were useful for transportation; and a mare or two for milking. From these animals the Mongols, like the other nomads of Inner Asia, obtained most of their food.3 In the words of John of Plano Carpini, who visited the Mongols in the 1240s: "[The Mongols] have neither bread nor herbs nor vegetables nor anything else, nothing but meat....They drink mare's milk in very great quantities if they have it; they also drink the milk of ewes, cows, goats and even camels."4 Although many nomads exchange animal products for goods, including foods, from settled peoples, those of Outer Mongolia, where the Mongols under Chinggis Khan got their start, were (and still are) a long way from the nearest substantial farmlands, and for them imported food would have been an expensive, and for the average family, no doubt rare luxury.5

The Mongols' main meat foods were mutton and lamb; although by all accounts, their favorite was horse-meat, it was a preference that the average family could seldom indulge. The other principal type of food was milk (in various processed forms), again chiefly from sheep, but mare's milk by preference. The predominance of sheep in the herd and the importance of mutton and sheep's milk in the diet, as well as the predilection for horse-meat, probably arose from the very high caloric value of these foods, a matter of central importance for practitioners of the hard--and in Mongolia, cold--nomadic life. Beef--our meat mainstay--has 1073 kilocalories [kcal--the same "calories" that we count when dieting] per lb; mutton has 1834, and horse-meat 1855. Likewise, cow's (whole) milk provides around 400 kcal/lb, and sheep's milk 511.6 Pound for pound, pint for pint, you get the best caloric return from sheep and horses.

Fat provides most of these calories: 89% of them in the case of mutton, which is 40% fat; and 67% with sheep's milk (7.5% fat). One low-fat food was available. Since most of the Mongols' animals provided milk for only about 5 months a year (cf. cows at 10-11 months), the Mongols had to process milk into forms that would keep well during the 7 "dry" months. They rendered cow's milk into a dried skim milk solid, the approximate equivalent of our non-fat milk powder. But they kept and ate the by-product, butter, offsetting the healthful effect of the dried skim.7

Interestingly, poor Mongols probably benefited from a better-balanced diet. After Chinggis' father died, most of his family's herds were stolen, so that his mother had to feed her children edible plants: wild pears, bird cherries, garden burnet root, cinquefoil root, wild onion, shallot, lily root, and garlic chives. Despite this diet of what the Mongols considered second-rate foods, Chinggis and the other boys "grew up into fine men" in the words of the Secret History.8

The Mongols, as mentioned, had (and have) a great liking for mare's milk. Not on account of richness of the milk, which, by comparison with the milks of other domesticated animals, is virtually a diet drink at only 214 kcal/lb., but because mare's milk (qumis) becomes alcoholic with fermentation.9 Not very alcoholic, however: ranging from 3.25% down to 1,65. Since, as Plano Carpini noted, "Drunkenness is considered an honorable thing by [the Mongols],"10 they had to develop high-volume drinking habits and customs to offset its weakness. Plano Carpini again: "They drink mare's milk in very great quantities if they have it...."11 And Rubruck amplifies this: "In summer they do not bother about anything except [qumis]....When the master begins to drink, then one of the attendants cries out in a loud voice 'Ha!' and [a] musician strikes his instrument. And when it is a big feast they are holding, they all clap their hands and also dance to the sound of the instrument, the men before the master and the women before the mistress. After the master has drunk, then the attendant cries out as before and the instrument-player breaks off. Then they drink all around, the men and the women, and sometimes vie with each other in drinking in a really disgusting and gluttonous manner.

"When they want to incite anyone to drink they seize him by the ears and pull them vigorously to make his gullet open, and they clap and dance in front of him."12

However, the pre-imperial Mongols were probably largely spared the perils of drink. Mare's milk is generally available only in summer, as Rubruck sugggests, during three to five months' of the mares' lactation period, and most of it is imbibed at that time. To live exclusively off of qumis, at, say, 2000 calories a day, at least nine pints per person would have been needed: that is, the daily milk production of two mares (above and beyond the needs of their foals). Two mares would have been about as many as an ordinary family would have kept.13 They would have sufficed to enable the man of the family to devote himself to qumis during the five milking months of spring and summer. (Observers of Inner Asian nomads have commonly remarked that the men have nothing to do in peacetime.)

So far, then, we have the early Mongols on a high-fat and high-cholesterol diet, somewhat checked by food shortages; and with a penchant for drunkenness offset by the limited supply and low alcoholic content of the only available beverage.

Then came the imperial period. Over the course of about two generations (1206-1279), the Mongols conquered a good part of the known world, including China, Russia and much of the Middle East. From the proceeds of this empire, the Mongol rulers then tried to make more food available to their nomad subjects, and to provide for themselves all they could eat of their favorite dishes, plus a staggering supply of intoxicating drink.

At first, the Mongols relied on commerce to enlarge food supplies. This was very expensive. In 1221 a Chinese traveler encountered a caravan bringing food to Mongolia and reported that "Eight catties [about 100 lbs at 1 1/3 lbs/cattie] of flour here [in western Mongolia] cost as much as fifty pounds of silver, for it is brought on the backs of camels from beyond the [Tien] Shan, some two thousand li [one li is 1364 ft--about a quarter-mile] away by foreign traders from the Western lands."14 By 1234, such prices apparently came to be considered excessive even by Ögödei Qa'an, Chinggis' spendthrift successor, since he established a state program to supplement the food-supply of Outer Mongolia. "[H]e had issued [an edict, (yasa)] to the effect that every day five hundred wagons fully loaded with food and drink should arrive [in Qara-Qorum in central Outer Mongolia] from the [Mongols' Chinese] provinces to be placed in stores and then dispensed therefrom. For [grain] and [wine] there were provided great wagons drawn by six oxen each."15

To calculate the amount of food supplied, we need to estimate the size of the wagon-load.16 Among the Inner Asian vehicles described by Pegolotti, a Florentine trader with knowledge of commerce to East Asia, are a wagon drawn by one ox carrying about 1000 lbs (10 Genoese cantaras), and a three-camel wagon with a load of about 3000 lbs (30 cantaras); Pegolotti does not mention a six-ox wagon.17 If three camels could pull 3000 lbs, it seems to me that six oxen might draw at least 4000 lbs, in which case the supply of food and drink to Qaraqorum could have amounted to 1000 tons a day. If the loads were two-thirds food and one-third drink, most of the people of Outer Mongolia could have received each day two pounds of food (probably grain, flour or pasta), and a pint of drink a day.18

For their own consumption, the Mongol leaders arranged lavish supplies of horsemeat and qumis. Ibn Baððñða, who visited the Golden Horde in Russia in the early fourteenth century, reports, for instance, that he "went one day to the audience of the [Mongol] sultan Uzbak [Özbek] during the month of Ramadan. There was served horse-flesh (this is the meat that they most often eat) and sheep's flesh, and rishta, which is a kind of macaroni cooked and supped with milk."19 On another occasion, the Mongol commander Tughluk Timur invited Ibn Baððñða to a religious ceremonial banquet: "The servants...brought in the dishes, consisting of the flesh of horses, etc., and also brought mares' milk. Afterwards they brought the buza [fermented millet], and when the meal was finished the Qur'an-readers recited with beautiful voices." After this and other religious presentations, "more food was served, and they continued in this fashion until the evening...."20

Qumis was no longer the only alcoholic drink available. Now that the conquered sedentary lands were paying tribute, much of it in kind, including drink (as we have seen in Ögödei's provisioning scheme), the imperial Mongols were also supplied with "rice mead"21 or "rice ale"22; with "honey-mead," that is, fermented honey (bal);23with a fermented millet drink (buza);24 and with a red wine "like the wine of La Rochelle," according to Rubruck.25 Most of these were surely stronger than qumis--much stronger in the case of the red wine--and they were available all year.26 Wine, rice wine, fermented honey, and distilled qumis (qara qumis) were all served at the khan's court in winter.27

The Mongol rulers served up these drinks in some style. "At Caracorum," says Rubruck, "[Möngke Khan (Chinggis' third successor, 1251-58)] has....a large which he holds his drinking festival twice in the year, once round about Easter when he passes by that way and once in the summer on his return....." (Like their Mongol subjects, the rulers were nomads; they only occupied their misnomered "capitals" part of the year.) "At the entrance to this palace, seeing that it would have been unseemly to put skins of milk and other drinks there, Master William of Paris has made for him a large silver tree, at the foot of which are four silver lions each having a pipe and all belching forth white mares' milk. Inside the trunk four pipes lead up to the top of the tree and the ends of the pipes are bent downwards and over each of them is a gilded serpent, the tail of which twines round the trunk of the tree. One of these pipes pours out wine, another [qara qumis], that is the refined milk of mares [distilled qumis], another boal [bal], which is a honey drink, and another rice mead....At the very top he fashioned an angel holding a trumpet; underneath the tree he made a crypt in which a man can be secreted, and a pipe goes up to the angel through the middle of the tree....Outside the palace there is a chamber in which the drinks are stored, and servants stand there ready to pour them out when they hear the angel sounding the trumpet.....And so when the drinks are getting low the chief butler calls out to the angel to sound his trumpet. Then, hearing this, the man who is hidden in the crypt blows the pipe going up to the angel with all his strength, and the angel, placing the trumpet to his mouth, sounds it very loudly. When the servants in the chamber hear this each one of them pours out his drink into its proper pipe, and the pipes pour them out from above and below into the [silver] basins prepared for this, and then the cup-bearers draw the drinks and carry them round the palace to the men and women."28

For similar large-scale entertainments, seating as many as 40,000 guests, according to Marco Polo,29 Qubilai Khan had "a very fine piece of furniture of great size and splendour in the form of a square chest, each side being three paces [about 8 ft] in length, elaborately carved with figures of animals finely wrought in gold. The inside is hollow and contains a huge golden vessel in the form of a pitcher with the capacity of a butt, which is filled with wine. In each corner of the chest is a vessel with the capacity of a firkin, one filled with mares' milk, one with camels' milk, and the others with other beverages....From [the chest] the wine or other precious beverage is drawn off to fill huge stoups of gold, each containing enough to satisfy eight or ten men. One of these [stoups] is set between every two men seated at the table. Each of the two has a gold cup with a handle, which he fills from the stoup. And for every pair of ladies one stoup and two cups are provided in the same way."30

Qubilai, probably like the other Mongol rulers, provided such lavish service frequently: at the New Year's festival; at the festivals for each of the thirteen lunar months; on assorted "festive occasions"; and on birthdays. The birthdays in particular must have added up to a lot of partying. Marco Polo describes only Qubilai's in detail, but remarks that "all the [Mongols] celebrate their birthdays as festivals"31--and Qubilai had four wives and 22 sons by them (along with an unspecified number of daughters), plus a number of concubines and 25 more sons by them (and surely more daughters as well). And then there were his other relatives, his great commanders, their wives, children, and so on.

