With the two great Mongol-Chinese-Korean fleets attempting an invasion of Japan, turned back by the brave resistance of the samurai, who were for once fighting a foreign enemy instead of each other. But the resistance to the Mongols, although successful, ended with a vast expenditure of resources and no real means of rewarding the participants with booty or confiscated lands.


It was not merely a wind that had smashed apart the Mongol armadas: it was a Divine Wind, a Kamikaze, demonstrating that Japan was the privileged “land of the gods.” This phrase, in fact, first achieved popular currency during the Kamakura period (1185–1333), where it began a genealogy of the emperors that was aimed at sorting out the succession issues which would characterize the end of the era.

Although the Mongol invasions would create ultimately fatal instabilities, and there were several uprisings requiring military action, the Kamakura period nevertheless saw new flourishing in the arts.


Katana signed by Masamune with an inscription in gold inlay, Kamakura period, 14th century, blade length: 70.6 cm[

The Kamakura period also saw a creative peak in sword making. Japanese swords had often been appreciated as fine artifacts, but a century of actual warfare had honed requirements to extreme levels. The contact with the Mongols, in particular, had confronted swordsmiths with the fact that many samurai had previously fought largely with arrows rather than swords. Mongol armor was thicker and harder, and was met by the swordsmiths by a new trend in thinner blades with a triangular cross-section, which might be more likely to punch through tougher armor.

It was not merely a case of evolving technology. The Kamakura smiths also benefited from an aristocracy that valued weapons but no longer required them to be churned out in quite such high quantities. For a century before (and, as it would turn out, for a century afterwards) battlefield conditions required swords that were made efficiently and quickly, in anticipation of high turnover, breakages, and losses. Now the swordsmiths could afford to take it easy, to experiment with new ideas, and to take money from their patrons, who were no longer troubled warriors but relatively wealthy men of leisure commissioning new family heirlooms or impressive curios.

In the Kamakura period we see refinements to the curved blade of earlier wars; the edge was tempered, and several variant hardnesses of steel were used, folded hundreds of times. The weapon could be honed to razor sharpness while remaining flexible enough to take some punishment. Blades were tested by their ability to cut through something with the consistency and resistance of a human body; this would sometimes involve testing on them corpses or even live human subjects, with convicted criminals meeting unpleasant fates in the service of science. It was not unusual for a sword cut a man’s body from shoulder to navel in a single strike. The best were certified for cutting through several bodies at once.

Despite their demonstrable practicality, many such swords were unlikely to see real action, and instead were cherished as family heirlooms. They hence gained an entire subculture of fittings and trappings, including ornate scabbards, highly decorative hand-guards (tsuba), and exotic materials binding the hilts.

China, 1280 CE—Marco Polo saw it with his own eyes. The river Yangtze was thick with ships great and small: ocean-going traders, robust war junks, and huge numbers of shakily repurposed river boats. All were being readied for the latest great enterprise of the new emperor, the Mongol Khubilai Khan: a massive armada that would cross the sea to annihilate the defiant island kingdom of Cipangu.

Nobody in the West had ever heard of this Cipangu before. Marco Polo’s account was the first to even mention it in a European language. When he did, he drew on years of propaganda designed to fire up the conscripts of Khubilai’s navy, as well as lies and spin concocted by reluctant Korean allies.

In his intimidating correspondence with the rulers of Cipangu, Khubilai had belittled the nation as a jumped-up barbaric kingdom, uncomprehending of courtly etiquette and ignorant of the trouble it would be in if it did not submit to him. However, in his exhortations to his armies, he made sure that everybody knew how incredibly rich Cipangu was.

“They have,” enthused Marco Polo on the basis of no evidence whatsoever, “gold in abundance, because it is found there in measureless quantities.”

Khubilai Khan had already made one attempt to invade the island kingdom after a decade of increasingly antagonistic diplomatic exchanges. Not only the people of Cipangu, but also their Korean counterparts, had literally spent years lying about the distance to the islands and the likelihood of strong resistance. Embassies had been fobbed off with numerous wily excuses, and often failed to work out whom they were supposed to be addressing. The natives infamously claimed to have an emperor of equal standing to the ruler of China, but the man who sat on their throne in 1274 was only a figurehead. His father, the former emperor, had abdicated, allowing him to meddle in politics from behind the scenes. His mother supposedly had no power of her own, but was a member of the powerful Fujiwara family, obliging the new emperor to listen to the wishes of his grandfather and uncles. His wife, meanwhile, was a scion of the Minamoto family, another powerful clan with vested interests.

And yet none of this really mattered, because foreign policy and many local issues were in the hands of the emperor’s barbarian-suppressing supreme general, the shōgun, in the town of Kamakura. But the shōgun was himself a puppet of yet another group, the Hōjō clan that had secretly run the islands for many decades. His own job was delegated to a regent, the shikken, who at the time was a callow youth of twenty-three, leaning on a shadowy council of advisers. Your guess is as good as mine, and certainly as good as Khubilai’s, as to who was really in charge.

Such obfuscations were not unique to Cipangu. The Mongols were getting a similar runaround far to the south in what is now Vietnam, where any request for a direct answer would be passed around a series of grandly titled bigwigs, any one of whom might waste another couple of months by sending back a request for clarification. The land that Marco Polo knew as Cipangu was impossible to understand, beyond a distant horizon, itself at the very edge of the known Asian world, and apparently controlled by a nebulous, invisible hegemony of power brokers and alliances. This would not be the last time it was described in such terms.

Still, at least Khubilai could mention all that imaginary gold. His troops were drunk on stories of it. The armies that had pushed Mongol rule all over Asia were ready to advance on these unknown islands, hoping thereby to shut down so-called pirate bases. What happened next is one of the greatest war stories of human history.

Khubilai’s first armada, in 1274, swiftly snatched the islands of Tsushima and Iki in the 200-kilometer (124 miles) strait. The huge fleet packed into the wide sweep of Hakata Bay, which had been for centuries the gateway to the islands for any foreign shipping.

The natives were waiting for them.

The country had not seen a meaningful battle for two generations, and the members of its warrior class, the samurai, were spoiling for a fight. They had, however, distinctly odd ideas about how a battle should proceed. The Mongols and their Korean and Chinese allies watched in bafflement as a soldier clad in strange armor tied with brightly colored silks shot a noise-making “humming-bulb” arrow over their heads. It screamed in the air on the blustery November day, intended by the defenders to signal a parley and a series of small bouts between champions.

The Mongols retaliated with a volley of deadly poison arrows. They were not there to take part in some local battle ceremony. They were there to invade.

The fighting that broke out was so fierce that it cost the defenders a third of their military manpower by the end of the first day. The Mongols, however, had no way of seeing over the wall. Unsure of their ability to hold a beachhead that was all but encircled by enemy fortifications, they retreated back to their ships to await the dawn.

The natives had other plans, taking to the waters in a swarm of little boats piled high with kindling. They crept aboard the enemy ships, starting fires and knifing sailors, although many of the flames were soon put out by heavy rain.

Soon the wind whipped up even further; the natives pulled back, looking on as the elements carried on the fight for them. The mother of all storms pounded down on the armada, overturning smaller boats and threatening to bash the close-packed ships into each other. Captains ordered their crews to push their vessels into deeper waters where the swell would not be so dangerous.

But it was too late. The storm had escalated into a monstrous typhoon that dashed the invaders against the coastline and into each other, swamping them and overturning them, crushing troop transports and supply vessels. It was soon too dark to see, but the defenders on land heard the powerful crack of timbers and the screams of men and horses.

The following morning as the sun parted the clouds, Hakata Bay was carpeted in driftwood. A handful of soaked survivors washed ashore at several points along the coastline, where they were swiftly put to death. A few of the larger Chinese ships made it out and ran straight for home.

A year later, a new embassy arrived from Khubilai, presenting the shikken regent with a golden scroll that offered to make him the “king of Cipangu.” It was a gesture of reconciliation, suggesting that everybody could spare themselves further trouble if the natives just bowed to the khan and admitted that he was their overlord. The regent made his feelings plain by having all the ambassadors executed.

It was Khubilai’s retaliation—an even larger fleet—that Marco Polo saw being assembled. Thanks to modern marine archaeology, we now understand the significance of his report of 15,000 ships on the river Yangtze—a great distance from the Korea Strait, and a reflection of numerous botched and poorly organized planning decisions. Khubilai’s second fleet was packed with everything the Mongol warlords could scrape up, including condemned barges and creaking riverboats. The timbers were warped; even the nails have been shown to have a high sulfur content, suggesting that corners were cut in every possible stage of sourcing and construction. Over-loaded with horses and supplies for a long campaign, packed so closely with men that three thousand troops were dead from disease before land was even sighted, the armada set sail in two task forces, one from Korea and the other up the coast from the mouth of the Yangtze.

Fearful of being boxed in again at Hakata Bay, the Mongols dithered offshore, many of their unseaworthy vessels lashed together in a vast floating fortress. They were particularly wary of a seaborne attack, since many of their troops were timid Chinese and Korean conscripts, worried about reports of “dragons in the water” and all too ready to surrender. They had started out with three months’ supplies, two-thirds of which they had already consumed, and they had still not made permanent landfall.

Samurai boats came out to the armada in ones and twos—numbers so small that the Mongols assumed they were there to offer terms. But the boats contained suicidal platoons on one-way missions; they cut down their own masts to make boarding ramps and stormed the larger warships.

And then the real storm came: a second typhoon, even more powerful than the earlier one. Modern marine archaeologists, examining the smithereens on the sea bed, estimate it to have been a Category 3 storm—a “major hurricane” with gusts of 199 kph (123.5 mph) that whipped up waves into storm surges of up to 4 meters (13 feet). Equivalent modern storms have been seen to strip the roofs off buildings, blow away mobile homes, and uproot trees. For the close-packed boats of the Mongol armada, it was a veritable apocalypse. Some 30,000 men managed to make it ashore from the sinking boats—bedraggled, starving, and without fresh water. They were easy pickings for the samurai defenders.