At these great banquets, the guests and hosts drank steadily. Plano Carpini attended the post-election and enthronement banquets for Güyük Khan, and reported that "The chiefs held their [electoral] conference inside [a] tent....There they remained until almost mid-day and then they began to drink mare's milk and they drank until the evening, so much that it was amazing to see.....[Some days later] they placed [Güyük] on the imperial throne, and the chiefs knelt before him and after them all the people, with the exception of us who were not subject to them. Then they started drinking and, as is their custom, they drank without stopping until the evening."32

Ibn Baððñða participated in a imperial banquet of the Golden Horde, to which Mongol commanders of a Thousand and above (perhaps 187 of these, assuming that the regular, nomad army of the Golden Horde consisted of 17 tümens, each of ten Thousands33), along with religious dignitaries and distinguished guests (like Ibn Baððñða) were invited. Boiled horse-meat and mutton were served first. "After this, drinking vessels of gold and silver are brought. The beverage they make most use of is fermented liquor of honey, since, being of the Hanafite school of [Islamic] law, they hold fermented liquor to be lawful. When the sultan [Özbek Khan] wishes to drink, his daughter takes the bowl in her hand, pays homage...and then presents the bowl to him. When he has drunk she takes another bowl and presents it to the chief [wife], who drinks from it, after which she presents it to the other [wives] in their order of precedence. The sultan's heir then takes the bowl, pays homage, and presents it to his father, then, when he has drunk, presents it to the [wives] and to his sister after them, paying homage to them all. The second son then rises, takes the bowl and gives it to his brother to drink paying homage to him. Thereafter the great [commanders] rise, and each one of [the 17 of] them gives the cup to the sultan's heir and pays homage to him, after which the (other) members of the royal house rise and each one of them gives the cup to this second son, paying homage to him. The [170] lesser [commanders] then rise and give the sons of the king34 to drink. During all this (ceremony), they sing (songs resembling the) chants sung by oarsmen."35

After this feasting and drinking--a celebration of the end of Ramadan, the Muslim month of fasting--Özbek Khan was supposed to attend prayers at the mosque. "The [khan] was late in coming, and some said that he would not come because drunkenness had got the better of him, and other said that he would not fail to attend the Friday service. When it was well past the time he arrived, swaying....We then prayed the Friday prayers and the people withdrew to their residences. The sultan went back to [his great tent]" and, until the afternoon prayers, "continued as before"--drinking, presumably.36

Guests at Özbek's celebration received gifts in addition to hospitality: "To the limit of vision both right and left I saw waggons laden with skins of qumizz, and in due course the sultan ordered them to be distributed among those present. They brought one waggon to me, but I gave it to my Turkish neighbors."37 If this wagon was one of the one-ox type, it could have carried 131 gallons of qumis.

Consumption at one royal Mongol party may be quantifiable. On 24 June 1254, Möngke Khan hosted a "great drinking festival" supplied, according to Rubruck, with "a hundred and five carts laden with mare's milk, and ninety horses [to be eaten]...."38 Mongolian ponies weigh on average around 600 lbs, of which about 240 lbs is meat,39 so 90 ponies would yield about 20,000 lbs of meat. Möngke's view of rations appropriate for his guests may be estimated from his allowance for Rubruck's travelling party of four: one sheep every four days.40 This would have provided a daily ration of three pounds of mutton--5502 kcal--for each man. At three pounds of horse-meat per guest, Möngke's 90 horses would have fed about 7000 persons with 5565 kcal apiece. Assuming 1000-lb loads on the carts carrying drink,each of the 7000 would also have been served about two gallons of qumis, the approximate equivalent of 19 shots of 80-proof whiskey.41

A guest-list of 7000 is plausible: the khan's entourage consisted in large part of his Imperial Guard, the kesig, a force of 10,000 men drawn from the Mongols' best families, and Möngke probably invited them all (Qubilai did so every month, as we are told by Marco Polo) save those on guard and catering duty.42 This duty fell to the night-guards and quiver-bearers, 1000 of each, leaving as likely guests the 8000 day-guards.43

Mongol leaders drank not only at mealtime, but during business hours. Rubruck had several interviews with the Khan Möngke, and at the first he observed that Möngke "appeared to me intoxicated,"44 while at the last, the khan "drank four times, I believe," during the meeting.45

It is important to emphasize that the women of the Mongol ruling establishment drank as heavily as the men. Wives attended and drank like their husbands at the parties of Möngke and Qubilai mentioned earlier. "[S]inging and loud shouting in not considered reprehensible either in man or woman."46 "[Mongol women]...may get very drunk, yet in their intoxication they never come to words or blows."47 Rubruck and his party were entertained by one of Möngke's wives, who served them "rice ale, red wine...and [qumis]. The lady, holding a full goblet in her hand, knelt down and asked a blessing, and all the priests sang in a loud voice and she drank it all. My companion and I were also obliged to sing another time when she wanted to drink. When they were all nearly intoxicated food was brought, [mutton and carp] and of this I ate a little. In this way they passed the time until evening. Then the lady, now drunk, got into a cart, while the priests sang and howled, and she went her way."48

A generation later royal women still engaged in heavy drinking. Ghazan Khan, ruler of the Mongol Middle East (1295-1304), attempted to set limits to expenditures by princesses on purchases of clothing, animals and provisions; allowances for children, salaries for servants--and the expenses of the sharabkhana, approximately "wine-cellar."49 Specification of wine-supplies as an expense item distinct from "provisions," and the need to control its costs, suggests that the Mongol princesses, like the princes, khans and indeed the Mongols in general, as far as they were able, were still drinking their heads off.

To the moral of the story. Regular and plentiful consumption of high-calorie foods--especially horse-meat--had predictable consequences. Gout, according to Ibn Baððñða, was a common affliction among the Mongols of the Golden Horde.50 In the Far East, Qubilai suffered from it for the last 27 years of his life; he also grew to be "grotesquely fat." Nevertheless, he lived a very long life, 1215-94, for a Mongol ruler.51 Mongol men were not alone in overeating: in Rubruck's view, the Mongol women were "wondrous fat."52 Cardiovascular problems, although not then subject to diagnosis, may be suspected as well.

Heavy drinking, in turn, often led to alcoholism. The Mongols recognized this early on, but were unable to deal with it, even given the warnings and example of Chinggis Khan. Chinggis drank, but in a controlled fashion, unwilling to suffer mental confusion; he knew the symptoms and consequences of binge drinking: dulled senses, impaired physical control, clouding of the mind, and addiction leading to impoverishment. He tried to set limits on indulgence: no more than three drinking binges a month, preferably fewer and best none.53 But custom, holding drunkenness an honorable condition, won out, and, with the ready availability of strong alcoholic beverages augmenting the Mongols' high-volume drinking practice, led many Mongol rulers to drink themselves to death.

Ögödei, Chinggis Khan's successor (1229-41), "drank continuously and to excess," and eventually died of it, despite the efforts of his brother, Chaghtai, who "appointed an watch over him and not allow him to drink more than a specified number of cups...[but] he used to drink from a large cup instead of a small one, so that the [amount was large although the] number [of cups] remained the same."54 Güyük, Ögödei's successor, likewise overindulged, undermining a weak constitution and leading to an early death and a short reign, 1246-48 (perhaps saving Europe from a second Mongol invasion). Abaqa, ruler in the Middle East from 1265 to 1282, died in delirium tremens, and one of his later successors, Öljeitü (1304-18), expired, at age 35, of "digestive disorder brought on by the intemperate habits common to all the Mongol princes."55

General if less conspicuous overindulgence may also account for a decline in dynastic longevity. Chinggis and two of his sons lived reasonably long lives for the thirteenth century: Chinggis for at least 60 years (perhaps having acquired a taste for vegetables after having to eat them as a child), Chaghatai for some 57 years, and Ögödei, despite his drinking, for 55 years. Tolui, however, died at only about 42.56 Of Tolui's principal sons who died of natural causes (Möngke fell fatally ill on campaign, and Ariq Böke conveniently died in awkward political circumstances), Qubilai, as mentioned, lived for 78 years, but Hülegü, founder of the ruling Middle Eastern lineage, died at 48.57 Most subsequent Middle Eastern rulers did not even approach Hülegü's unremarkable standard. Abaqa also died at 48.58 Hülegü's third successor, Arghun, apparently already concerned about poor health when only in his early thirties, began to take a longevity-medicine made of sulfur and mercury--and soon died.59 Ghazan lived to only 32, Öljeitü to 35, and Abñ Sa‘Ìd, the last real Middle Eastern sovereign, to 30.60 The successors of Qubilai likewise declined rapidly in lifespan. Timur died at age 42, Qaishan at 31, Ayurbarwada Buyantu at 35, Yesün Timur at 35, Tugh Timur at 28, Irinjibal Qutuqtu at 7, and Toghon Timur at about 50.61

Male alcoholism complemented by heavy drinking on the part of Mongol women may have compromised fertility as well as longevity. Just as Qubilai's long life of 78 years shows what might have been, so does his procreativity. He had, as mentioned, 47 sons (and probably about as many daughters), by four wives and numerous concubines. Hülegü had 21 children by 5 wives and some concubines. Abaqa fathered 9 children by 15 women. Arghun begat 8 children, one of whom died as a child, by over 9 women. Ghazan had 7 consorts but only 2 children; one died in infancy. Of Öljeitü's 12 women, 3 had no offspring, and of his 9 children by the others, 6 died as infants. Abñ Sa‘Ìd had only one (posthumous) child by at least two wives.62 Given what we are now told about fetal alcoholism syndrome and the diminution of male fertility from binge drinking, we can perhaps understand this unimpressive record.

-----------[continue, p. 23 0f AOS version; continue, p 17, reconsidering and perhaps adding material on drink from latter part of AOS n.46]------

The corollary of short lifespans was short reigns, as mentioned at the outset. Short reigns, in turn, made for frequent successions, which were often disruptive, since Inner Asian tribal peoples had no firm rules or principles governing the the transmission of chiefly authority. Chinggis had insisted that successions should be settled at a leadership conference (quriltai), but even his decree did not hold for long. After the death of Möngke in 1259--by which time the Mongols had clearly taken the measure of all their opponents and felt less compulsion to solidarity--the Mongol princes increasingly resorted to force or the threat of force in claiming chieftaincy, and increasingly this armed competition led to military stalemate and political fragmentation.

The corollary of infertility in the Middle Eastern lineage was dynastic extinction. After the death of Abñ Sa‘Ìd, no reputable descendants of Hülegü could be found to replace him, except a candidate of desperation, his sister, Sati Beg, whose brief reign was not a success; other failed figureheads included four Hülegüid nonentities, a distant nephew of Hülegü, a descendant of Chinggis' brother, Jochi-Qasar, an obscure individual awarded the great Persian royal name "Anushirvan", and finally, a second "Ghazan", known only from a few coins.63 All of these ephemeral sovereigns were puppets of Mongol generals who, between them, pulled apart the Middle Eastern realm.

Had the descendants of Chinggis spent less time at the table, they might have lasted longer on the throne, and produced more stable, more capable, and even farther-flung government. But in enjoying too thoroughly the pleasures enabled by the vast empire they had seized, their even greater original intention, to conquer the world, became once again only the subject of drunken boasting, as it had been among Inner Asian nomads during the millenium before the coming of Chinggis Khan.64*


1 A.K.S. Lambton, Continuity and Change in Medieval Persia (Albany: State University of New York Press for Bibliotheca Persica of the Persian Heritage Foundation, 1988), p. 296.

2 L. Krader, "Ecology of Central Asian Pastoralism", Southwestern Journal of Anthropology, 11/4 (1955), 301-326; p. 309, profiled the (Outer) Mongolian domestic animal population as 55% sheep, 22% goats, 9% bovines, 10% horses, and 4% camels (I have rounded off his percentages). Ten sheep or goats may be taken as equivalent to one camel, and one horse or cow equal to five sheep or goats. The average herd suggested by H. H. Vreeland, Mongol Community and Kinship Structure, Third ed. (New Haven: HRAF Press, 1962), 31-32, for the Narobanchin Temple community had 193 sheep, and about half of the families owned between 200 and 300 sheep; most families kept rather few cows and horses.

3 P. Buell, in "Mongol Empire and Turkicization: The Evidence of Food and Foodways", in R. Amitai-Preiss and D. O. Morgan eds., The Mongol Empire and Its Legacy (Leiden: Brill, 1999), 200-223, pp. 206-208, considers game animals and vegetables to have provided important additional foods. I doubt that they were reliable, large-scale sources, as I have attempted to quantify in the case of marmots in "Mongol Campaign Rations: Milk, Marmots, and Blood?" Turks, Hungarians and Kipchaks: A Festschrift in Honor of Tibor Halasi-Kun, vol. 8 (1984) of the Journal of Turkish Studies, 223-228.