The Mongols had every intention of coming back for a rematch, but they never did. An office within Khubilai’s administration was supposedly tasked with putting together a third armada, but it was underfunded and overlooked, and largely powerless after 1286. Khubilai died in 1294, and his descendants more or less ignored the indomitable islands in the east.

The great storms gained their own legendary status among the natives. Before long, certain religious cults were claiming that the no-show by a third armada was the result of their intensive prayers. The Mongols, it was now claimed, had been fought off not merely by the warrior elite, but by the combined efforts of the entire nation, and even by the very elements themselves. In sending the weather to deliver the death blow, the gods had sent a Divine Wind, or Kamikaze. Their country, the locals believed, was special; it was unique; it was blessed by its particular gods and would never know defeat.

For decades afterwards, guards on the shores watched the seas for a new attack, but none came. Meanwhile, the defense took its toll in a different, mundane way. It was not merely the loss of life; it was the colossal expense of the project, which bankrupted many local lords. When, in earlier times, the samurai had been fighting one another, there was always a loser whose lands could be confiscated as the spoils of victory. But when the enemy came from another country, they left nothing behind but their dead; there was no reward that could be bestowed. Within a generation, the Kamakura shōgunate’s hold on power had collapsed, and the samurai had turned upon each other in another civil war, this time in the name of two rival emperors.

As for Marco Polo’s Cipangu, word of it was carried back to Europe. His tall tales of great riches and fierce knights would enter popular parlance. Two centuries later, Christopher Columbus would set out in search of the Spice Islands and this legendary Cipangu, sailing west in the hope of reaching the East, and finding something entirely unexpected.


The Mongol threat to the Latin West down to 1323

In its interventions in each of these areas, the Golden Horde was active on behalf of, or exerting pressure upon, schismatic rulers, and hence constituted an additional obstacle to the ambitions of popes and Western secular princes. But to what extent did the Mongols threaten the security of Latin states themselves? The incidence of Mongol raiding during the first two decades after 1260 is difficult to gauge: as we saw, even when the Tartars are expressly mentioned this does not necessarily indicate operations by the forces of the Golden Horde. Hungary may have suffered a series of raids from the Horde after 1272, since András III’s charter of liberties (1291) tells how the Mongols and Cumans had taken advantage of the minority of his predecessor, László IV, to attack the kingdom on frequent occasions. For what it is worth, Rashīd al-Dīn’s informants around the turn of the century were likewise under the impression that Noghai had launched repeated inroads into Hungary.

These were probably only minor raids; but the mid-1280s appear to have witnessed a shift in policy, with major attacks upon both Hungary and Poland in succession, which as we saw coincided with advances in the Balkans and should possibly also be viewed against the background of the Golden Horde’s relations with other Mongol powers. In Hungary, the dispute between King László and the Church, and subsequently the upheavals caused by the Cuman revolt, furnished new opportunities for the Mongols to exploit, and in the fourteenth century it was believed that they had been summoned by elements among the defeated Cumans who had taken refuge among them. The khan may also have been provoked by King László’s action, during the campaign to suppress the Cumans, in leading his forces into what would later become Wallachia – in the sonorous words of his charter, ‘beyond the mountains, around the confines and frontiers of the Tartars, which none of our predecessors had penetrated’.

If the Mongol invasion of Hungary at the onset of Lent 1285 was not on the scale of 1241–2, it was nevertheless a major enterprise. It was led by two prominent figures – Noghai and the future khan Töle Buqa – and was accompanied by Lev Daniilovich and others from among their Rus′ satellites. Even though the figures given by German annalists smack of hyperbole, the language of Hungarian charters certainly indicates that the numbers involved were considerable. The invaders ravaged as far as the Danube and entered Pest; and László’s consort Elizabeth, from the safety of the walls of Buda, witnessed a spirited and effective sally by members of her household. The Mongols may still have been present in the kingdom in June. Although László himself headed an expedition into Transylvania from May to August, he probably did no more than harass their withdrawal; according to a contemporary letter and reports that reached Germany, it was the local troops – Saxons, Vlachs and Székely, the last fighting as light cavalry – who cut off their retreat in Transylvania and inflicted on them a serious reverse. Polish sources allege that the Mongols also suffered considerably from famine and some kind of epidemic.

In 1287 Noghai and Töle Buqa invaded Poland; according to the fourteenth-century Vita of St Kynga (Kunigunde, widow of Bolesław the Chaste), they were in the country from 6 December until early February 1288. Töle Buqa failed to take Sandomir, while Noghai headed a similarly unsuccessful attack upon Cracow, which Polish annals place around Christmas. For their spirited resistance the citizens of Cracow would later be rewarded by Leszek the Black with tax exemptions. We learn more about the campaign from a charter of László IV of Hungary, dated 1288, in which the king rewarded György, son of Szymon, for his services. While Leszek sought refuge in Hungary from the Mongols, his kinsman László had despatched a corps of Hungarian troops under György to aid the Poles. György engaged a force of about a thousand Mongols near Sandecz (now Stary Sącz), killing their commander. In February 1288 Leszek in turn expressed his gratitude by giving György a villa in Sandecz. It was perhaps in reprisal for the aid given by László that the Mongols attacked the Szepes (Zips) region of Hungary later in the year, albeit on a smaller scale; here György again distinguished himself.

During those years when the Golden Horde was prey to civil war between Töle Buqa and Noghai (1290–1) or between the latter and Toqtoʾa (1298–9), the Latin world seems to have been spared Mongol raids; but they recommenced each time the Mongols’ internal conflicts had been resolved. A diploma of András III relates how, around winter in the second year following his coronation (i.e. 1291–2), the Mongols raided the Mačva (‘Macho’) region and he had despatched troops against them. This incursion, from the south, indicates that Noghai’s forces were now using Bulgaria, or perhaps Serbia, as a base to attack Hungary. A charter of 1296 refers to a recent raid on Hungary, though it furnishes scant detail and in any case could conceivably refer to the inroad of 1291–2. There is documentary evidence, lastly, of a Mongol attack on the Leles region in Zemplén (in present-day Slovakia) in 1305. As for Poland, Mongol troops are found ravaging Sandomir in 1293, doubtless profiting from Łokietek’s war with King Václav of Bohemia. After this episode, no more Mongol incursions into Poland are mentioned until Uzbek’s reign (1312–41).

Frontier conditions and mentalities

If, by the last decades of the thirteenth century, Ilkhanid diplomacy was turning the Mongols of Persia into potential allies, no such aura attached to those of the Golden Horde. In Europe fear of the Mongols was widespread, surfacing on one occasion in the most incongruous of places. During a widespread popular rising in the Utrecht region in 1274, it was apparently natural for the citizens, confronted by an unexpected and formidable attack by the rebels, to assume that the assailants were the Tartars. As late as 1330 Marino Sanudo, like some latter-day Carpini, was warning that the divisions in the Latin world would enable the Mongols to advance into France, Germany and Italy. In Eastern Europe, apprehensions of an imminent Mongol attack reverberated along a vast frontier – Latin Christendom’s longest land frontier with a pagan enemy. In 1286 the Teutonic Knights evacuated four of their Prussian strongholds on reports of the Mongols’ approach, sparked off, in all likelihood, by the preparations to invade Poland. The same invasion of Poland in 1340–1 that spawned rumours of an attack on Brandenburg also elicited an urgent appeal to the pope on the part of Prussian bishops and prompted a papal collector in western Hungary to despatch his funds to the greater security of Zagreb. The manner in which, alongside this Mongol campaign, contemporary chroniclers describe a near-simultaneous attack on Christian Spain by the Muslim ruler of Granada and his Moroccan allies and an Ottoman Turkish advance against the Greeks, suggests a general sense of crisis. Late in August 1340 Pope Benedict XII himself juxtaposed these other enemies with the Mongol threat in a letter urging the French king to make peace with Edward III.

The Hungarians alone shared an eastern and south-eastern border with the Mongol world that extended for hundreds of miles. For them, the ‘frontier’ was likely to be a yawning wilderness, from which the enemy might appear with no warning. In 1264 Pope Urban IV assigned the parish of Wynch (Felvincz, in Aranyos) to the archdeacon of Szatmár (now Satu Mare in Rumania), on the grounds that he was based ‘in the furthest part of the realm of Hungary, so that between him and the Tartars’ territory there is absolutely no human habitation’. The city of Milcov, once the centre of a bishopric, was described as ruined and devoid of Christian inhabitants in 1278. These eastern regions were bleak terrain for Latin forces. When the Hungarian King Louis (Lájos) crossed the Carpathians early in April 1352, on his way back from a campaign against the Lithuanians and Mongols, his horses had to feed on branches and for an entire week the men ate nothing but beans. According to the Franciscan János of Eger, the king’s confessor, who has left us an account of this expedition, he and a colleague were so weakened by hunger that they were unable to mount or dismount without help. The climate did not necessarily smile upon the enemy either, of course. During the Mongol retreat from Hungary in 1285, Noghai made off to the safety of his winter quarters, but Töle Buqa’s troops were decimated in the freezing cold and were reduced to eating their mounts, dogs and dead comrades. The Volynian Chronicle has him arrive back with few survivors of his original force after crossing the Carpathians.