4 John of Plano Carpini, History of the Mongols, in Christopher Dawson ed., The Mongol Mission (London: Sheed and Ward, 1955), 16-17.

5 Plano Carpini, 17: "[The Mongols] do not have wine, ale or mead unless it is sent or given to them by other nations."

6 On horse-meat, see W. Martin-Rosset et al., "Rendement et composition des carcasses du poulain de boucherie," Bulletin Technique, Centre des Recherches Zootechniques et V¾t¾rinaires de Theix, 41 (Beaumont, 1980); and [no author given] Nutrient Requirements of Domestic Animals, Number 8, Nutrient Requirements of Dogs (Washington: National Academy of Sciences, 1974), 46.

On milk and meat from Mongolian cattle, see H. Epstein, Domestic Animals of China (Farnham Royal, 1969), 2-3; I. Kh. Ovdiyenko, Economic-Geographical Sketch of the Mongolian People5s Republic (Bloomington: Mongolia Society Occasional Papers, No.3, 1965), 65; G. Dahl and Hjort, Having Herds, (Stockholm, 1976), 25 and 170. Figures for the caloric value of cow's milk vary considerably from region to region and breed to breed.

On sheep, see Dahl and Hjort, 216 on milk (given a range of 1050-1200 kcal/kg, from African rather than Inner Asian sources in this case; cf. S. K. Kon, Milk and Milk Products in Human Nutrition, 2nd rev. ed. [Rome: FAO, UN, 1972], 3); and, for meat, Epstein, 34 and Dahl and Hjort, 201 and 204.

7 William of Rubruck, The Journey of William of Rubruck, in Christopher Dawson ed., The Mongol Mission (London: Sheed and Ward, 1955), 99. (Citations of Rubruck below will continue to refer to the translation in Dawson; the preferable translation by P. Jackson, in P. Jackson and D. Morgan ed., The Mission of Friar William of Rubruck, P. Jackson trans; (London: Hakluyt Society, 1990), was not available to me when preparing this paper.)

8 The Secret History of the Mongols, sect 74. The plant names are from P. Buell, "Pleasing the Palate of the Qan: Changing Foodways of the Imperial Mongols," Mongolian Studies, XIII (1990), 57-81; p. 60.

9 For the caloric value of mare's milk, see Kon, 3. Fermentation works with any milk, but best on high-lactose mare's milk (over 6%, while others are under 5%). The Mongols had little choice but to acquire a taste for qumis because, as Plano Carpini observed (see note 4 above), other alcoholic beverages had to be imported. The high prices and problematic transportation of food imports are considered below.

10 Plano Carpini, 16.

11 Plano Carpini, 17.

12 Rubruck, 96-97.

13 Labor requirements markedly increase as a horse herd grows beyond the basic needs of the family for riding animals, which amount to one to three mounts per family, preferably geldings (which, in the good old days, could serve at need as war-horses); see Vreeland, 32-33, 40-42. In Chinggis' youth, his family, headed by his widowed mother, Hö'elün, had eight geldings, and apparently at least two other mounts, even after thefts of their animals by former camp-mates; see Secret History, section 90.

14 Chih-chang Li, Travels of an Alchemist, A. Waley trans. (New York: AMS Press, 1979), 71. Commercially-supplied grain remained, or, possibly as a result of shortcomings in Ögödei's public supply program, again became, expensive in Mongolia by comparison with China: five to seven times as costly in Qaraqorum as in Ta-t'ung during Qubilai's reign, according to Ch'i-ch'ing Hsiao, The Military Establishment of the Yuan Dynasty (Cambridge MA: Harvard UP, 1978), 60.

15 RashÌd al-DÌn, The Successors of Genghis Khan, J. A. Boyle trans. (New York, 1971), 62-3. Notice that Ögödei, not Chinggis, instituted this program; obtaining food for the Mongols was an afterthought of conquest, not a motive for it.

16 Other important characteristics of the supply-system may also be estimated: grain-wagons took four months for the round-trip between Ta-t'ung, the Chinese frontier city that was the starting-point for the Mongolian supply system, and Qaraqorum, according to Hsiao, pp. 59-60. This was a distance, round-trip, of some 1500 miles, thus covered at an average pace of 12.5 miles per day [mpd]--not counting time for loading, repairs, etc. Compare the 17 mpd pace of commercial travel across Inner Asia, from Tana on the Black Sea to Kanchow in China, as listed by Francesco Balducci Pegolotti, La Pratica della Mercatura, Allan Evans ed. (Cambridge MA: Mediaeval Academy of America, 1936), 21; this East-West transport probably moved faster because it used smaller wagons and faster draught animals. At this pace, the 900-odd miles between Beijing and Qara-Qorum could have been covered in 53 days; the round-trip (without allowance for delays in turn-around) in 106 days. So that 500 wagons should arrive at Qara-Qorum each day, 60,000 wagons would have been needed (without allowance for down-time), along with 360,000 oxen, at six per wagon (without allowance for replacements or relays), plus at least 60,000 teamsters (assuming a minimal one per wagon).

17 Pegolotti, 22. Assuming the Genoese cantara weighed 104.83 lbs (47.65 kg); a different value for this cantara, yielding a load of 1250 lbs (presumably for the one-ox wagon) seems to have been used by R.S. Lopez and I.W. Raymond in Medieval Trade in the Mediterranean World (New York, 1955), 353 n. 43 and 357-58.

18 Chinggis Khan's army in 1206 numbered some 135,000 men, approximately the whole adult male population of Outer Mongolia, which implies, multiplying by five, a total population of perhaps 675,000; see J. M. Smith, Jr., "Mongol Manpower and Persian Population," Journal of the Economic and Social History of the Orient, 18 (1975), 271-299; pp. 282-283. A pound of bread provides about 1000 calories. The weight of the drink-containers has not been included in the calculation.

19 Ibn Baððñða, The Travels of Ibn Baððñða, H.A.R. Gibb trans. (Cambridge UK: Cambridge UP, 1962), II, 474. Buell, "Foodways", pp. 205, 208 and 211-213, describes the new foodstuffs and recipes taken up, at least by Mongol royalty, after the attainment of empire and its resources.

20 Ibn Baððñða, 477.

21 Rubruck, 176.

22 Rubruck, 163.

23 Rubruck, 154; Ibn Baððñða, 477.

24 Ibn Baððñða, 477.

25 Rubruck, 163.

26 Paul Buell points out this important consideration in "Pleasing the Palate of the Qan: Changing Foodways of the Imperial Mongols," Mongolian Studies, XIII (1990), 57-81; p. 78 (n. 24 for p. 61).

27 Rubruck, 154. On p. 99, Rubruck characterizes qara qumis as "a very pleasant drink and really potent," but says it was available only to the "great lords".

28 Rubruck, 175-76.

29 Marco Polo, The Travels of Marco Polo, R. Latham trans. (Harmondsworth: Penguin, 1958; reprint of 1980 [n.b. pagination varies among reprints]), 137f.

30 Polo, 136. According to the American College Dictionary, a butt equals two hogsheads, each of 63 to 140 gallons; a firkin is one-fourth of a barrel, and a barrel equals 31.5 US gallons.

31 137f.

32 Plano Carpini, 62-63.

33 Smith, "Manpower", pp. 289-290.

34 "Kings" in the translation.

35 Ibn Baððñða 495-6.

36 Ibn Baððñða 496. I have somewhat rearranged the elements of the last sentence of the quotation.

37 Ibn Baððñða, 496.

38 Rubruck, 202.

39 Epstein, 100; Martin-Rosset, "Rendement."

40 Rubruck, 206 in Dawson; clearer in Jackson's

translation, p. 253.

An alternative, less lavish scale of provisioning--and a much longer guest-list--may be estimated from Rubruck's statement, p. 98, that "[The Mongols] feed fifty or a hundred men with the flesh of a single sheep...." Mongolian sheep average 121 lbs in weight (Epstein, 34), of which about 44%, or 53 lbs, is edible, and provides 1834 kcal/lb (Dahl and Hjort, 201, 204). At the more generous of these rates, the 90 horses could have served 20,000 guests--and each of them could have had about 5 pints of qumis, assuming 1000-lb loads on the carts carrying drink.

41 Lactating mares produce only about 2 qts (4.5 lbs) of milk a day surplus to the needs of their foals--see Epstein, 101 (cf. Vreeland, 40)--so Möngke would have needed a sizeable herd of milking mares to provide the roughly 13,000 (U.S.) gallons of qumis for this party, as well as for another, similar one he held five days later, on 29 June. Consider, for example, the 3000 mares that supplied milk to Batu, the khan of the Golden Horde (Rubruck, 99).

42 Polo, 140, for Qubilai's entertainment of his kesig.

Ration-data for the week-long feasting that followed the election of Möngke at the kuriltai in 1251 are also available, in JuvainÌ, The History of the World-Conqueror, J. A. Boyle trans. (Cambridge MA, 1958), 2 vols., II, 573 and RashÌd al-DÌn/Boyle, 207. The feasting Mongols, they say, were provided each day with 2000 wagon-loads of drink: qumis and wine; and 3000 sheep and 300 cattle and horses to eat. The liquid provisions must have consisted in large part of wine, since it would have required about an enormous number of milking mares, giving about two quarts a day (surplus to their foals' needs), to provide the 1.75 million gallons of milk/qumis needed for the week's feast. The number of animals eaten, equivalent to the property of 45 or fewer families, is a reasonable provision. The quantity of rations that so many animals would have yielded is huge, around 240,000 lbs of meat a day, but proportionate to the demand that can be projected from stated or suggested figures for the troops in attendance on the principals at the feast. Batu had provided Möngke with three tümens as guards: perhaps 21,000 men by the customary seven-out-of-ten rule of thumb for calculating troop readiness. Then there were 20 princes and commanders named as attending; three are said to have brought a thousand troops with them, and probably the rest did too. And Möngke probably had his own guard tümen. Finally, if all these troops travelled with their families--say a wife and three children per soldier--the scale of the supply makes sense. By way of comparison, Qubilai, on occasion, hosted parties with more than 40,000 guests (Polo, 136).

43 Secret History, sections 226 and 232. The numbers of day-guards present might have been diminished by the absence of some on the special duties that often fell to these picked men, or the rations might have been somewhat diminished, from three pounds of meat to two and a half, and from two gallons of drink to one and two-thirds.

44 Rubruck, 155.

45 Rubruck, 195.

46 Rubruck, 167.

47 Plano Carpini, 15. Plano Carpini was much impressed by the peaceableness--among themselves--of the Mongols in general.

48 Rubruck, 163.

49 "Approximately" because among the migratory Mongols the "wine-cellar" would have been a "wine-wagon." Lambton, 293-94, translates sharabkhana as "pantry."

50 Ibn Baððñða, 489: "We went to visit...the sultan's daughter....Her husband...was present, and sat with her on the same rug. He was suffering from gout, and was unable for this reason to go about on his feet or to ride a horse, and so used to ride only in a waggon....In the same state too, I saw the amir Naghatay, who was the father of the second [wife of the sultan], and this disease is is widespread among the [Mongols]."

51 M. Rossabi, Khubilai Khan: His Life and Times (Berkeley: University of California Press, 1988), 98, 227.

52 Rubruck, 103. A portrait of Qubilai's wife, Chabui, makes her appear distinctly chubby, supporting Rubruck's appraisal; see W.C. Fong and J.C.Y. Watt, Possessing the Past (New York, Metropolitan Museum of Art, and Taipei, National Palace Museum: Abrams distrib., 1996), 265, pl. 137.

53 P. Ratchnevsky, Genghis Khan: His Life and Legacy (Oxford: Blackwell, 1991), 192, citing Yuanshi, 118, 10b; and RashÌd al-DÌn, Sbornik Letopisei (Russian translation of the Jªmi‘ al-TawªrÌkh/Collected Chronicles by Yu. P. Verkhovski [Moscow-Leningrad, 1960]), II, 47.