Documents from thirteenth-century Hungary bear vivid testimony to the psychological impact of Mongol inroads. Nora Berend has drawn attention to the way in which the experience of 1241–2 had seared itself on the Hungarian collective memory, to the extent that it inaugurated a new semi-official chronology. Throughout the rest of his reign Béla IV’s chancery employed phrases like ‘at the time of the Tartar persecution’ or ‘at the pestilential advent of the Tartars’; and the simple words ‘at the time of the Tartars’ became entrenched in the language of record. During the process for the canonization of Béla’s daughter Margaret in 1276, a number of witnesses established their ages by reference to the Mongol invasion. Subsequent onslaughts intensified the sense of disruption and loss, particularly that of 1285, which passed into Hungarian historiography as ‘the second Tartar assault’ and obliged chancery scribes to devise phrases like ‘the time of the first’ (or ‘former’) ‘Tartars’ – or even, in one case, ‘the main Tartars’ – for the invasion of 1241. In Poland the emotional imprint of Mongol devastation is less clearly discernible; but it manifests itself, perhaps, in the way that annalists regularly couple the name of Duke Henry II of Lower Silesia, for some years after 1241, with the poignant formula ‘who was slain by the Tartars’.

There is no shortage of documentary evidence for the material and economic damage perpetrated by Mongol attacks after 1242. In 1296 the church of St Mary in Sandomir, which had been burned down in 1259, was still not fully rebuilt, and Pope Boniface VIII granted indulgences to anybody who assisted in the task. Pope Clement VI was told in 1343 that the abbey of St Andrew near Visegrád (in western Hungary), which had flourished before the Tartar invasions, had housed no monks for more than forty years. At some point in the later 1280s László IV remitted half the revenue due from the inhabitants of Beszterce (Bistritz), ‘in very great measure annihilated or impoverished by the devastation and burnings of the Tartars’. Whether mounting lightning raids or, as in 1285, wide-ranging campaigns of devastation, the enemy were intent on acquiring able-bodied captives in large numbers, and Hungarian charters regularly give great prominence to the liberation of some thousands of their unhappy countrymen by those notables who defeated the Mongols. Serfs whom the invaders abducted, but who subsequently escaped back to their homes without the aid of a ransom, were legally free in the duchy of Sandomir, though an appeal addressed to Pope John XXII in 1327 suggests that their lords (in this case the church of Sandomir) were but imperfectly acquainted with this custom. Arrangements were made, presumably, for the ransoming of Christian prisoners (and would have imposed an additional burden on local communities); but regrettably the only extant charter documenting such efforts, in Hungary, is an eighteenth-century forgery.

Naturally the Mongols were not the only agents of destruction and sacrilege in Hungary. Many elements within the kingdom profited from the upheavals caused by the Mongol attacks to misappropriate ecclesiastical property, so that Clement VI would complain in 1344 that more than forty Benedictine houses had been illegally occupied over the past hundred years. In 1277 the archbishop of Kalocsa recounted the bloodthirsty career of a Saxon rebel whom he accused of adopting ‘Tartar’ practices; and eight years later King László himself referred to the mutual strife of the Hungarians in the same breath as Tartar and Cuman attacks. For a Hungarian cleric writing in 1321, past decades were an era of wickedness characterized by ‘both Tartar invasion and oppression by the tyrants of the land’ – a situation that, in his view, Divine Grace combined with King Carobert’s energies had done much to ameliorate. Other churchmen were less ready to discriminate among Hungary’s various afflictions and less fulsome about the king’s role. Protesting to Pope Benedict XII in 1338 about Carobert’s erosion of ecclesiastical rights, the kingdom’s prelates spoke of the vulnerability of their churches. Their sole means of resisting the encroachments of the lay power was to produce written privileges which had in fact been destroyed by fire in the course of two Tartar invasions. There were those in Eastern Europe who had clearly learned to turn Mongol visitations to good account.


Mongol Invasions of Vietnam and Java

The Battle of Bạch Đằng (1288) during the Third Mongol invasion.

Map depicting Mongol campaign in Đại Việt in the north and Champa in the south.

In order to conquer south China, Mongols took the city of Yunnan and sought Trânsit via Dai Viet (modern North Vietnam), in 1257, to enter Guangxi and trap the southern Song in a pincer movement. Trân Thai Tông (r. 1225-1258) refused, so the Mongols invaded and burned the capital, Thang-long (Ha-noi). Lack of supplies, tropical climate, and a counterattack by Trân Hu’ng Đao drove them back to Yunnan. Further war was delayed by Khubilai Khan’s succession and the invasion of Korea in 1259 and Japan in 1274 and 1281. Trân Thanh-Tong (r. 1258-1278) and Nhăn Tông (r. 1279-1293) refused demands to come to the Yuan court, or to send sons and brothers as hostages and men as troops.

In 1281, Nhăn Tông reconsidered and sent a royal family member, Trân Di Ai, to the Yuan court, where he was named king and sent home with 1,000 troops. After he was killed, Mongols invaded Champa (present-day South Vietnam) in 1281 and 1283; long an enemy of Dai Viet, Champa forced the Mongol troops northward. Then, the Mongol leader Toghan and 500,000 troops invaded Dai Viet from China in 1284, sacking Thang-long and killing many people. Dai Viet’s military leader, Trân Hu’ng Đao, retaliated with a force of 200,000 troops, as King Nhăn Tông and the court fled to Thănh-hoa. While Toghan took the rich Hong delta, other Mongols advanced north. However, Trân Hu’ng Dao defeated the Mongol forces with 50,000 troops, made his way to the capital, and, by July, 1285, took more than 50,000 prisoners. Tradition says Toghan escaped by hiding in a bronze drum.

Khubilai was forced to abandon a planned third invasion of Japan in order to seek revenge against Dai Viet. In 1287, Toghan and 300,000 troops burned Thang-long, but fled when their supply ships were sunk. When a Mongol fleet of 500 ships entered the Bach Dang River, Trân Hu’ng Đao used a traditional strategy. Sharp poles put in the river bed at low tide sank 100 ships, and 400 others were captured when the tide ebbed again. Trân Hu’ng Đao put his troops’ backs to the river, and forced them to swear to defeat the Mongols or die in the effort. Toghan withdrew in 1288, and plans for a further invasion died when Khubilai died in 1294. Vietnam was the only place where the Mongols were defeated militarily.

During the same period, Mongols were also making incursions in other areas of Southeast Asia. Troops from Yunnan took the Kingdom of Pagan in Burma in v 1287. Mistreatment of Mongol envoys to Java in 1289 by King Kertanagara led to invasion in 1293. The king died before the invasion took place, and Crown Prince Vijaya joined with the Mongols to overthrow a usurper. However, Vijaya later betrayed his Mongol allies, and thus their invasion of ]ava ultimately failed.


While Vietnam was a Mongol tributary from 1258, Qubilai Khan’s effort to integrate Vietnam into the Yuan Empire resulted in a great defeat.

In 1225 Trân Thu –Dô (d. 1264) placed his nephew Trân Nhât Qu´ynh (posthumous title, Trân Thai Tông, r. 1225-58, d. 1277) on the throne, ending the Lỳ dynasty (1009-1225) and beginning the Trân (1225-1400). The Trân strictly separated civil and military functions and furthered the bureaucratization of administration with an examination system based on Confucian, Buddhist, and Taoist (Daoist) classics.

When the Vietnamese imprisoned Mongol envoys sent from YUNNAN to find a route to attack the Song, the Mongol general Uriyangqadai (1199-1271) and his son AJU invaded in December 1257 with 3,000 Mongols and 10,000 Yunnanese Yi tribesmen. After the Mongols routed the Vietnamese and massacred the inhabitants of the capital, Thănh Long (modern Hanoi), Thai Tôong abdicated in March 1258 in favor of his son Quang Bính (posthumous title, Trân Thănh Tông, r. 1258-78, d. 1291). Thănh Tông paid tribute to Uriyangqadai, who had quickly evacuated Vietnam to escape malaria.

After QUBILAI KHAN’s election as khan in 1260, Thănh Tông, enfeoffed as prince of Annam, sent tribute every three years and received a DARUGHACHI (overseer). By 1266, however, a standoff developed, as Thănh Tông sought to return to a loose tributary relationship while Qubilai demanded full submission to Mongol rule. The remoteness of communications through Yunnan, however, delayed armed conflict.

By winter 1278-79, with the conquest of South China, Qubilai ordered Mongol Yuan troops stationed along Vietnam’s borders. Vietnam’s new ruler, Trân Nhat Huyên (posthumous title, Trân Nhân Tông, r. 1278-93, d. 1308), resisted renewed Mongol demands for personal attendance at court but dispatched his uncle Trân Di Ai as hostage. In 1281 Qubilai tried to enthrone Di Ai as prince of Annam in place of Thânh Tông, but the plan failed miserably.

In summer 1284 Qubilai appointed his son Toghan to conquer Cham-pa, south of Vietnam. That December the Yuan general Sodu (d. 1284), defeated in Cham-pa, proposed the occupation of Vietnam as the key to pacifying all Southeast Asia, and Toghan was ordered to implement this plan. While Nhân Tông considered surrender, Prince Hu’ng Đao (1213-1300) rallied his troops, who all tattooed their arms with “Death to the Mongols.” After defeating Prince Hu’ng Đao ‘s army, Toghan, with Sodu and Li Heng and naval forces under `Umar Ba’atur, reoccupied Thânh Long in June 1285, while the Vietnamese court fled. As the Yuan troops advanced down the Hong (Red) River, however, Prince Quang Khai counterattacked at Chu’o’ng Du’o’ng, forcing Toghan to evacuate Vietnam, while Prince Hu’ng -Dao’s armies annihilated the isolated vanguard at Tăy Ket (near modern Hu’ng Yên), killing Sodu and Li Heng. The next March Qubilai enfeoffed Nhân Tông’s younger brother Trân Ích Tăc, who had defected to the Yuan, as prince of Annam, but hardship in the Yuan’s Hunan supply base aborted Qubilai’s planned invasion. Finally, in 1287 Toghan invaded with 70,000 regular troops, 21,000 tribal auxiliaries from Yunnan and Hainan, a 1,000-man vanguard under Abachi, and 500 ships under `Umar and Fan Ji. Toghan reoccupied Thănh Long, but the Vietnamese captured the Mongol supply fleet and defeated the navy at Bach- Đang River (near modern Haiphong), forcing Toghan to evacuate in March 1288. Abachi and Fan Ji died in the bloody retreat, and `Umar was captured. Qubilai angrily banished Toghan to Yangzhou for life.

While Nhăn Tông was willing to pay tribute to the Yuan, relations again foundered on the question of attendance at the Yuan court, and invasion plans continued. Qubilai’s successor, Emperor Temür (1294-1307), finally recognized the futility of these plans and released all detained envoys, settling for Vietnam’s traditional loose tributary relationship, which continued to the end of the Yuan. Prince Hu’ng Đao’s command of the resistance – became legendary in Vietnamese history.

Mongol invasion of Java


With the conquest of South China, the Mongols entered actively into the commerce, diplomacy, and wars of the South China Sea and the Indian Ocean. In 1278 QUBILAI KHAN (1260-94) appointed Mangghudai (d. 1290) of the TATARS, a general in the conquest of the Song, to handle overseas trade. As head of the Maritime Trade Supervisorate in Zaytun (modern Quanzhou), which MARCO POLO called perhaps the world’s most flourishing port, Mangghudai followed the Song regulations, levying a 10 percent tax on high-priced commodities and 6.7 percent on bulk goods. Merchants going abroad had to register their ships and declare their destinations, with deviations allowed only in emergencies. By 1293 six additional ports, from Shanghai in the north to Canton in the south, had Maritime Trade Supervisorates.

The court actively funded overseas expeditions, issuing government capital to privileged ORTOQ merchants. However, mercantilist anxieties showed in the prohibition on the export of metals, both precious and base; foreign merchants had to sell their goods for paper currency (chao). Foreign policy and prestige considerations prohibited the export of slaves and weapons and occasioned temporary embargoes against hostile states. Desire to profit by a government-carrier monopoly and vague worries over luxury exports led in 1285, 1303, and 1320 to prohibitions on all foreign trade by private domestic merchants. None lasted long, and in any case foreign merchants were never affected.

In 1278 Mangghudai’s colleague Sodu (d. 1284) of the JALAYIR dispatched edicts to 10 South Seas kingdoms, from Cham-pa in present-day south-central VIETNAM to Quilon on India’s southwest coast, demanding submission. By long-standing practice the South Seas realms were accustomed to paying nominal tribute to China, receiving investitures and gifts in return. Now, demanding that their rulers attend his court, Qubilai in January 1280 dispatched Sodu to Cham-pa and a Canton DARUGHACHI (overseer), Yang Tingbi, to Quilon. By 1286 Yang Tingbi had reached India’s Maabar and Quilon coasts several times, collecting eager professions of nominal submission of rulers from Kerala to Malaya. Cham-pa had, however, turned hostile, and in December 1282 Sodu led a maritime invasion with 5,000 men. The Yuan troops occupied the capital, Vijaya (near modern Qui Nho’n), but the king, Jaya Indravarman IV (1266-c. 90), retreated to the mountains. Stymied by this withdrawal, Sodu eventually sailed home in March 1284, just as AqTaghai (1235-90) embarked with another 5,000 men on a fruitless mission to reinforce him. Sodu’s subsequent plan to invade Cham-pa through Vietnam resulted only in his death and a 10-year quagmire for the Mongol YUAN DYNASTY. In 1293 Ighmish (fl. 1266-1311), an Uighur envoy and Fujian high official with experience in Champa (1281), Vietnam (1284), and Maabar (1287), led an expedition against Java with 20,000 men. Ighmish occupied the capital, Kediri, but was soon driven out.

Qubilai’s successor, Temür (1294-1307), abandoned the aim of conquering the South Seas and, content with only nominal tribute, received envoys from previously hostile Siam and Cambodia. The account of the Yuan’s 1296 envoy to Cambodia by the envoy Zhou Daguan (Chou Ta-kuan), is a major source on Cambodian history and society.

The Mongol IL-KHANATE in Persia also bordered the Indian Ocean. Trade between India and the Middle East passed through Hormoz, a port city theoretically tributary to the Il-Khans but effectively independent. From the reign of Abagha on (1265-81) the sea route from Zaytun to Hormoz was favored for embassies between the Il-Khans and the Yuan. Marco Polo and MUHAMMAD ABU-`ABDULLAH IBN BATTUTA left vivid accounts of the South Seas commerce and diplomacy

BIBLIOGRAPHY Kim, Trân-trong. Viet-Nam Su-Luoc. Saigon: Tan-Viet, 1964. Le, Thanh Khoi. Histoire du Vietnam: Des origines a 1858. Paris: Sudestasie, 1981. Masson, Andre. Histoire du Vietnam. Paris: Presses Universitaires de France, 1967. Janet L. Abu-Lughod, Before European Hegemony: The World System A. D. 1250-1350 (Oxford: Oxford University Press, 1989); David Bade, Khubilai Khan and the Beautiful Princess of Tumapel (Ulaanbaatar: A. Chuluunbat, 2002).

Mongol Otrar Campaign 1219-20

Khwarezmid Empire (1190–1220), on the eve of the Mongol conquests

It is sometimes said that Genghis’s reponse to the Otrar atrocity was tardy and that two years went by before he made his move. In fact his riposte was remarkably rapid. While making elaborate plans for a grand rendezvous of the majority of his forces on the upper Irtysh, he ordered Jochi and Jebe, already in Qara Khitai, to take their 30,000 troops and begin the march west immediately. The idea was that when these two reached the Ferghana valley, ‘arrow messengers’ or fast-travelling couriers from the main force converging on Otrar would be in touch with the khan’s latest orders. The orders to set out at once committed Jebe and Jochi to a gruelling trek in winter over high mountain passes, but there was no gainsaying the khan’s commands. The barrier of the Altyn-Tagh range forced travellers to take a route either north of the T’ien Shan or south of the River Tarim through the fearsome Taklamakan Desert. The rule of thumb was that trade caravans took the southerly route to avoid the worst mountain passes, but large groups of migrants, needing more water than the desert could provide, went north. With 30,000 men Jebe and Jochi had no choice but to go north, yet they do not seem to have followed the conventional route through Dzungaria; instead they veered slightly south-west and found a pass between the Pamirs and the T’ien Shan, most likely through the Altai range (the sources are anything but pellucid). It was probably the Terek-Dawan defile, an all-year pass at 13,000 feet, which later became the principal route from Qara Khitai and was used by Marco Polo.3 On the way to this pass the Mongols rode through snowstorms and snow 5–6 feet deep, their horses wrapped in yak-hides and the riders wearing double sheepskin coats. Shortage of food meant they often had to open the veins of their mounts, drink the blood then close the veins up again. Not surprisingly, many horses dropped dead from the snow, ice and blood letting; any that did were devoured instantly.

Finally the Mongols reached the fertile valley of Ferghana in spring 1219 after an exploit that easily rivals Hannibal’s crossing of the Alps. Turkestan was then the name given to the entire area stretching from modern China to the Caspian, so it is a high commendation to say that the Ferghana valley was the commercial jewel of the entire region. It produced gold, silver, turquoise, quicksilver, iron, copper, naphtha, bitumen, millstones, perfume, cloth, weapons, needles, scissors, pots, bows, quivers, dyed hides, cloaks, flax, cotton, had a vast acreage of rice fields, extensive orchards and vineyards and a thriving pastoral sector concentrating on goats, horses and mules. The Mongols could raid and seize all they needed in such a milk-and-honey land.

The news of a Mongol army in Ferghana seriously disconcerted the shah Muhammad II, who had thought that any army coming from the east would have to take a more northerly route, through the Dzungaria Gate. Here was confirmation of Jalal al-Din’s opinion that the Mongols should be opposed on the eastern frontier. But already defeatism was evident at Muhammad’s court. The majority wanted to abandon Transoxiana and retreat to Khorasan or the Ghazni area of Afghanistan and build an invincible stronghold against which the Mongol hordes would fling themselves in vain until fatally weakened. The shah provided no proper leadership but instead declared that Allah had told him to attack the Mongols; he ranted against Genghis as an idolater and complained to his entourage that the Mongols had ‘unfairly’ beaten him to the punch by invading China.6 Yet the provocation of learning that the Mongols were laying waste Ferghana was too much to bear. Muhammad assembled a large army and marched against them.

Jochi’s orders from his father were not to allow himself to be sucked into pitched battles with the shah; his role was as a diversion, to keep the Khwarezmians occupied while Genghis came through the Dzungaria Gate. The headstrong Jochi never liked obeying his father’s orders, and this occasion was no different. Jebe strongly urged that the Mongols should retreat, if necessary up into the foothills of the mountains, so as to lure the shah further away from Otrar, where Genghis intended to strike. Jochi took a perverse pleasure in overruling a superior general (Jebe), exercising his prerogative as a prince of the blood, saying that such a course of action would be arrant cowardice.

The sources differ in their accounts of the battle. One version is that the Mongols were in a poor state to receive the enemy after their exertions on the long trek and, instead of their usual guileful manoeuvres, simply charged the shah head on. Another is that the Mongols gave a textbook demonstration of their tactics – the light cavalry appearing to discharge their usual arrow cloud, with the heavy cavalry waiting to deliver the killer blow. It is even suggested that Muhammad came within an ace of being captured. At all events, night came down on a battle that was still indecisive, but with the heavily outnumbered Mongols (perhaps 25,000 to twice that number) having outpointed the enemy in every area: speed, mobility, imagination.

This was the second time Muhammad had taken a mauling, and it reinforced what was becoming an idée fixe with him – that it was always folly to engage the Mongols in open battle. Jebe and Jochi meanwhile followed the time-honoured tactic of withdrawing under cover of darkness, managing to take most of their cattle and horses with them. Muhammad’s failure to pursue has puzzled some analysts, but at least three major factors were responsible. He was unsure of the true strength of the Mongols and could not know for certain that the army he had fought was not just a vanguard, with the main army lying in ambush, waiting for him to pursue. Then, in order to campaign effectively, the shah had to raise taxes and this in turn led to open rebellion among some already disaffected towns; to deal with these insurrections Muhammad had to divert his army from pursuit of the Mongols. Thirdly, by late summer he learned that the vanguard of another Mongol army was already pouring through the Dzungaria Gate in the north. He now had his answer. The Jebe–Jochi force had been a classic diversion.