54 RashÌd al-DÌn, The Successors of Genghis Khan, J. A. Boyle trans. (New York, 1971), 65.

55 J. A. Boyle, "Dynastic and Political History of the •l-khªns," Ch. 4 of The Cambridge History of Iran, V, The Saljuq and Mongol Periods (Cambridge, 1968), 406.

56 Rossabi, Khubilai, 8.

57 Boyle, "•l-khªns," 354.

58 Lambton, 251.

59 Boyle, "•l-khªns," 371-372.

60 Boyle, "•l-khªns," 396: Ghazan; 406: Öljeıtü; 406, 412: Abñ Sa‘Ìd. Cf. Lambton, 251-252. Those dying violent deaths (Ahmad, Geikhatu and Baidu) and Argun, done in by his longevity-medicine (incl. sulfur and mercury), are omitted; Abñ Sa‘Ìd is included although it was rumored that he was poisoned.

61 For these ages at death, except Irinjinbal's, see R. Grousset, The Empire of the Steppes: A History of Central Asia (New Brunswick NJ: Rutgers UP, 1970), 320-321: Grousset believed that "their lives were shortened by their hedonistic excesses." For Irinjinbal, see H. H. Howorth, History of the Mongols, Part I (London: Longmans, Green, 1876; rpt Taipei: Ch'eng Wen, 1970), 310. (N.B. The ages given by Howorth vary from Grousset's in some cases, usually by only one year.) Shidebala Gegen and Qoshila, who were assassinated, are not counted.

62 Lambton, 295-296.

63 For a brief account of most of these last •l-khªns, see Boyle, 413-416. "Ghazan II" is the as-yet unpublished numismatic discovery of Stephen Album (; a gold

dinar of Ghazan II, struck at Tabriz and dated 757 hijri, has been published and illustrated in Spink & Son, Auction 27: Coins of the Islamic World, 1 June 1988, lot 360.

64 Both the Huns and the Orkhon Turks employed a rhetoric of world-conquest; see C. D. Gordon, The Age of Attila (Ann Arbor, MI: Michigan UP, 1966), 59, 93; and T. Tekin, A Grammar of Orkhon Turkic (Bloomington IN: Indiana University Publications, 1968), 261, 263, 268 and passim.

Hülegü Moves West: High Living and Heartbreak on the Road to Baghdad

Hülegü Moves West: High Living and Heartbreak on the Road to Baghdad

John Masson Smith, Jr.

University of California, Berkeley

Hülegü traveled to the Middle East at the command of his elder brother, Möngke, the newly-elected Qan of the Mongol empire, who decided, in 1253, to take care of unfinished business there. Mongol armies had established pastoral bases in Azerbaijan starting in 1229, on the orders of Ögedei, and subjugated many smaller potentates of northern Iran as well as the Cilician Armenians and Rum (Anatolian) Seljuks. But they had failed to overawe or defeat the Caliphal state in Iraq, or the Ismaili "Assassins" in their strongholds scattered across northern Iran. The Caliphate, although no longer the Islamic empire it had once been, still possessed fertile Mesopotamia, a strong army and a great, fortified capital city at Baghdad. The Mongols attacked Baghdad in the 1230s and ca. 1242, but were repulsed.1 The Assassins relied, and had for over a century, on their inaccessible mountain castles and the suicidal murderers whom they dispatched against enemy leaders.2 Möngke, wanting to deal with persistent "rebels" against his empire, now sent Hülegü to the Middle East, and their brother Qubilai ("Kubla Khan") against Song China, the other great hold-out against Mongol supremacy.

Möngke authorized both Hülegü and Qubilai to form an army by taking two men in every ten of the imperial forces, which, by this time, were extremely numerous. They included the current descendants of Chinggis' original army--call this the "Army of Mongolia"--which numbered 145,000 at the time of Chinggis' death. Chinggis' will apportioned this force among his close relatives. 101 regiments (hazaras, Thousands) were based in Mongolia (and to some extent in North China), and commanded at first by Tolui, Chinggis' youngest son by his first wife, Börte. An additional 28,000 men went other members of the imperial Family; each of Tolui's older brothers, Jochi, Chaghadai and Ögedei received four regiments. During Chinggis' lifetime, these brothers had been assigned territories as well, west of Mongolia: Jochi's west from roughly Lake Balkash to the Volga (this came to include not only most of Kazakhstan, but Russia, Ukraine, etc.: the realm of the so-called Golden Horde); Ögedei's ulus (realm) centered on the valleys of the Emil and Qobaq rivers, east of the Ala Kul (Lake) and included the area beyond (north of) the Tarbagatai range; and Chaghadai held the Ili valley, Almalyk and Transoxiana.3 Thus, some 133,000 troops of the Army of Mongolia were stationed in Mongolia or to its east and south and only 12,000 from Mongolia west to Anatolia and Ukraine.

These westerly realms, like Mongolia, supported large nomad populations, predominantly Turkic but otherwise quite like the Mongols; many of them were conscripted into the imperial army, complementing the brothers' four original regiments. Rashiduddin says that in his own time (early fourteenth century) the army of the Ulus Jochi (the Golden Horde) consisted of descendants of these four thousand Mongols, plus Russians, Qipchaq Turks, Circassians and Magyars. It probably numbered some fifteen (nomad) tümens (divisions of Ten Thousands); each region seems to have been limited to a regular nomad army of fifteen tümens, even if more manpower were available, so as not to outnumber the forces of Mongolia proper.4 Ögedei and Chaghadai surely enlarged their followings in the same way.

Hülegü also had to raise most of his army outside Mongolia. Few Mongolian troops could be spared him because Möngke intended to use most of them against China, in a campaign engaging 90 tümens--a nominal 900,000 men. Of these, some considerable part was non-Mongolian: "Jauqut"--Chinese, Tangqut, Manchurian Jurchens, and Koreans. Since 24 Mongol commanders are named (including Möngke and Qubilai), I assume that the 90 tümens included 24 Mongol (cavalry) units and 66 of Jauqut (mostly infantry).5 This suggests a doubling--or more: some forces remained in Mongolia--of the Mongolian population in about 50 years, an increase not implausible in light of Fredrik Barth's study of a nomad tribe that grew by a factor of three per generation.6 Rashiduddin says that the 101,000 men bequeathed by Chinggis to Tolui, and those assigned his other children "have multiplied and become many times the number they were originally."7 Population increase could more than account for the Mongol force used in China. But with so many from the Army of Mongolia involved in this campaign, it seems unlikely that it could have spared a fifth ("two in ten") for Hülegü.

The Journey

Möngke sent Hülegü back to his own ordu in February of 1253, the Ox Year, to prepare for the campaign. In autumn of 1254, the Leopard Year, Hülegü led out his army on a journey that was to cover about 5000 miles.8 He left his a’urughs behind (more on these below).

The long journey took what might seem a very long time. The army reached Baghdad only on 22 January 1258, about 3 years and 3 months (some 1200 days) after it had set out, thus moving at an average pace of only about 4 miles per day (mpd). Compare the progress of the Mongol army sent against West Inner Asia, Russia and Eastern Europe starting in 1236. That army reached Bulghar on the Volga from Mongolia by autumn, 1236, about 3000 miles** in 6 months (assuming a start in spring) at 16.7 mpd. This would have been about top speed for a Mongol force: ponies should only travel for about 4 hours a day to leave time for grazing, and should walk (4 mph) to avoid overexertion.9 Chinggis required his cavalry to travel with their cruppers (which help secure the saddle) and bridles removed, so that they could not “horse around” and exhaust their mounts.10 Let us examine the route and timetable, and discuss factors that slowed Hülegü’s march.

Assuming that Hülegü started in autumn, probably October as in Juvaini (despite Juvaini's mistake about the year11), I suggest that his army made its way from somewhere near Qaraqorum, across (Outer) Mongolia, through passes of the Altai and along river valleys leading to Lake Zaysan. These suggestions, and others below, are based on some guesswork, often guided by convenience. The sources seldom specify or clearly indicate the route taken, but do name significant way-stations: Almalyk, Samarqand, the Oxus river, Shiburgan, etc. I have connected these points by routes that are more or less measurable: more when detailed in guide books, and less when measured by me on a map. This procedure shows about 824 miles (1327 km) from Qaraqorum (I assume Hülegü 's ordu was in central Mongolia) to the western Mongolian border at the Bulgan Nature Preserve.12 From there, a plausible route, with pasture and water, of some 224 miles (approx. 360 km), leads to Lake Zaysan.13 Next, on to the modern town of Ayagoz, about 202 miles (ca. 325 km) from Zaysan; thereafter to Aktogay, roughly another 75 miles (ca. 120 km); then, Ayagoz to Alma-Ata/Almaty, 350 miles (565 km).14 Almaty is fairly near ancient Almalyk, where Hülegü and his force arrived in 255, presumably in spring, since he spent that summer in the mountain pastures of the region.15 This estimate yields a journey of about six months, October to “spring”—say April. The total distance is around 1675 miles (ca. 2697 km). The pace of march, on these data and assumptions, was 9.3 miles per day (mpd).

This part of the journey took place in winter, as often on Mongol campaigns. The steppe climate is fairly dry, and snow accumulations usually slight, so that the larger animals, like ponies, can dig through to grass. Winter even aided the armies, as in Russia, where the cold froze rivers that otherwise would have presented obstacles, or in the Middle East, where the winter temperatures are safer for horses, and the rains provide them better pasture and more water. And the army could keep warm, even on the frigid steppes: each soldier had several ponies, and the animals and men, traveling together in close formation, generated heat enough for comfort.16

Hülegü and the army summered in 1255 on the mountain pastures around Almalyk where the expedition’s ponies could fattened up; the Mongols took great care of their mounts, indeed of all their livestock.17 In late summer, Hülegü left his a'urughs (again, see below) at Almalyk and proceeded on to Samarqand, 772 miles (1243 km), arriving in Sha'ban 653/1255 (5 September- 4 October).18 The rate of march cannot be ascertained; there is no clear start date. They should have left by the end of July to reach Samarqand in Sha'ban at a likely pace.

Hülegü, after spending "nearly forty days," moved on from Samarqand, sometime after 3 November 1255.19 He stopped briefly at Kish (now Shahrisabz, the birthplace, later, of Tamerlane), then proceeded to the Oxus/Amu Darya river and crossed into what is now Afghanistan on 1 January 1256.20 The distance covered is about 200 miles (322 km); the pace 2.4-3.8 mpd, depending on the start date. The army then proceeded to Shiburghan (perhaps 70 mi in 10 days), where heavy snow- and hail-storms shortly set in and compelled it, contrary to plan, to camp for the winter.21

Leaving this involuntary qishlaq in spring, Hülegü led his force to Tun in the region of eastern Iran called Quhistan, where followers of the Assassins held many strongholds. He besieged Tun on 4 May 1256, and took it on 16 May.22 Ket Buqa had left Mongolia before Hülegü to pacify Quhistan in advance of Hülegü 's passage, but had not finished the job.23 Quhistan later became a favorite Mongol qishlaq.24 From Shiburghan to Tun is ca. 530 miles (some 853 km); no date is given for the departure from Shiburghan.

From Tun, the army travelled via Tus (near Mashhad), Radkan, and Khabushan (modern Quchan) to reach Bistam on 2 September 1256. The distance covered was about 697 miles (1123 km), as measured along modern highways and railroads.25 If the army left Tun on 20 May, the average pace was 6.6 mpd.26 Along the way, the army lingered at Tus for a few days, and at Radkan for "some time," to obtain supplies, and fodder for the animals, among other things, from Merv, Yazir and Dihistan, districts north of the mountains separating Iran from Turkmenistan.27 Then they spent a month at Ustuva beyond Khabushan, until the grazing was exhausted.28 Discounting the time at Ustuva, the army moved at 9.3 mpd.