Genghis set off with the main army in May 1219, following the Orkhon and Tula Rivers. Angling south-west, he crossed the Khangai Mountains through passes ranging from 8,000 to 10,000 feet and reached the Altai Mountains by mid-July. There is much scholarly wrangling about the exact route he took thereafter (geography was not the medieval chroniclers’ strong point); he may have used the Dabistan-Daban Pass, though at least two other defiles in this area are open from May to September. He made camp on the upper Irtysh in summer 1219, to give his men and horses rest and recreation and await the advent of his allies to this rendezvous. While encamped there, the Mongols experienced a freak summer snowstorm.

To confuse the shah still further Genghis sent a small detachment (maybe 5,000 strong) on a circuitous route south to enter Turkestan by the famous Dzungaria Gate. This (on the modern China–Kazakhstan border) was already known in ancient times to Herodotus and Ptolemy and thought to be the home of Boreas, the North Wind, on account of the fierce and constant winds encountered there. Basically a small rift valley, the Dzungaria Gate is a six-mile-wide, 46-mile-long gap between the lakes Alakol and Ebi Nur, the most important mountain pass between China and Central Asia and the one gateway in a mountain wall that otherwise stretches 3,000 miles from Afghanistan to Manchuria. This is the route Muhammad would have expected Genghis to take, on the expectation that he would be marching west from a base in Qara Khitai.

Meanwhile on the upper Irtysh Genghis took stock of his position and reviewed his strategy. In his retinue were Qulan, his favourite wife, his sons Tolui, Chagatai and Ogodei, and all his important generals and advisers except for Jebe and Jochi, already engaged on the western front, and Muqali in China; the most important personality of all may have been Subedei, who acted as Genghis’s chief of staff and is usually credited with the brilliant strategy used against the shah. (The government of Mongolia had been left to Genghis’s brother Temuge.)

Quite how many troops Genghis led is a vexed question, as are all issues relating to numbers in Mongol history. Estimates range from the grotesquely impossible 800,000 to the absurdly low 80,000. The lunatic figure of 800,000 mentioned by some popular writers is implausible on a number of grounds, chiefly that this would imply also herds of 800,000 horses and 24 million sheep and goats all on the march. Much depends on what figure we assign to the total population of Mongolia, and here again estimates range from 700,000 to two million. Given that the pastoral economy of Mongolia is inelastic and therefore can support only a constant population, and given also that the population of Mongolia in 1967 was three million, there is every justification for accepting the higher figure of two million in the thirteenth century. This might give us a total military strength of 200,000 and take us close to some of the higher estimates. Yet we must remember that large numbers of troops were still waging war in China, and that some of the newly conquered regions in Genghis’s rear could not be totally counted on and needed garrisons to keep them loyal. All in all, counting allied contingents, Chinese sappers, engineers and siege experts, we might settle for a total force of 120,000 effectives, including the 30,000 under Jochi and Jebe.

The most alarming news that reached Genghis at his summer camp on the upper Irtysh was that the expected Tangut contingent would not be coming. At first the campaign of 1209–10 seemed to have borne ripe fruit, for to start with Hsi-Hsia stayed loyal. There was an open usurpation of the crown in 1211 when a new ruler, Shen-Tsung, a man in his late forties, secured the throne by a coup, but he confirmed the Mongol alliance and remained steadfast until 1217. But then he repudiated all his commitments, under the influence of the virulently anti-Mongol general Asa Gambu. Together ruler and general offered the Jin an anti-Mongol alliance to take advantage of Genghis’s absence in the west; Asa Gambu, moreover, was convinced the Mongols would lose the war against Khwarezmia.

The Jin refused, on the basis that both the Mongols and the Tangut were their sworn enemies. The Tangut had better luck with the Song, but the latter told Shenzong they could not formally commit to an alliance until 1220 at the earliest. When the Mongols officially protested at Hsi-Hsia’s perfidy, Asa Gambu replied with heavy irony that since Genghis Khan styled himself the Khan of Khans (though actually he never took this title), he scarcely needed the help of the Tangut, as Heaven was already on his side. When this reply was conveyed to Genghis, he is said by some sources to have become apoplectic with rage. He asked one of his secretaries to remind him at noon and dusk every day thenceforth that the treacherous Tangut realm still existed.

Siege of Otrar

Soon it was time to move on, to the first target, Otrar. Genghis ordered his commissariat to make the most careful and meticulous preparations for the march ahead, factoring in all known wells, waterholes and oases. Every ten horsemen had to carry three dried sheep, with the mutton salted and dried in the sun, and an iron cauldron in which to cook the meat; similar ‘slide-rule’ projections were formulated for all other food. Genghis’s itinerary next took him across the Irtysh, past Lake Zaysan, then, by way of the River Emil and the Tarbaghatai Mountains, and passing the eastern shore of Lake Balkhash, one of the world’s great inland seas, he came to an autumn rendezvous on the plain of Qayaliq south of the lake; here he was joined by Arslan of of the Qarluqs, Suqnaq-tigin the new ruler of Almaliq, and his great friend the idiqut Barchuq.

Ten thousand Uighurs, 6,000 Qarluqs and a contingent from Almaliq made a hefty reinforcement; Ongud, Khitans, Solons, Kirghiz and Kem Kemjiut are also mentioned among the recruits. The allies were all much impressed with the Chinese engineers and the heavy equipment they brought for siegecraft. At Qayaliq Genghis sent Chagatai ahead with the vanguard to build bridges to take them across the remaining rivers, making sure they could bear the weight of heavy transport wagons. Chagatai had many faults but he completed this task with supreme efficiency, building forty-eight timber bridges wide enough for two heavy carts to drive across side by side.

The army proceeded south-west, reached the Ili River and followed it down to Almaliq, the final significant stop before their destination. Passing to the north of the Lake Issyk Kul, they reached the River Chu (in today’s northern Kyrgyzstan and southern Kazakhstan), the last significant obstacle before Otrar itself. Genghis gave strict orders that from now on there should be no hunting, so as not to tire the horses; he made sure food supplies were adequate, then struck due west for Otrar. Once across the Chu they were in the realms of the shah.

In October 1219 the Mongols finally arrived outside Otrar on the banks of the Syr Darya, the mighty river known to the ancients as the Jaxartes. (Alexander the Great fought a famous battle on the Jaxartes in 329 bc and proclaimed it the northern limit of his empire.) Genghis had spent three months on the march, excluding stopovers on the Irtysh and elsewhere, and had covered over 2,500 miles. Now he decided to leave the siege of Otrar to Ogodei and Chagatai while he waited with a large reserve force in a pass at the top of the Arys valley, in the foothills of a nearby mountain range.

As always, his strategy was masterly. Knowing that the shah was based at Samarkand, he sent 5,000 men upstream along the Syr Darya to seize Banakat (near Tashkent), where the road from Samarkand reached the river and where any army coming from that direction would have to approach. He hoped to lure Muhammad into an expedition to relieve Otrar. If that happened, the 5,000 Mongols at Banakat were to leave and link up with Ogodei and Chagatai outside Otrar. Genghis’s plan was to tempt the shah to imagine that the besieging army at Otrar could be caught between two fires, between his army advancing from Samarkand and the powerful garrison in Otrar, which would then sortie and assail the Mongols in the rear as they turned round to face the new army. If that happened, Genghis hoped to destroy the Khwarezmian military power in one go, using his expertise in uniting far-flung detachments of his army with lightning speed. The shah was unaware that there was a second army in the north, lurking in the foothills, and had lost sight of Jochi. With Genghis appearing unexpectedly on the flank of the sallying garrison and Jochi in the rear of Muhammad’s army, the stage would be set for a victory that would echo down the centuries. It would be more complete than Gaugamela, Cannae, Zama or any of the great battles of history.

But the shah would not take the bait. He was confident that the huge force of defenders at Otrar could hold out, and he wanted to be able to locate Jochi and Jebe accurately before committing himself to a clear course of action. He dithered and procrastinated, a true martial Hamlet, while his son Jalal tore his hair out that his earlier advice – to oppose the Mongols at the Syr Darya – had been rejected. In fact merely by abandoning the Syr Darya to Genghis, Muhammad had lost the first round of the struggle.

There was some rationality in his decision not to endorse Jalal’s plan. Since all the cities on that river (including Otrar) were on the north bank, any army defending them would have the river at its back and nowhere to escape to if defeated. On the other hand, if he used the river as a defence, defying the Mongols to cross it in the face of strong forces on the south bank, he would have to abandon all his northern cities. Moreover, even if he was victorious on the north bank, the Mongols would retreat into the mountains, and it was too dangerous to follow such a foe into that kind of terrain. Muhammad’s strategy therefore was to place such an enormous garrison in Otrar that the Mongols, already weary after a long march, would tire themselves out trying to take it. When he deemed that the besiegers were sufficiently exhausted, Muhammad told his advisers that he would indeed order the march from Samarkand to Banakat. This, too, was not entirely irrational. The garrison at Otrar contained no fewer than 60,000 fighting men, with the cavalry and infantry stationed all round the walls.

Genghis waited patiently for two months while the siege of Otrar dragged on, but finally concluded that the shah would never be tempted into battle. He therefore left express instructions with Ogodei and Chagatai to press the siege with all their might, assisted by Barchuq and the Uighurs, and sent orders to Jochi to advance from Ferghana and conquer all cities along the north bank of the Syr Darya. Sadly for his own ambitions, Muhammad had imbibed the myth that the Mongols were hopeless at siegecraft, which his agents based on the lacklustre performance in the campaign against Hsi-Hsia in 1209–11. He had no idea that as a result of their war with the Jin the Mongols’ expertise had proceeded almost exponentially, and the well-defended fortress of Otrar held no terrors for them. Inalchuq, the governor responsible for the original atrocity, and general Qaracha, sent by the shah with 50,000 men to bolster the governor’s original 10,000 garrison, are said to have been caught completely off guard by the Mongol host appearing outside the walls, with the neighing of armoured horses and the braying of chain-armoured mules.35 Naturally, the Mongols used all kinds of tricks to exaggerate their numbers. Gradually they pounded the walls and cut off all supplies of food and water. By their fierce discipline the numerically inferior nomad army triumphed over an enemy who should have been able to resist.

Nonetheless, it took five bitter months of fighting before Otrar finally cracked, in February 1220. In January, Qaracha, foreseeing the inevitable end, tried to make his escape with a bodyguard but was captured and executed; Ogodei fully shared his father’s belief that a general should never abandon his master.37 After this debacle large numbers of the shah’s dreaded mercenaries deserted. Some civilians, tired of the privations of a five-month siege, opened a side gate and let the attackers in, but Inalchuq, after abandoning the city to the Mongols, withdrew into the citadel with 20,000 of his crack troops; many of them soon deserted and in the end he was left with just 6,000.38 It took another month for the Mongols to winkle them out. When the citadel fell Inalchuq and his diehard loyalists retreated into a central tower. The defenders fought bitterly and in the end, desperately short of firearms, were reduced to showering the attackers with tiles. The Mongols mined the tower and, when it collapsed, dug a still living Inalchuq out of the ruins. All the Turkish deserters and any other soldiers left alive were instantly slaughtered. Ogodei and Chagatai ordered the city razed to the ground; it was never rebuilt, and its ghostly ruins attested to the folly of opposing the greatest power on earth. Inalchuq was taken and held for Genghis’s pleasure whenever he should appear. He was of course executed, but the story that Genghis first tortured him by having molten silver poured into his eyes is apocryphal.

With the fall of Otrar there was now no obstacle to the systematic reduction of all the cities and towns along the Syr Darya. Jochi and Jebe decided they should split up, with Jebe striking south, intending to cross the River Zerafshan and bar any southern escape route from Samarkand whenever Genghis decided to assault it. Jebe had with him somewhere between 10,000 and 20,000 men, scarcely enough to engage a large army; nevertheless when he encountered a larger Khwarezmian force, he attacked it and put it to flight. It was a great exploit but Genghis was none too happy when he heard of it. He always tried to avoid heavy casualties and to win by mobility and other indirect means.

A Brief History of WMDs before 1900

This is a Mongol style siege but not the siege of Kaffa [Caffa] from the early 14th century Jami al-Tawarikh  (Compendium of Chronicles) by Rashid ad-Din.  Edinburgh University Library

Transmission from Kaffa. (Wheelis, 2002)

In terms of referring to nuclear, chemical-and by inference, biological-weapons, the term “weapons of mass destruction” first came into use in 1956 when it was used in a speech by Soviet Red Army Marshal Georgi Konstantinovich Zhukov. In fact, it was this speech that highlighted for U. S. policy makers the real or perceived threat from the Soviet Union, particularly in terms of the latter’s presumed arsenal of chemical and biological weaponry. As such, Zhukov’s speech invigorated United States Cold War research into WMD, including biological weaponry. During the Cold War, the United States-and, to a much greater extent, the Soviet Union-amassed large chemical and biological weapons stockpiles. The threat posed by these stockpiles has diminished greatly since the crumbling of the Berlin wall.

Regional threats posed by state-funded militaries from chemical and biological weapons also have declined. By the end of 2003, the U. S. government had admitted that there was little evidence that Iraq had possessed large chemical or biological weapon stockpiles after the mid-1990s. This has since led both the United States and British governments to begin inquiries into the faulty prewar intelligence on Iraq that was in large part the basis for justifying Operation Iraqi Freedom in March 2003. Other regional threats, however, still remain. Among these, states such as Syria and North Korea are suspected of possessing chemical and biological weapons. Their bellicose posture regarding their immediate neighbors and regional rivals, as well as their possession of long-range delivery systems (such as Scud missiles), make these threats impossible to ignore. By contrast, Libyan leader Mohamar Qaddafi stated in early 2004 that he would renounce the possession of WMD, which demonstrates how quickly the threat of weapons of mass destruction seems to rise and fall on the global agenda.

The historical record shows that mass poisonings and the occasional plot to spread disease among armies and civilian populations go back many centuries. Still, chemical and biological warfare (CBW)-sometimes referred to in military parlance as “bugs and gas”-is essentially a modern phenomenon. It is modern in the sense that the science and industry required to produce these types of WMD have only existed since the early 1900s. However, there may indeed have been designs to use chemical or biological agents as a means of warfare (or possibly terrorism) before the Industrial Revolution. Before the late nineteenth century (the time of Louis Pasteur and many developments in chemistry), however, the requisite scientific knowledge and engineering capacity were insufficient to bring any such ideas to fruition. Obviously, this is no longer the case.

Many books and articles that discuss CBW often introduce the subject by bringing up past examples of chemical or biological warfare. In an excellent introduction to chemical weapons, a short book published by the Chinese People’s Liberation Army discusses a case of CW (chemical warfare) from China’s early history: In the Zuochuan, it is written that in the sixth century to about the fifth century B. C. E., “An official of the noble princes of the Xia, came from the Jin to attack the [forces of] Qin, and poisoned the Jing River, killing more than a division of men.” Another case is cited: “In the year 1000 [ C. E.], there was one named Tangfu, who made poi- son fire grenades and gave them to the Chao court of the Song dynasty. The poisonous smoke ball, containing arsenic oxide (As 2 O 3 ) and a type of poison derived from crotonaldehyde, looked a bit like a precursor to a chemical gas grenade. After alighting, this weapon would issue forth smoke to poison the enemy and thus weaken their ability to fight.”

These same authors also point out that this is a far cry from what one expects in modern times, for back then chemical warfare “was just in its infancy, and not only were its methods crude but its utility in actually killing people was limited. Because of this, chemical weapons were regarded as a method to generally assist in conducting warfare, and at the time did not draw any particular attention. Coming into the recent era, as the developments in technology continued, chemical weapons then really began to demonstrate their real menace.”

Another premodern military tactic that is often described as a form of BW (biological warfare) is the siege of Kaffa (1346 C. E.), in modern Feodosia, Ukraine. During a campaign by Mongol forces to defeat a heavily defended city of mostly Genoese merchants, bubonic plague struck the area: “The Tartars died as soon as the signs of disease appeared on their bodies: swellings in the armpit or groin caused by coagulating humors, followed by a putrid fever. The dying Tartars, stunned and stupefied by the immensity of the disaster brought about by the disease, and realizing that they had no hope of escape, lost interest in the siege. But they ordered corpses to be placed in catapults and lobbed into the city in the hope that the intolerable stench would kill everyone inside. . . .” We note here that “stench” was considered in the pre-germ theory era to be responsible for disease. Thus, miasmas, “noxious effluvia,” or “corrupt vapors” (febres pestilentiales) were synonymous with the spread of deadly epidemics-plague (causative organism: Yersinia pestis) being among the most notorious.

The suggestion later made by historians that the Mongols were in fact able to spread bubonic plague by hurling disease-ridden corpses over the fortress walls is an intriguing one. During the fourteenth century, however, a germ theory of disease did not exist. How would the people of that era have known exactly how the disease could spread? What they could not have known is that bubonic plague is spread by fleas, which collect the bacteria Yersinia pestis (the causative organism of plague) through feeding upon infected rats. Fleas do not linger near the body once the temperature of the host (be it rodent or human) cools following death, making it rather unlikely that the cadavers would have done much to spread the plague. In the end, it was not the use of projectile cadavers, but more likely the exceptionally large rat population around the Black Sea that led to a pandemic throughout the region (and indeed much of Europe). One could probably conclude, however, that the Mongols did have the intent to spread disease among their enemy, and at least in this respect they conducted an early form of BW.

Sixth Century B. C. Assyrians reportedly used ergot fungus (Claviceps purpurea) to poison their enemy’s water wells

431-404 B. C. Spartan armies use sulfur and toxic arsenic smoke during Peloponnesian War

Fourth Century B. C. Chinese engineers use arsenic against underground sappers.

Circa 200 B. C. Officers in Hannibal’s army adulterate the wine of African rebels with mandrake, which contains belladonna alkaloids causing hallucinations.

187 B. C. Ambraciots (Greece) employ irritating smoke against Roman soldiers

7th Century C. E. The Byzantine architect, Callinicus (“Kallinikos”), reportedly invents the first liquid incendiary-“Greek Fire.”

Circa 1040 Scottish king poisons wine using a belladonna-like (“sleepy nightshade”) herb and gives to Norwegian enemies as “provisions” under pretense of surrender. Scots then slaughter the incapacitated Norwegians.

1347 Mongolians lay siege to Kaffa (in modern Ukraine) and throw corpses over city walls to spread bubonic plague. May have contributed to Black Death, which killed approximately 50 million people through the fourteenth century.

1672 Bishop of Münster attempted the use of atropine-like drug in grenades in siege against city of Groningen. Attack backfires.

1767 British plot to supply cloths from a smallpox hospital ward to American Indian tribes in hopes of spreading disease. Unknown if this strategy was ultimately successful.

1855 Sir Lyon Playfair suggests using cyanide-containing chemicals against Russian troops during Crimean War, but this tactic never found approval by the British High Command.

29 July 1899 First Hague Convention signed, prohibiting “the use of projectiles the sole object of which is the diffusion of asphyxiating or deleterious gases.”(Mauroni, p. 81)


The Mongol siege of the Crimean city of Kaffa in 1346 is often cited as one of the first recorded incidents of biological warfare-and perhaps even the cause of the spread of bubonic plague to Europe.