From Bistam, the army, now dividing into several corps traveling separately but to a fixed timetable, approached the Assassins, setting out on 2 September 1256. Hülegü reached Maymun Diz, the Assassin Master's castle, on 7 November 1256.29 The march from Bistam had taken 67 days, covering about 400 miles (ca. 644 km) at a pace of about 6 mpd. The other corps arrived almost simultaneously; the routes taken by the Right and Left were about the same length as those of the Center; forces from the Golden Horde coming via the Caucasus joined too, probably by prearrangement at the kuriltai of 1253.30 The army surrounded the castle; the Master was summoned to yield, and refused (his officers claimed he was not there). On 13 November the Mongol attack began, and on 19 November the Assassin Master, weakened by Mongol cajolery and catapults, finally surrendered.31 He and all his people came to a bad end. The Assassins were finished in Iran, and in history, although some survived in Syria, tolerated and used for a while by the Mamluks, and later restrained by them after an agreement with the Mongols.32 After his victory, Hülegü camped for the winter of 1256—57 near Qazvin and Lammassar.33

Hülegü started for Baghdad from Qazvin in March 1257.34 The direct distance is 511 miles (823 km).35 The journey was complicated, however, by diversions. Hülegü reached Dinavar on 26 April—and then decided to go to Tabriz. He came back to Hamadan on 26 July—and returned to Tabriz. He next reached Hamadan on 21 September, and finally set out for Baghdad.36 Subtracting the time spent traveling back and forth to Tabriz, Hülegü’s 603-mile journey from Qazvin to Baghdad may be divided into two parts, an estimate for the Qazvin-Dinavar part—say 254 miles (409 km) covered between sometime in March and 26 April; if in 30 days, then 8.5 mpd (but adjust to please)—and a fairly precise measure for Hamadan-Baghdad: 349 miles (563 km) in 73 days (10 November-22 January) at 4.8 mpd.37

The distance covered between central Mongolia and Baghdad, adding up the figures given above, was 4947 miles (7966 km). The parts of the journey for which reasonable estimates of the pace may be estimated are: Qaraqorum-Almalyk, 7-8 mpd; Samarqand-Oxus, 2.4-3.8 mpd; Tun-Bistam, 7.2 mpd; Bistam-Maymun Diz, 6 mpd; Hamadan-Baghdad, about 5 mpd. (Low) average mpd: 5.5.

This time-and-motion study shows that, while the journey from Mongolia to Baghdad proceeded at an average speed of about 4 mpd, the movement involved was not continuous, but interrupted by a number of halts of some duration. When in motion, the army proceeded at a pace that averaged 5-6 mpd, where this can be calculated. This was far slower than usual for campaigning Mongols, who could, as noted at the beginning, average 15-16 mpd. At the other extreme, an ordinary Inner Asian nomadic migration, with its full panoply of livestock, sheep, goats, cattle, horses (ponies) and camels, averages only 2-3 mpd.38 Why did Hülegü move at the pace he did?


Sieges were awkward for the Mongol cavalry. They required positional warfare instead of the campaigns of movement most congenial to cavalry, especially nomad cavalry operating on steppe with grazing at the end of each march. Möngke had reserved pastures for Hülegü, but during the siege of Maymun Diz, some generals complained that "the horses are lean. Fodder has to be transported from [the regions] from Armenia to Kirman."39 Accessible grazing had clearly been used up. One might speculate that, to reduce reliance on pasture, Hülegü took fewer mounts than the five per soldier usual for a Mongol army; the Crimean Tatars (descendants of the Golden Horde), for instance, took only three.40 Fewer mounts meant more work for each, and care had to be taken not to overwork them.41 Steppe-raised, grazing-dependent equines do not grow very large; the average Mongol pony weighs around 600 lbs (cf. modern, fodder-fed riding horses at 1000-1500 lbs) and cannot therefore bear heavy burdens for long. An appropriate burden for a 600-lb pony is 102 lbs, 17% of body-weight; Mongol riders overburden their ponies, but only for a day at a time, and then give them several days off and ride others (hence the five ponies per soldier).42 If a day's march for troops with five ponies each was 16 miles (at 4 mph; then eight hours of grazing and eight of sleeping), then the length of march appropriate with three each might be around nine, reducing the work-load and burden on pasture, while increasing grazing time. This might help explain the slowness of the journey.

Another explanation involves the pace of ox-wagons. Hülegü expected to conduct sieges, and brought along Chinese artillerists--1000 "families" of them--probably by wagon. They might have walked, given the army's pace, but if their "families" were in fact whole "military households" with three adult males (one a soldier, the others supporting him) and their wives and children, they would have needed wagons.43 Two oxen could draw a cart carrying seven persons, who, if mounted would have needed three to five ponies per person. Cavalry needed ponies to function, artillerists did not. 44

The artillerists did need equipment requiring wagons. They used large engines to project a variety of heavy missiles over considerable distances. Although they constructed some of these weapons on the spot, using trees growing around Maymun Diz to build catapults, they had to bring along, for instance, saws, adzes, hammers, pulleys, ropes, javelins and even some of the major weapon-components: naphtha for incendiaries, and the oversized bow component of the "oxbow" engine would have required manufacture involving special woods given special treatment long in advance to give them the 'Cupid's bow' shape their name suggests.45 All this artillery gear required transportation by wagon. It could, in theory, have been broken down into camel-loads, but, at 300 lbs per camel-load, as against 1000-1500 lbs on a two-ox cart, or 3000 lbs on a three-camel wagon, wheeled vehicles were the obvious choice to reduce the numbers of animals.46 And ox-wagons would have slowed the army: without relays of oxen, they would have made <10 mpd.47

Because the army was possibly not taking the usual complement of ponies for riding and if so probably none for eating, and since (as we shall see) it did not bring along the sheep, goats and cattle that provide subsistence in normal nomadism, Möngke made other arrangements. "In advance of the army envoys were sent to reserve all the meadows and grasslands from Qaraqorum to the banks of the Oxus that had been calculated as lying in the path of [Hülegü 's] army and to build strong bridges across deep canals and rivers ..... From all lands for every individual one taghar of flour and one skin of wine as troop provisions were to be made ready."48 This was not to be a conventional, self-sufficient nomad campaign.

Business and Pleasure

The army made many stops on its way west that afforded opportunities, among other things, for lavish partying, which the Mongols loved.49 While the army rested at Almalyk, Orqina Qatun, the regent of the Ulus Chaghatay, treated Hülegü and his establishment to a series of banquets.50 At Samarqand, Hülegü and his entourage enjoyed some proper high living. The local administrator of “Turkestan and Transoxiana” for the Mongols, Mas‛ud Beg, had summoned all the commanders (amirs) of those regions, and erected a tent of gold brocade with a white felt exterior, in which Hülegü received them. The assembly engaged in "constant merrymaking and revelry" (Juvaini) and "constant drinking" (Rashid) for the nearly forty days they stayed there.51 Möngke had admonished Hülegü, as they parted, to "be awake and sober in all situations," but no source indicates that he departed from Mongol norms either by abstemiousness or excessive drunkenness.52

Somewhat later, after crossing the Oxus river, Hülegü and some of his men enjoyed a different entertainment. While out for a ride by the river, Hülegü noticed tigers (or lions: shir can mean either) in the woods nearby, and sent his guards to encircle them in the usual Mongol fashion (the circle was called nerge and was used in battle as well as hunting). Their ponies sensibly refused to face the tigers, so the soldiers mounted camels and managed to kill ten.53

While snowbound at Shiburghan, Hülegü and his entourage "constantly engaged in pleasure and enjoyment."54 In spring, Arghun Aqa produced an enormous tent, made of embroidered “gold-on-gold” linen cloth, and held up by a thousand tent-pegs. All the princes, officers , administrators, grandees, and regional rulers gathered in it to perform rituals and banquet on food and drink served on and in gold and silver plates and cups embellished with gems.55

Near Tus, Arghun Aqa pitched yet another fine tent for Hülegü. "For a few days they feasted and revelled " there.56 Then they moved on to Radkan, stopping "for a while in order to enjoy the scenery,"--and wine imported from Merv, Yazir and Dihistan (along with fodder for the animals).57 Further on, at Ustuva by Khabushan, the army's animals received their feast, spending a month grazing until "mountain and plain had been denuded of grass" and the army moved on.58

The army now had no rest until Maymun Diz had surrendered. Then, on the way to Alamut, which, like several other Assassin castles, still held out, Hülegü halted at Shahrak nine days for a victory feast.59 After that, having visited Alamut, which shortly yielded, and Lammasar, which did not, Hülegü camped for the winter of 1256—57 near Lammasar and then, from 13 January 1257, near Qazvin, where he celebrated the (Mongolian) New Year with a week of banqueting.60

In March of 1257, Hülegü started for Baghdad.61 No more banquets are reported for the rest of the campaign. Hülegü fell sick on the way back from Baghdad to Hamadan, and had earlier been ill while proceeding to Tun; he may have felt disinclined to engage in the copious drinking obligatory at such events.62

But banquets were not just for hard drinking and heavy eating. Some were also business meetings. Samarqand saw working parties. Amir Mas'ud Beg, "the [Mongol-appointed]lord of Turkistan and Transoxiana," and the region's amirs (commanders) had joined Hülegü. Möngke had ordered them--and all authorities along Hülegü 's intended route--to prepare supplies for the army, and to get ready to accompany Hülegü to Iran with their forces.63 Hülegü had come to collect.

Samarqand lay in the Ulus Chaghatai. To get there, as we have seen, Hülegü had proceeded through the Zaysan region, belonging to the (former) Ulus Ögedei, and along the way, in both uluses, he would have encountered in their winter-quarters (qishlaqs), and with their families, so that they could not escape, the nomads who lived in those regions (and summered in the adjacent highlands). Hülegü conscripted large numbers of them (more on this below).

At Kish, Hülegü stopped for a month and did more business. Arghun Aqa, the Mongols' administrator for Iran, joined Hülegü, along with "the grandees and nobles of Khurasan."64 They ordered the Mongols’ vassal "monarchs and sultans" in Iran to mobilize soldiers and supply weaponry and provisions for the campaign against the Assassins. Local rulers from Anatolia, Fars, Persia, Khurasan, Azerbaijan, Arran, Shirvan and Georgia obeyed.65 Many of them joined Hülegü and his entourage at Shiburghan, to banquet and perform rituals in Arghun Aqa's thousand-peg tent.66 They also must have taken care of business, as the next stages of the journey would bring the army into Iran, where the men and supplies summoned up from Kish would be awaiting further orders, which they would receive at Bistam.

We can see the growth of Hülegü's forces in the increasing number of commanders named as the army proceeds west. For example, Ket Buqa, with Köke Ilgei, started off from Mongolia ahead of Hülegü with 12,000 men; Ket Buqa was in charge, probably commanding a tümen, while Köke Ilgei led two hazaras which were to serve as cadre for future conscripts filling out a tümen.67 Hülegü followed with, presumably, a tümen (it is mentioned later). Thus, the army apparently consisted of two tümens and two hazaras at the outset. By the time the army arrived at Bistam in Iran, it had five tümens, with two more unit commanders named: Tegüder Oghul and Buqa Timur.68 And more were coming.

After taking Maymun Diz, and wintering near Qazvin, Hülegü proceeded to Hamadan in March 1257, and "began outfitting the army," which was apparently completed by mid-April, as he had moved on to Dinawar by 26 April on his way to Baghdad.69

Besides the play-and-work stops, some delays were forced by bad weather: by the blizzard at Shiburghan, and by the dangerous (especially for ponies) summer heat at Baghdad, which Hülegü apparently discovered as he neared Iraq and which caused him to meander during the summer, 26 April (hot weather arrives early in Mesopotamia)-21 September, 1257, between Dinavar, Tabriz, Hamadan, Tabriz and Hamadan, before resuming his march on Baghdad.