The city of Kaffa (or Caffa, now Feodosija, Ukraine), established in 1266 by agreement between the Mongols on the Black Sea and the Genoese, was an important trading hub between Genoa and the Far East. In 1289, the city fell under the suzerainty of the Khan (Toqtai) of the Golden Horde. The relationship between the Genoese and the Khan, however, was an uneasy one. Kaffa was first besieged in 1308 after the reported displeasure of Khan Toqtai over Genoese trading in Turkic slaves. (The sale of these slaves to the Marmelake Sultanate in Egypt reportedly upset the Khan by depriving him of an important source of foot soldiers for his own army.) The Genoese set fire to Kaffa and fled. After Toqtai’s death, Khan Uzbeg allowed the Genoese to rebuild their trading colony in 1312.

In 1343, after a brawl between Christian locals and Muslims in the Italian enclave of Tana inflamed the ire of Khan Janibeg, the Italians fled from Tana to Kaffa, bringing the Khan’s army to the gates of Kaffa behind them to besiege the city. In February 1344, the Italians managed to break the siege after killing 15,000 of the Khan’s Tartars and destroying their siege machines. Janibeg renewed the siege the following year, but the residents of Kaffa were able to maintain their position because they retained access by sea to supplies. In 1346, the Khan’s army besieging Kaffa suffered a natural outbreak of plague. The Tartars catapulted the plague-infected corpses of their dead comrades over the city walls. According to one historical account, the Tartars’ tactic finally broke the 3-year stalemate; the Genoese were crippled by the plague and fled Kaffa by sea-taking the disease to Europe with them.

The most contemporaneous account of the siege was written by Gabriele de’ Mussi, a notary of the town of Piacenza, north of Genoa. There is some debate as to whether de’Mussi witnessed the events at Kaffa. Written in 1348 or 1349, the account de- scribes the “mysterious illness “that struck the Tartar army besieging Kaffa. De’ Mussi recounts how the Tartars, desperate from the devastation of the disease on their army, thought to kill the inhabitants of Kaffa with the stench of their diseased dead. According to the de’ Mussi account, the people of Kaffa had no hope once the air and water had been contaminated, and only one in 1,000 was able to flee the city. Those that did flee took the plague with them as they left.

De’ Mussi’s account suggests that not only did the Tartars deliberately hurl their diseased dead over the city walls of Kaffa with the intent to kill their enemies, but those escaping Kaffa brought the disease into the ports of Europe. The disease was most likely brought within the walls of Kaffa through flea-infested rodents from the Tartar camps, or possibly through the transmission of the disease from direct contact with infectious body fluids from the Tartar dead.

Most scholars believe that the Genoans brought the plague with them to Naples, from where it then spread throughout Europe. Others have recently suggested that although the use of plague corpses against Kaffa was a true act of biological warfare, the siege had no significant impact on the spread of the Black Death through Europe. As Wheelis suggests, Kaffa was certainly not the only Tartar port that could have transmitted plague into European ports. Wheelis further argues that the rate and pattern of plague transmission suggests that it took 1 year to spread the plague into different European ports.

Though Kaffa may not have been the precise source of the Black Death that spread into Europe, the use of infected cadavers against its besieged inhabitants remains one of the most important instances of the intentional use of disease in warfare.

References McGovern, Thomas W., and Arthur M. Friedlander, “Plague,” in Russ Zajtchuck and Ronald F. Bellamy, eds., Medical Aspects of Chemical and Biological Warfare (Washington, DC: Borden Institute, 1997), pp. 479-502. Watts, Sheldon, Epidemics and History: Disease, Power, and Imperialism (New Haven, CT: Yale University Press, 1998). Wheelis, Mark, “Biological Warfare at the 1346 Siege of Caffa,”Emerging Infectious Diseases, vol. 8, no. 9, September 2002, http://www. cdc. gov/ncidod/EID/ vol8no9/01-0536. htm.

Russia and the Mongol Empire

The Golden Horde. Russian name for the Mongol domains of
Russia and neighboring areas. The name derives from a nickname for the palace tent (Middle Mongolian hordo) of the Golden Horde ruler.

The Mongol conquest of the Russian duchies ironically led to the emergence of Muscovy as the center of a new united Russian state. At the time of the Mongol invasions the Russians, or East Slavs (including the Ukrainians and Belarussians), were ruled by princes (or dukes, kniazi) of the Riurikid family, dating back to the ninth century. Since no system of primogeniture existed, the number of new appanages (udel) multiplied steadily. Politically, the unifying institution was the grand duchy, that is, the special title of grand prince (or grand duke, veliki kniaz’), which accrued, after 1169, to the possessor of the city of Vladimir in the northeast. However, this position of grand prince had no set rule of succession, leading to complex family politics. The real unity of the Russian land, surrounded by Catholic, Muslim, and pagan neighbors, was provided by the Orthodox Church, conducting its services in Old Church Slavonic and headed by the metropolitan in Kiev. Economically, the Russian duchies had ceased coinage in the 12th century, using furs for money instead. The largest city, Novgorod, which traded Russia’s furs, falcons, and lumber to the Baltics, had an estimated 22,000 persons.

The Conquest

The Russians first collided with the Mongols (or TATARS, as the Russians always called them) through their Turkish- speaking Qipchaq (Polovtsi or Cuman) allies. When SÜBE’ETEI BA’ATUR and JEBE led a reconnaissance force of three tümens (10,000s) through Qipchaq territory in 1223, the latter appealed to allied southwest Russian princes, who joined the QIPCHAQS, slaying the Mongol ambassadors. Sübe’etei and Jebe crushed the Russian- Qipchaq force on the Kalka River (May 31, 1223).

Thus, in 1235 ÖGEDEI KHAN added the Russians to the list of targets of the Mongols’ great Western campaign. As usual, the Mongols campaigned principally in the winter. The main Mongol force, headed by the Jochid princes BATU and Hordu, the future great khans GÜYÜG and Möngke, and several others, arrived at Ryazan’ in December 1237. Once Ryazan’ refused to surrender, the Mongols sacked it and stormed through the north-eastern district of Suzdalia, sacking its cities and defeating Russian field forces before leaving that summer. Among the casualties was Grand Prince Iurii (1217–38), killed on the River Sit’ (March 4, 1238). The Mongols reappeared in southern Russia in 1239, sacking Pereyaslavl’ (March 3) and Chernihiv (Chernigov, October 18). Finally having smashed the Russians’ Qipchaq allies, the Black Caps, the full Mongol army sacked Kiev (December 6, 1240), Halych (Galich), and Volodymyr (Vladimir) before passing on to Central Europe. Throughout the campaign the Russians showed neither unity of purpose nor any sense of the enemy they were facing. No Russian princes surrendered to the Mongols, but most fled when it became clear resistance was futile.

When the Mongols destroyed Kiev in December 1240 and sacked Halych (Galich) and Volodymyr (Vladimir), Prince Daniel of Halych (d. 1264), his brother Vasil’ko of Volodymyr (d. 1269), and Michael of Chernihiv (Chernigov, d. 1246) all took refuge in Poland.

The princes of the dynasty, who survived and succeeded those killed defending the Rus’ lands, faced multiple problems of consolidating their own positions, reconstructing their lands, and also developing working relationships with their conquerors. In the first years after the invasion, while the Mongols were pursuing military campaigns elsewhere, the remaining Riurikids were virtually left to their own devices to recover and restore order.

Mongol suzerainty

Nevertheless, the invasion of the Rus lands, although not followed immediately by any formal treaty or Mongol occupation, constituted for practical purposes a conquest. A new, political relationship had to be forged between the Mongol khans and the Riurikid princes. The khans had to find methods of exercising their authority over the defeated lands; and the Rus princes and their populations had to accept and adjust to the demands of the Horde. The princes of Rus recognized Batu and his successors as their overlords. The first evidence of this relationship became manifest almost as soon as the Mongols’ western campaigns ended. According to chronicle accounts, they began the practice of traveling to the Mongol khan or “going to the Horde” to receive the khan’s iarlyk or patent, which was an official appointment or confirmation of each prince’s right to rule his domain.

As early as 1243, Prince Iaroslav Vsevolodich, who had replaced his brother Iurii as prince of Vladimir after the latter’s death at the Battle on the Sit, made such a trip to the Horde. He was awarded not only the title of grand prince of Vladimir, but also that of grand prince of Kiev. Three years later Iaroslav returned to Sarai. On that occasion he was sent on to the Mongol capital at Karakorum. He did not survive the journey. Iaroslav’s sons, Andrei, who served as prince of Vladimir from 1249 to 1251/52, and Alexander Nevsky, who replaced him in 1252, also traveled to both Sarai and Karakorum. In subsequent years other princes also repeatedly made the trip to the Horde. The Suzdalian princes alone, according to one count, made nineteen visits between 1242 and 1252.

By 1280 Noqai, leader of a junior Mongol line, had established a virtually independent realm from the Dnieper to the Danube, ruling Ossetes, Vlachs (Romanians), and Russians of Halych and Volodymyr. King Ladislaus IV (r. 1272–90). The figure who divided the Golden Horde was Nogai. Leader of the Mangkyt clan and himself a descendant of Juchi, Nogai emerged during the reign of Mengu-Timur (1266/67–81) as a powerful military commander with virtually autonomous control over the western territories of the Golden Horde, i.e., the lands west of the Dnieper reaching to the lower Danube. Nogai’s influence was such that he engaged in direct interaction with Byzantium, Egypt, and Hungary and conducted military expeditions as far west as Serbia, Transylvania, and Hungary. When Tuda-Mengu succeeded his brother Mengu- Timur, Nogai’s power as clan leader, Horde elder, and military commander was so great that he has been characterized both as a “virtual co-ruler” with the khan and as an independent ruler. Nogai continued to play a powerful role when Tuda-Mengu, having personally converted to Islam (1283), delegated much of his authority and responsibility to his nephew Telebuga.

The south-western Rus’ principalities of Galicia and Volynia continued to be entangled with Lithuania as well as Hungary and Poland after Daniil’s death in 1264. But it was not uncommon in the 1270s for the Tatars to assist Daniil’s heirs in their campaigns against their western neighbors. When Nogai and Telebuga conducted their campaigns in Hungary in 1286 and 1287, troops from Galicia and Volynia joined them. By that time, however, the unity and power of the two principalities were declining. Tatar troops passing through and wintering in Galicia and Volynia during those campaigns destroyed agricultural fields and looted the region. Unable to protect their lands, the Rus’ princes lost credibility and authority, and Mongol officials became more directly involved in the south-western principalities, especially during the height of Nogai’s power, than they did in the northeast.

Not long afterward, however, as the Sarai khans consolidated their authority after Nogai’s death, their grip on the western principalities relaxed. By the end of the thirteenth century, Daniil’s son Lev (d. 1301) and grandson Iurii (d. 1308) were temporarily able to recombine the Volynian and Galician lands. But it was Poland and the rapidly expanding state of Lithuania that ultimately not only replaced the Riurikid princes, but also competed effectively and successfully with the Mongols for control over the region. Poland absorbed Galicia by 1349; Lithuania gradually acquired the remainder of the south-western lands, including Kiev. The divergence of the south-western from the north-eastern Rus’ principalities that had become apparent before the Mongol invasion was thus intensified; the two sectors that had previously formed Kievan Rus’ split apart.

Golden Horde (Qipchaq Khanate, Ulus of Jochi)

The Golden Horde, founded by CHINGGIS KHAN’s eldest son, Jochi, unified for the first time the lands around the Kazakh, Caspian, and Black Sea steppes. Successors to the Golden Horde ruled under Russian sovereignty into the 20th century. The name Golden Horde derives from the gold-hung palace-tent (horda or ORDO) at which ÖZBEG KHAN (1313-41) received visitors. When Russian chronicles mentioned “going to the Horde,” horde was being used in its proper sense of a nomadic palace, not the later European sense of a mass of people. As the realm disintegrated, the chroniclers referred to other ordos at the center of splinter regimes: the BLUE HORDE, the Volga Horde, the Great Horde, and so on. Implicitly, the palace-tent stood for the people gathered around their ruler. Not until the 16th century, however, did Russian chroniclers begin explicitly using Golden Horde to designate this Mongol successor state. Persian sources of the 13th and 14th centuries, undoubtedly reflecting Mongol usage, spoke either geographically of the Dasht-i Qifchaq, “Qipchaq Steppe” or dynastically of the “ulus (realm) of JOCHI,” Chinggis Khan’s eldest son and ancestor of its khans.


The Golden Horde army was the largest of the three western khanates but neither as battle worthy nor as well equipped as those of the CHAGHATAY KHANATE and the IlKhanate. During the 1357 invasion of Azerbaijan, the conventional wisdom said the Horde’s vast army was “horsemen without weapons.” The bulk of the army must have been Mongol clans and their native Turkish subjects. Still, Rashid-ud-Din speaks of Russians, Hungarians, and Circassians being brought into both right- and left-hand armies, and they do figure occasionally in battle accounts. In 1277 the Russian prince of Rostov won distinction in the siege of an Ossetian fortress.


A constant of Golden Horde foreign policy was hostility to the Il-Khans. While some rulers did not actively pursue their claims beyond the Caucasus, none, except possibly Toqto’a (1291-1312), ever considered abandoning them. Ultimately, Sultan Özbeg’s sons and grandson pursued this claim to ultimate success by occupying Tabriz, in modern Iran, in 1357-59, although they could not hold on to their conquest. To outflank the Il-Khans, the Golden Horde maintained the Egyptian alliance begun by Berke and Baybars. Although the alliance never produced the hoped-for military benefits, it did play a substantial role in the cultural and religious life of the Horde. Geography dictated that this alliance needed the Byzantine Empire as a third partner, and keeping Byzantium in line necessitated land access to Constantinople, which in turn necessitated control over Bulgaria. From Berke to Özbeg these requirements were more or less maintained.

Eastward, the Golden Horde’s primary interest lay in thwarting the Chaghatayids’ ambitions toward Khorazm and the Syr Dar’ya cities. From 1269 to 1284 the khans pursued this aim by encouraging QAIDU and the Chaghatayids’ ambitions to the south and east. From 1284, however, the Golden Horde became wary of Qaidu’s expansion and opened friendly relations with the YUAN DYNASTY.

Central Europe impinged on the Golden Horde primarily as a source of instability among the Russian principalities. Local Jochid princes and noyans met Lithuanian and Polish raids on southwestern Russian towns (modern western Ukraine) with counterraids, which were, however, often as damaging to the local Russians through whom the Mongol soldiers passed as to the Poles or Lithuanians.

The most potent weapon of the Mongols was terror

The twelfth- and early thirteenth-century walls of Merv (Turkmenistan). The photograph shows how the twelfth-century walls, with their hollow towers and arrow slits, have been strengthened by a massive outer covering, probably hastily built in the face of the Mongol invasions. It was all to no avail: the city fell without any serious resistance and most of the inhabitants were massacred.

To the local people they appeared completely strange and alien; and, unlike other adversaries, they were non-Muslims with no respect for mosques or holy places. Furthermore, with the exception of a small number of artisans, and selected attractive girls and boys, the Mongols did not regard the conquered populations as assets whose talents could be exploited but rather as wasteful occupants of good grazing space and, in addition, potentially dangerous. Other military adventurers would protect a city to enjoy its revenues, if for no more elevated reason. Not so the Mongols, who seem only to have wanted plunder. On a number of occasions when cities were taken, the inhabitants were ordered out into the surrounding plains for six or seven days so that the Mongols could pillage their houses thoroughly and at leisure. If the locals accepted this without resistance or protest, they might be allowed back to the remains of their dwellings.

The survivors, however, were the lucky ones. As the conquest proceeded, the Mongols became even more ferocious. At Bukhara, which was conquered early on, only the Turkish soldiers were systematically slaughtered. At Urgench, where there had been fierce hand-to-hand fighting through the streets, the people were driven out of the city, artisans were separated and taken away, the children and young women were reduced to slavery and ‘the men that remained were divided among the (Mongol) army, and to each fighting man fell the execution of twenty-four persons’. After the fall of Balkh (in northern Afghanistan), Genghis Khan commanded that ‘the population, ‘small and great, few and many, both men and women should be driven out on the plain and divided up according to the usual custom into hundreds and thousands to be put to the sword; and that not a trace be left of fresh or dry. For a long time wild beasts feasted on their flesh.’ After the people of Merv had agreed on terms for surrender,

the Mongols entered the town and drove all the inhabitants, nobles and commoners out on to the plain. For four days and nights the people continued to come out of the town: the Mongols detained them all, separating the women from the men … the Mongols ordered that, apart from 400 artisans they specified and selected from among the men and some children, girls and boys, who they bore into captivity, the whole population, including women and children, should be killed, and none, whether man or woman, should be spared. The people of Marv [MervJ were then distributed among the soldiers and levies and, in short, each man was allotted the execution of three or four hundred persons.

The historian Ibn al-Athir was a typical product of the Muslim bourgeoisie of the early thirteenth century. He was immensely learned and well-read and had the ability, not shared by all his colleagues, to demonstrate this learning in simple, straightforward prose. He had visited north-eastern Iran in the years immediately before the Mongol invasions and had been impressed by the size and wealth of the cities and the richness of their libraries. Whether by good luck or good judgement he had returned to his native Mosul (a city never taken by the Mongols) shortly before the storm broke. His appalled reaction to the invasions show the terror the Mongols inspired among people who had never seen them.

Ibn al-Athir tells us:

I have heard that one of them took a man captive but did not have a weapon to kill him with, so he said to his prisoner, ‘Lay your head on the ground and do not move’, and he did so and the [Mongol] went and fetched his sword and killed him. Another man told me the following story: ‘I was going with seventeen others along a road and we met a Mongol horseman who ordered us to tie up each others’ arms. My companions began to do as he said but I said to them, “He is only one man, why don’t we kill him and escape?” but he replied, “We are afraid.” I then said, “This man intends to kill you immediately so let’s kill him and perhaps God will save us.” But I swear by God that not one of them dared to do this so I took a knife and slew him and we fled and escaped.’ There were many such events.

Whether the anecdote is true or not, it shows how the sinister reputation of the Mongols spread. It also highlights a phenomenon known from other war situations, the passivity and hopelessness which can overcome people when faced with an enemy they believe to be stronger, leading to a meek acceptance of their fate. These attitudes provide some insight into the secrets of Mongol success. The Mongols certainly gloried in and publicized their reputation for terror. When Genghis Khan took Bukhara, he gathered the survivors in the great mosque. The scene was one of complete desecration. The chests in which the great old Qurans were kept had been tipped out so that the leaves were lying around in the ·dust while the boxes themselves were, used as troughs for the Mongols’ animals. He addressed his cowed audience:

‘O people, know that you have committed great sins and that the great ones among you have committed these sins! If you ask me what proof I have for these words, I say it is because I am the punishment of God. If you had not committed great sins, God would not have sent a punishment like me upon you!’ One of his audience told his friend who wanted to object, ‘Be silent! It is the wind of God’s omnipotence that blows and we have no power to speak.’

Revisionist historians have questioned the extent of Mongol ferocity and destructiveness, suggesting that such accounts are largely rhetoric and hyperbole. However, the weight of contemporary evidence is very strong and it is backed up by the archaeology. Of the great cities sacked by the Mongols, only Bukhara and Urgench were rebuilt on the same site: Balkh, Otrar and Nishapur were ruined for ever and at Merv a new town was founded two centuries later well away from the remains of the old. Samarkand was rebuilt outside the old walls while the ancient city remained as it is today, a desolate waste of mud-brick ruins.