Finally, all these play-and-work stops helped the diverse elements of the army become familiar with one another, and their leaders to learn to work together, as they did so well when the time came.

Weaponry, Strategy and Tactics

Military operations also slowed Hülegü 's march. Since his mission targeted mountain strongholds (the Assassins' lairs) and a fortress-city (Baghdad); he had prepared for siege warfare. By this time, the Mongols had conducted many sieges during their campaigns in northern China, Central Asia, Russia and Hungary. Their Mongol (and Turkic) soldiery, all archers, could outshoot the cities' defenders, few of them skilled in archery, while Chinese, Manchurian and other conscripts provided engineers and artillerists; laborers; and "arrow-fodder" for storming fortifications--and, in the worst case, for besieging them until the defenders yielded.70 Hülegü started off with only a few of these troops, about two tümens of Mongols and the 1000 "families" of Chinese soldiers discussed above. He found more manpower en route.

The Chinese soldiers with Hülegü employed state-of-the-art siege engines. One was a catapult of the sort called trebuchet or mangonel in Europe, manjaniq or ‛arradah by the Muslims, pao in China, and orbu’ur by the Mongols.71 By the beginning of the thirteenth century, two types of these had been developed in China. Both had a long pole pivoting on an axle set on a scaffold; the axle divided the pole into a short and a long section; the long section terminated with a sling holding a missile, the short attached to a power-source that turned the pole on its axle and impelled the sling and its shot. The first, older "traction" type was powered by teams of men pulling on ropes tied to the short end. It was labor-intensive, none too powerful, and dangerous to use. A team of 250 men could propel a stone of only 90 lbs for only 33 yards; the enemy's archers could shoot accurately and pierce armor at 50 yards. The other type, the "counterweighted" catapult, resembled the traction variety, except that a weight replaced the ropes at the short end of the pole, and gravity, not muscle, impelled the missile. A crew of only 10 to 15 men could shoot such a catapult 167 yards; others ranged to180 and 233 yards, at the limit or beyond of useful archery.72 No crew sizes are given for these latter; and although no missile weights are given,* later Chinese records, contemporary illustrations, and modern replicas, show that they could reach 250 lbs.73 These weapons could thus operate in safety and hurl an irresistible missile.

The Mongols probably acquired this Chinese counterweighted catapult during their conquest of North China.74 The defenders of Lo-Yang had used them against the Mongols in 1232.75 Earlier, Chinggis had taken a whole tümen of artillerists to Central Asia in 1219-23--and they had failed to take Ashiyar castle in Gharchistan,76 probably because they still used the old-style high-manpower, short-ranged and light-missile traction catapults, no more than 40 of them if 250-man teams were drawn from the 10,000-man unit. Hülegü's siege train included only a thousand families, but they produced better results at Maymun Diz and Baghdad, probably because they could man some 65 of the low-manpower, long-ranged, heavy-missile counterweighted catapults.

Besides catapults, Chinese artillery also included over-sized crossbows, mounted on a stand or cart (that the Romans called arcuballista).77 Hülegü had some: "a kaman-i-gav ["ox's bow"], which had been constructed by Khitayan craftsmen and had a range of 2,500 [gâm], was brought to bear on those fools [the Assassins in Maymun Diz] ... and ... many soldiers were burnt by those meteoric shafts." 78 Boyle, Juvaini's translator, gives "pace" for "gâm," which can also mean foot, cubit, step or pace.79 But the only remotely plausible of these measures is the foot; a gâm of 12 inches would make the range of the "oxbow" 833 yards, about a half-mile (0.47).

No such performance is recorded. Franke documents a Chinese arcuballista with an effective range of 300 pu/paces, which he converts at 2 feet to the pu/pace into "something like 200 yards."80 Needham lists a "large winch-armed" crossbow (which he appears to think was an arcuballista) as shooting 1160 yards, a distance he acknowledges "seems credible only with difficulty."81 More likely, incredible. Compare a European winched crossbow with a 1200 lb-draw and range of (only) 460 yards.82 And consider Needham's "arm-drawn" crossbow with a range of 500 yards (300 pu) in light of the difficulty of drawing (i.e. cocking) by hand a crossbow drawing more than 150 lbs.83 The discrepancies arise from different conversions of the Chinese measure pu. Franke has it as a pace of two feet, and Needham as a "double-pace" of five feet. Using a two-foot pu, Needham's arm-drawn crossbow had a range of 200 yards, and (getting back to the main matter) his large, winch-armed weapon's range was about 460 yards. But with the double-pace dismissed and the gâm reduced from pace to foot, the oxbow still shoots nearly twice as far as any recorded arcuballista.

Juvaini's 2,500 gâm shot may have been powered by hyperbole, but rocket-assistance should also be considered. The thirteenth century saw the development of weapons using the propulsive effect of low-grade gunpowder: "ground-rats," a bamboo tube containing powder which escaped through an aperture on combustion and propelled the tube along or about on or above the ground, originally as an entertainment, eventually to frighten or injure men and horses; and "fire-lances," flame-throwers best visualized as reversed rockets attached to a lance and projecting flaming gas toward the enemy.84

The next step, according to Needham, was taken between 1150 and 1350 when this explosive force was applied to propel arrows and javelins independently of bows or crossbows.85 I imagine an intermediate step, the addition of rockets to the conventional fire-missiles shot from arcuballistas (not from hand-held bows or crossbows because of the rockets' fiery blowback). Winch-cocked siege crossbows and arcuballistae shot about 460 yards, and wholly rocket-powered Chinese missiles by the late sixteenth century flew for 400-467 yards.86 These combined powers should account for the 833-yard range of the missiles of Hülegü's oxbow. After Maymun Diz, no more is heard of the oxbow, perhaps because pre-modern rocketry performed unreliably, and any inconsistency would have produced inaccuracy in the long shots at Maymun Diz. At Baghdad, on level terrain, the range of the unassisted missiles of the arcuballistae sufficed.

As for strategy, Hülegü 's plan for the defeat of the Assassins--and of the Caliph-- was a standard Mongol method: surround the target, immobilize (and demoralize) the enemy, and employ all weapons and as much manpower as possible.87

At Bistam, Hülegü arranged the attack on Maymun Diz, the Assassins' headquarters, arraying the army in the conventional Mongol order of battle: Left Wing, Center and Right Wing. Hülegü’s tümen (here explicitly mentioned) constituted the Center; the Left included forces—probably tümens, as also in the Right—under Ket Buqa and Tegüder Oghul, a Chaghadaid prince, who presumably brought his force from that realm; the Right was made up of the units of Buqa Timur, who had come "with an army of Oyirats," and Köke Ilgei, his original two hazaras now apparently fleshed out into a tümen (Buqa Timur's force may also have needed bulking up).88 In addition to these units that were more or less directly under Hülegü’s management, three more tümens were on their way from the Golden Horde. Assuming that each named commander led a tümen, Hülegü now commanded eight tümens.

Hülegü marched via Firuzkuh, Demavend and Rayy toward the Assassins’ castle of Maymun Diz, where their Master was resident, arriving on 7 November 1256.89 The Left traveled toward the same objective via Khwar and Semnan. The Right proceeded to “Mazandaran,” 90 a region including the extensive pastoral zone where the Gurgan and Atrek rivers approach the Caspian—later a favorite and ample Mongol qishlaq91--and also the narrow zone between the Elburz mountains and the Caspian sea. Hülegü’s Right moved west by this shore route, to cross the mountains and attack Maymun Diz if possible, and at least to block the road, lest the Assassin Master obtain support or shelter from the Assassins' castle at Girdkuh. The Golden Horde units also appeared, a tümen under Quli b. Orda arrived via Khwarezm and Dihistan, and two more, led by the Batuid princes Balaghai and Tutar approached the Assassin strongholds from the east after entering the Middle East through the “Qipchaq Strait” past Darband. The enemy was encircled.92

After the Mongol troops "formed seven coils around" the castle, a circuit "nearly six leagues around," Hülegü surveyed its defenses.93 Some commanders suggested postponing the siege because of the daunting prospect (on which the Assassins were banking94) of becoming snowbound (again). It was already late fall, fodder could not be found, and grazing was apparently inadequate, as the animals were losing weight; preparations were being made to requisition flour for the troops and fodder for the animals, and to seize all animals for transportation and as rations, from all over northern Iran.95 But Maymun Diz, although well-fortified and difficult of access, appeared vulnerable.96 The Mongols' catapults could be placed within range of the defenses, probably within about 200 yards, and if the shots could reach the defenses, they could break them. Hülegü, supported by several of his generals, decided to persist, and the following day fighting began. On the second day of combat, 13 November, the "oxbows" went into action, picking off the defending Assassins as they exposed themselves.97 The catapults needed more time, as they were built (at least in part) from local trees, but with 1000 Chinese artillery specialists on the job, work went quickly; the weapons went into action only six days after the start of construction (12-17 November).98 Once the parts were ready, teams of haulers stationed at about 300-yard intervals moved them "to the top of the hill."99 From there, the catapults began to smash the Assassins' artillery and walls. Two days later, on 19 November 1256, the Assassin Master surrendered.100

Hülegü led a still larger army to Baghdad. Fifteen commanders are named, including Hülegü himself. Two generals, Chormaqan and Baiju, led tümens (presumably) from Anatolia, where they had long been stationed. Hülegü 's Right Wing included Sönitei, Balagha, Tutar, Quli, Buqa Timur, and Su'unchaq. The Left had Ket Buqa, Qudusun and Ilgei. The Center comprised Hülegü himself, Köke Ilgei, Uruqtu and Arghun Aqa; some administrators are also named: Qaraqai, Sayfuddin, Nasiruddin Tusi and ‛Alauddin ‛Ata-Malik (Juvaini, our historian), although it is doubtful whether they led troops. Hülegü 's entourage also included "all the sultans, maliks and atabegs of Iran," who probably were commanders--of "arrow-fodder."101 These forces encircled the city. The two tümens coming from Anatolia via Irbil, crossed the Tigris and approached West Baghdad; they fought, inconclusively at first, with the Caliph's field army, but finally swamped it by opening dams of a lake. Six tümens, Hülegü’s Right wing, came to the left bank of the Tigris above Baghdad, where four of them proceeded down that bank toward the city, and two crossed to join the Anatolian corps. Three tümens, forming Hülegü’s Left wing, moved up the Tigris from Khuzistan. And Hülegü, leading four tümens and the "sultans, maliks," et al., proceeded from Hamadan to Dinavar (where he left his a'urughs), Kirmanshah, Khanaqin and Ctesiphon. They reached Baghdad on 22 January 1258. 102 The attack began on the 29th, the artillery broke down the walls and towers of the city, and the Caliph surrendered on 7 February 1258.103 A massacre followed.

The Mongols relied in their siege warfare on their catapults, on overwhelming manpower--and, if necessary, on large expenditures of it.104 Because heavy losses could be expected in taking a city or fortress, the Mongols would take no chances of needing a second siege, when adequate manpower might no longer be available; the enemy soldiery and militias that had resisted their attack were therefore killed at the end.105 The Caliph too was killed.

This effectively ended the Caliphate. Members of the house of ‛Abbas served the Mamluks as puppet-Caliphs, and the Ottomans later claimed the title, but the perpetuation was only notional.


These time-and-motion studies show that Hülegü’s army did not take with them their sheep and goats--and families. Nomad wives, helped by their children, manage their families’ subsistence animals. Each family needs about 100 sheep or equivalents to supply its food, so a tümen (10,000 men) with their 10,000 families would have, in addition to its 50,000 ponies, at least a million sheep (the equivalent of 200,000 more ponies). Hülegü did not want to take them. One tümen with ponies, sheep and goats would require as much pasture, or fodder, as five tümens with ponies only. Moreover, sheep and goats cannot travel more than about 2-3 mpd and still obtain adequate nourishment; short sprints may be manageable, but not long distances for a long time. With the sheep and goats, Hülegü might have taken twice as long to get to Baghdad. And if the sheep and goats were left behind, the women and children had to stay to take care of them—and obtain their own support.

The sources say nothing directly about the familial circumstances of the ordinary soldiers, but Hülegü himself had to leave some of his wives behind in Mongolia. He had four principal wives living when the expedition departed; two came with him: Öljei Qatun, and Doquz Qatun (his widowed step-mother), “influential, and extremely domineering”—Möngke ordered Hülegü to heed her advice. The two others, Yesunjin and Qutui, stayed in Mongolia with Möngke.106

Families and flocks left behind as Mongol armies marched to war were called a’urughs. The practice was traditional in Inner Asia, and the Mongolian term frequently appears in our story. Hülegü, setting out from Mongolia in 1254, left his a’ughruqs, including Yesunjin and Qutui, behind with Möngke. Ket Buqa, who had already gone, had doubtless done likewise. More a'ughruqs, presumably those of soldiers newly-recruited in Central Asia, stayed at Almalyk. Others stopped at Hamadan and Khanaqin on the approach to Baghdad, probably those of the troops from Iran that Hülegü had impressed into the Mongol Center, as well as the establishments of his own wives, Dokuz and Öljei, who had traveled with the army.107

Whatever plans Yesunjin and Qutui, and the families of the ordinary soldiery may have made to rejoin Hülegü in Iran were spoiled by the succession-struggle following the death of Möngke in 1259 that stranded them in Mongolia. Not until 1263 was Hülegü able to summon his wives. They started, travelling with Hülegü’s son, Jumqur, who had also been left behind, and reached Samarqand, only to be further delayed when Jumqur died. They eventually reached Iran in 1268, where “Qutui Qatun was told of Hülegü Qan’s death [4 years before], and she cried her eyes out.”108

If reunion was problematic for Mongol rulers, it was still harder for many ordinary soldiers. The troops raised in Iran would have had no difficulty rejoining their families. Of the families left in Mongolia by the (nominally) 22,000 men who departed with Ket Buqa and Hülegü --say 22,000 wives and 66,000 children at the time of separation--some wives and younger children might have been able to leave for Iran when Yesunjin and Qutui did. They could have left their subsistence animals with their older children, some of which would have formed families of their own by that time; this would have speeded the journey and reduced its logistical difficulty. However, other troops in Hülegü 's army had no such possibility.

Missing Persons

When the Ögedeid prince, Qaidu, began his effort to reconstitute the Ulus Ögedei in the late 1260s, he could scarcely scrape up two or three thousand men from its former territories. Its nomads had originally constituted the army of the Ulus, but the army had been disbanded after the attempted Ögedeid coup against Möngke in 1251, seventy-seven of its top officers had been executed--probably the top officers of seven tümens (seven tümen commanders and seventy hazara commanders of a nominal 70,000 men) and their soldiers "distributed" in such a way as left almost none for Qaidu.109 Hülegü, who shortly came through the region, moving from western Mongolia to the Lake Zaysan area to Almalyk, charged to assemble an army for occupation of the Middle East, very likely assisted in the "distribution" by taking with him perhaps 50,000, re-forming and rearming them at Hamadan under five new commanders, adding them to his forces, and leading them against Baghdad.110

Heartbreak II?

Although Hülegü 's wives, and perhaps those of his Mongolian soldiers, eventually reached Iran, those of the ex-Ögedeid soldiers probably did not. They had been left in the a'urughs around Almalyk, and probably remained there. Their husbands were now part of Hülegü 's army, and part of his program, as ordered by Möngke, to occupy Iran, establish themselves in suitable pastures, and stay.111 They could not go home, and it was Mongol policy, probably for the logistical reasons discussed above, not to send their wives to soldiers permanently based abroad. For example, Möngke assigned Sali Noyan (Tatar) to Hülegü and gave him command of two tümens previously stationed in the Qunduz-Baghlan-Badakhshan region to hold the Indian frontier. Sali asked how long he would be there, and Möngke replied, "You will be there forever."112 And his men stayed there without their (original) wives: Sali's troops, and others stationed in eastern Khurasan and even western Iran came to be known as qaraunas , and, according to Marco Polo, "They are called Karaunas, that is mongrels, because they are the offspring of Indian mothers and Tartar fathers."113 The soldiers had had to find new wives.

1 Usman Juzjani, Tabaqat-i Nasiri, H.G. Raverty trans. (London: Gilbert & Rivington, 1881), two vols., II, 1117-8; Rashiduddin Fazlullah, Jami‛u’t-Tawarikh: Compendium of Chronicles [hereinafter RaD], W. M. Thackston trans. (Cambridge MA: Harvard, 1998-99), three vols., II, 397.

2 Since 1090, according to ‛Ala al-Din ‛Ata-Malik Juvaini, The History of the World-Conqueror, J. A. Boyle trans. (Cambridge MA: Harvard UP, 1958), two vols., II, 670.

3 Juvaini, I, 42-3.

4 When Möngke assigned "two men in ten" from the imperial armies to Hulegu, the Golden Horde sent him three tümens (3 x 5 = 15): Juvaini, II, 607-8.

5 RaD, II, 414-15 mentions that Taghachar, commander of the Left Wing for a time during this campaign, led "100,000 horsemen"; these would be the soldiery commanded by the eleven Mongols named by Rashid as commanders in the Left Wing.

6 F. Barth, Nomads of South Persia: The Basseri Tribe of the Khamseh Confederacy (Oslo UP, 1964), 115.

7 RaD, II, 282; see also, 279-80: the hazara (nominally a Thousand) of Müge Noyan had 4000 even in Chinggis' time, and had further "multiplied and increased" by Rashid's day.

8 The date usually given for the start of Hülegü’s march is 1253. This is consistent both with Juvaini’s information (II, 611): 19 October 1253, and Rashid’s statement ( II, 413: Ox Year). But as Rashid returns to the story ( II, 479) when taking up Hülegü’s history, the departure is restated as autumn of the Leopard (Bars) Year, that is, 1254.

** The direct (airline) mileage Ulaanbaatar-Moscow is 2,889 miles.

9 The campaign of Samuqa in northern China (1216-17) covered 14 mpd; and Ghazan's in Syria (1299-1300) moved at 15.1 mpd: H. Desmond Martin, The Rise of Chingis Khan and his Conquest of North China (Baltimore: Johns Hopkins, 1950), 191 and map of "Chingis Khan's Campaigns in China, 1209-1227"; J. M. Smith, Jr., "‛Ayn Jalut: Mamluk Success or Mongol Failure?" Harvard Journal of Asiatic Studies, 44:2 (1984), 307-45, esp. 335-7.

10 The Secret History of the Mongols, section 199.

11 Juvaini, II, 611 has 24 Sha'ban, which, for 1254 instead of 1253, would have been 10 October.

12 Lonely Planet: Mongolia (Hawthorn, Vic., Australia: Lonely Planet, 1997), 108, 174, 237.

13 I considered sending Hülegü to the upper Irtysh region, where Chinggis Qan assembled his army in preparation for the Khwarezmian campaign, but my sources give no distances for this route, and I have tried to rely as little as possible on map-and-ruler estimates (which I qualify as "approximately," "about," or label ca.).

14 Lonely Planet: Central Asia (Hawthorn, Vi., Australia: Lonely Planet, 1996), 216.

15 RaD II, 480 (the date should be 653); Juvaini, II, 612 (he is also mistaken about the date, which should be 653/1255).

16 L.J.D. Collins, "The Military Organization and Tactics of the Crimean Tatars during the Sixteenth and Seventeenth Centuries," in War, Technology and Society in the Middle East, V.J. Parry and M.E. Yapp eds. (London: Oxford UP, 1975), 265.

17 RaD, II, 480; Juvaini, II, 612. According to the Armenian prince Haithon, who had campaigned with the Mongols, "The Tartars have much more consideration for their horses than for themselves." See É. Dulaurier, "Les Mongoles d'après les historiens Arméniens," Journal Asiatique, 5th ser., 11 (1858), 172.

18 Juvaini, II, 612. D. Streatfield-James, Silk Road by Rail (Hindhead, Surrey, UK: Trailblazer, 1993), 279, 281.

19 Juvaini, II, 612; RaD, II, 480.

20 RaD, II, 480.

21 RaD, II, 480.

22 RaD, II, 482. Juvaini, II, 615, has Hülegü take Tun "at the beginning of Rabi' I," which was 29 March.

23 RaD, II, 481.

24 Smith, "Mongol Nomadism and Middle Eastern Geography: Qishlaqs and Tümens," in The Mongol Empire and Its Legacy, R. Amitai-Preiss and D.O. Morgan eds. (Leiden: Brill, 1999), 39-56, esp. 53-54.

25 Les Guides Bleus: Moyen-Orient (Paris: Hachette, 1956), 774 (Gurgan-Bistam), 871-2 (Tun-Mashhad), 896 (Mashhad-Gurgan).

26 For the likely date, see RaD, II, 482.

27 Juvaini, II, 616-7;RaD, II, 482.

28 Juvaini, II, 616-18.

29 Juvaini, II, 717.

30 The commanders who led these forces through the Caucasus, Balagai and Tutar, had been with Hülegü in Mongolia, and had set out, presumably for home and with instructions from Hülegü, at the time of Hülegü's departure. See Juvaini, II, 612.

31 Juvaini, II, 630-4.

32 The low-keyed survival of what might be called "post-Assassin" Nizari Ismailism in the regions around Alamut and in Quhistan after Hülegü's campaign is treated by Shafique N. Virani in "The Eagle Returns: Evidence of Continued Ismaili Activity at Alamut and in the South Caspian Region following the Mongol Conquests," Journal of the American Oriental Society, 123:2 (April-June 2003), 351-70.

33 Juvaini, II, 721-2.

34 RaD, II, 486.

35 Les Guides Bleus: Moyen-Orient, 711.

36 RaD,II, 488.

37 RaD, II, 493 and n. 4; Guides Bleus: Moyen-Orient, 705 (Dinavar), 706 (Hamadan).

38 For example, 200 km. in one and a half months: 2.8 mpd, for the modern Yoruk in southern Turkey. D. G. Bates, Nomads and Farmers: A Study of the Yoruk of Southeastern Turkey (Ann Arbor: Michigan UP, 1973), 5, 7.

39 RaD, II, 484.

40 Collins, "Crimean Tatars," 261; but cf. p. 267: some Tatars changed horses five times a day, and one Tatar force moved at 20 mpd: both items imply five horses per soldier, the Mongol standard.

41 A pony measures less than 14 "hands" (56 inches) at the withers--the high point of the back.

42 John of Plano Carpini, History of the Mongols, in Christopher Dawson ed., The Mongol Mission (London: Sheed and Ward, 1955; rpt as Mission to Asia), 47 on rotation; on loads: H. Epstein, Domestic Animals of China (Farnham Royal, UK, 1969): 100; D.W. Engels, Alexander the Great and the Logistics of the Macedonian Army (Berkeley: University of California Press, 1978), 128. Also, Smith, "Mongol Society and Military in the Middle East: Antecedents and Adaptations," in War & Society in the Eastern Mediterranean, 7th-15th Centuries, Y. Lev ed. (Brill: Leiden, 1997), 249-66, esp. 250-1.

43 Ch'i-ch'ing Hsiao, The Military Establishment of the Yuan Dynasty (Cambridge MA: Harvard UP, 1978), 18-19. The Mongols counted on one soldier per nomad family, but knowing sedentary families tended to be more than nuclear, and needed more subsistence labor, conscripted only one man from Chinese three-man families. Since Hülegü's forces were intended as a permanent garrison for the Middle East, the 1000 "families" of Chinese artillerists, et al., might have been whole extended families.

44 The artillerists included "catapult men, naphtha throwers, and crossbow men": RaD, II, 478. The naphtha throwers, I believe, would have prepared incendiary missiles for use with the catapults and crossbows, which latter would have been the large "ox's bows" mentioned by Juvaini, II, 631, rather than weapons for individual soldiers, since the Mongols' conventional archery was more than sufficient: faster shooting and accurate to greater range.

45 Juvaini, II, 631.

46 Engels, 14 and 14 n.11 for oxen and pack-camels. Pegolotti's fourteenth century commercial manual states that three-camel wagons traveled between Urgench and Otrar (ca, 650 miles) in 35-40 days (16-18 mpd; probably with relays of camels) drawing 3000 lbs; his text on Silk Road trade in the Mongol period in Francesco Balducci Pegolotti, La Pratica della Mercatura, Allan Evans ed. (Cambridge MA: Mediaeval Academy of America, 1936) is translated in R. S. Lopez and I. W. Raymond, Medieval Trade in the Mediterranean World (New York, 1955), 355-58.

47 Draught oxen move at 2 mph and work a 5-hour day, according to Engels, 15 (but cf. Epstein, 3: 30 km per day). Using relays of animals, ox-wagons delivering supplies to Outer Mongolia under a program initiated by Ögedei averaged 12.5 mpd,; Western traders traveling to China in the fourteenth century averaged 17 mpd, using horses and camels for the most part instead of oxen. See Smith, "Dietary Decadence and Dynastic Decline in the Mongol Empire," Journal of Asian History, 34/1 (2000), 35-52, and Hsiao, 59-60.

48 RaD II, 478. See also Juvaini, II, 609-10. Ibn Battuta, traveled with the entourage of Bayalun Qatun, the Byzantine wife of Özbek, the Golden Horde ruler, as she journeyed to Constantinople, and "At every halting-place in this land there was brought to the khatun a hospitality-gift of horses, sheep, cattle, dugi, qumizz, and cows' and sheep's milk." See The Travels of Ibn Battuta, H.A.R. Gibb trans. (Cambridge UP, 1962), II, 498-9.

49 Smith, "Dietary Decadence," 35-52.

50 RaD, II, 480; Juvaini, II, 612.

51 RaD, II, 480; Juvaini, II, 612.

52 RaD, II, 479.

53 RaD, II, 480; Juvaini, II, 613-4.

54 RaD, II, 480; see also Juvaini, II, 614.

55 RaD, II, 480-1.

56 Juvaini, II, 616.

57 RaD, II, 482. Juvaini, II, 616-7.

58 Juvaini, II, 617.

59 Juvaini, II, 718.

60 RaD, II, 485; Juvaini, II, 718-9.

61 RaD, II, 486.

62 RaD, II, (Tun): 482; (Hamadan): 500.

63 RaD, II, 479

64 RaD, II, 480. Mongol Khurasan included not only the modern province, but parts of Turkmenistan and much of Afghanistan. See Smith, "Qishlaqs," 53, map 3.

65 RaD, II, 480.

66 RaD, II, 480-1.

67 RaD, II, 481.

68 RaD, II, 483; Juvaini, II, 618

69 RaD, II, 487-8.

70 On Mongol archery, Smith, "'Ayn Jalut," 315-6. Worst case: after the fall of Maymun Diz, the Mongols besieged Assassins in Girdkuh for fifteen years (RaD, III, 535-6). Mongols were also prepared to participate in sieges: every Mongol soldier's equipment included " ropes for hauling [shooting] engines of war": Plano Carpini, 33.

71 Thomas T. Allsen, "The Circulation of Military Technology in the Mongolian Empire," Warfare in Inner Asian History (500-1800) (Leiden: Brill, 2002), 265-93, esp pp. 267-9.

72 Herbert Franke, "Siege and Defense of Towns in Medieval China," in Chinese Ways in Warfare, F.A Kierman, Jr. and J.K. Fairbank eds. (Cambridge MA: Harvard UP, 1974), 167-69.

73 J. Needham, Science and Civilization in China, section 30, vol. 5, part 6: Military Technology: Missiles and Sieges (Cambridge UP, 1994), 217. Also "Secrets of Lost Empires: Medieval Siege" (Nova video, available from WGBH Boston). Modern craftsmen using medieval techniques built two large catapults, one very like the Mongol catapult depicted in Rashid al-Din (D. Talbot Rice and B. Gray, The Illustrations to the 'World History' of Rashid al-Din [Edinburgh UP, 1976], ), 52-3, 146-7, 156-7; also The Legacy of Genghis Khan: Courtly Art and Culture in Western Asia, 1256-1353, L. Komaroff and Stefano Carboni eds. [New York: Metropolitan Museum; New Haven: Yale UP, 2002], p. 36, fig 33, cat. no. 24), both of which, from 200 yards, hit and broke a wall copied from a Welsh fortress of Edward I. The counterweights, one fixed and one--the Mongol type--pivoting, weighed 6 and 6.5 tons. Construction took forty craftsmen about two weeks (Hülegü's larger work force worked much faster) after obtaining the materials, including poles fashioned from single tree-trunks (weighing one ton on the fixed-counterweight engine). The poles of the Mongol catapults depicted in RaD/ Rice & Gray, 52-3 and Legacy, fig. 33, unlike the replicas', were composites of several shafts bound together, simplifying procurement and fabrication.

* [Additional, post-publication note] Douglas S. Benson, The Mongol Campaigns in Asia [and Europe] (Chicago: Bookmasters, 1991)], 338 provides this information from the Russian Patriarchal Chronicle, Year 6748 and 10 Polnoe Sobranie Russkikh Letopisei 115-17: At the battle of Chernigov, 1240, "they [the Mongols] hurled [catapulted] at [the defenses] with stones of one and a half shotweight: stones requiring four strong men to lift." [stones of ca 200 lbs?]

74 The common impression that the Mongols adopted the counterweighted catapult in the Middle East (see Thomas T. Allsen, "The Circulation of Military Technology in the Mongolian Empire," Warfare in Inner Asian History (500-1800) (Leiden: Brill, 2002), 265-93, esp pp. 267-9; also Needham (1994), 218-20), derives from Qubilai's importation in the 1270s of Arab engineers to design catapults for his Sung China campaigns (Marco Polo falsely claimed the credit). The Arabs' contribution was most likely the metal bushings and axles used in Iranian Mongol catapults: see Rice and Gray, loc. cit. In Chinese catapults, wooden axles turned in sockets or notches in the supporting wooden scaffold. The bearings of the Iranian catapults would have turned more smoothly and suffered less wear from the tremendous forces exerted upon them (by a counterweight of perhaps 6.5 tons to swing a beam of perhaps a ton) and would therefore have shot more reliably, more consistently, more accurately, and farther than their Chinese counterparts.

75 Needham (1994), 218. Clearly a counterweighted catapult from its small crew, but still short ranged.

76 Juzjani, II, 1074-77. Ashiyar eventually fell to starvation.

77 Franke, 166-67. Their effective range, according to Franke, was about 200 yards, similar to that of the counterweighted catapult, and about twice that of a hand-drawn war-bow.

78 Juvaini/Boyle, II, 631. Boyle translates gâm as "pace." For the name "oxbow," compare the limbs of a multiple "composite crossbow" illustrated in Franke, p. 162, with the bows of ox-harness depicted in the fourteenth century paintings of Eighteen Songs of a Nomad Flute: The Story of Lady Wen-Chi (New York: Metropolitan Museum of Art, 1974), scenes 16-18.The working of the "composite crossbow" is explained in Needham (1994), 194-95, figs. 63- 64.

79 F. Steingass, Persian-English dictionary (London: Routledge & Kegan Paul, 1957), 1072.

80 Op. cit, 166.

81 Needham (1994), 176, 177.

82 W. F. Paterson, A Guide to the Crossbow (Society of Archer-Antiquaries, 1990), 31.

83 Needham (1994), 176, 217. Paterson, 40. Increasing draw-weight does not translate directly into greater range: much of the added power goes into moving the heavier limbs of the bow rather than the impulsion of the projectile.

84 J. Needham, Science and Civilization in China, section 30, vol. 5, part 7, Military Technology; The Gunpowder Epic (Cambridge UP, 1986): ground rats, 135; fire-lance, 222-23 and figs. 44-45 (pp. 224-25).

85 Needham (1986), 477.

86 Needham (1994), 176-77, Table 3 for a European siege-crossbow and a large Chinese winch-armed crossbow; (1986), 484, note e for Chinese rockets. I have use the two-foot pu to correct Needham's Chinese ranges.

87 Plano Carpini, 36-8.

88 RaD, II, 483; Juvaini, II, 607-8, 618. Supplemental troops could have been impressed as Tainal Noyan had done during Chinggis' Khwarezmian campaign; he conscripted ten thousand Türkmen to enlarge his army as he passed through these same regions ( they deserted at the first opportunity, only to be caught by Tainal and slaughtered): Juvaini, I, 90.

89 Juvaini, II, 717.

90 RaD, II, 483.

91 Smith, "Qishlaqs," 52-4.

92 RaD, II, 361; Juvaini, I, 626-27; II, 716-7.

93 RaD, II, 484: Juvaini, II, 628.

94 Juvaini, II, 619.

95 Juvaini, II, 621-2; RaD, II, 484.

96 The description, in P.Willey, The Castles of the Assassins (London: Harrap, 1963 rpt Fresno CA: Linden 2001), of other Assassin castles, Alamut (214-24) and the fortress on Mt. Nevisar Shah (238), in the vicinity of Maymun Diz, shows they were much stronger--essentially inaccessible. The Mongols were fortunate that the Master had not holed up in one of these.

97 Juvaini, II, 631.

98 Juvaini, II, 629-32.

99 Juvaini, II, 630. The map in Willey, 171 shows trails ascending the SE side of Maymun Diz to what must have been meant by "the top of the hill." Peter Willey agrees with my view that the Mongol catapults bombarded the NE corner of the fortress (personal communication).

100 Juvaini, II, 634.

101 RaD, II, 493. The nominal strength of the putative fifteen tümens was 150,000 men--plus the "arrow-fodder."

102 RaD, II, 493-495.

103 RaD, II, 495-97.

104 For instance, eight tümens out of ten were lost by Qubilai during his siege of Yauju during Möngke's war against Sung China in 1256-7. See RaD, II, 415.

105 Smith, "Demographic Considerations in Mongol Siege Warfare," Archivum Ottomanicum, XIII (1993-1994).

106 RaD, II, 471-2, 479-80.

107 RaD, II, 473, 479-80, 493, 495, 500. The forces sent by the Golden Horde must have left their a'ughruqs at home too; this might help explain their reluctance to remain subordinate to Hülegü.

108 RaD, III, 519-20.

109 "Distributed" is the word used by J. A. Boyle in his (partial) translation of Rashiduddin, The Successors of Genghis Khan (New York: Columbia UP, 1971), 23. "Dispersed," is Thackston's translation: RaD, II, 306. For the 77 executed officers, see RaD, II, 407

110 Rearming: RaD, II, 487: " Hülegü ...camped in the plains of Hamadan... [and] began outfitting the army." New Commanders: RaD, II, 493: Sönitei, Su'unchaq, Qudusun, Uruqtu, Arghun Aqa; except for Arghun Aqa, these generals had not been mentioned previously in Rashid's account of Hülegü's expedition.

111 RaD, II 479:"From the River Oxus to the farthest reaches of the land of Egypt [said Möngke to Hülegü] .... Conquer the realm of the rebellious through the might of the great god so that your summer and winter pastures may be many ..... [Möngke] imagined that Hülegü would always remain in the realm of Iran ...."

112 RaD, I, 49.

113 Marco Polo, The Travels (Harmondsworth UK: Penguin, rpt 1980), 64-5 [N.B. Pagination varies in these reprints].