Everything in this course so far has assumed a route crossing dozens of independent polities, each taking its cut and enforcing its own law, and for about a century that assumption was false.
The Mongol conquests put the entire span from Korea to the Black Sea under a single family. That makes the thirteenth and fourteenth centuries the one period where the effect of political fragmentation on Eurasian exchange can be estimated, by the crude but genuine method of taking it away. The results are instructive in both directions, because the same century that produced the safest roads in the subject's history also produced the worst destruction in it, and then killed a third of the population of the western end.
What was unified
Temujin was proclaimed Chinggis Khan in 1206. Within seventy years his successors held China, Central Asia, Persia, Mesopotamia and the Russian steppe, some 24 million km² at the greatest extent, roughly five times the area of the Roman empire at its height and the largest contiguous land empire there has ever been.
It did not stay one empire. After the succession struggle of 1260 it functioned as four: the Yuan in China and Mongolia, the Ilkhanate in Persia, the Chagatai khanate in Central Asia and the Golden Horde on the Russian steppe. They fought each other, sometimes bitterly. But they shared a ruling family, a language of administration, a legal tradition and a set of institutions, and for most of the period a merchant crossing between them was crossing between relatives rather than enemies.
The yam
The institution that mattered most for exchange was the postal relay, the yam. Stations were placed roughly a day's ride apart along the main routes, each obliged to keep horses, fodder, food and lodging ready, supplied by the local population as a tax in kind. Yuan records indicate something like 1,400 stations in China alone. Authorised travellers carried a paiza, an inscribed tablet of wood, silver or gold according to rank, which entitled the bearer to draw on the stations.
The gain is not a modest improvement, and it can be quantified.
Example. A single rider with one horse sustains about 40 km a day over a long journey, because the horse must rest and graze. A relay courier changing horses at every station manages 300 km a day. Karakorum to Tabriz is about 4,630 km. Compare the two.
The lone rider takes days, close to four months. The relay courier takes days. The ratio is 7.5, and the reason is worth stating precisely: the limit on a long ride is the animal's endurance, not the rider's, so an institution that supplies fresh animals converts a biological constraint into a logistical one. Nothing about the technology changed, no faster horse was bred, and no road was built. What changed is that the state guaranteed remounts, which is an administrative achievement rather than a technical one. The consequence for the empire is that orders could reach a frontier and a reply come back within a season, and the consequence for merchants, who could not use the yam but travelled roads that its stations made safe and supplied, was almost as large.
Now you. The yam was paid for by requisitioning horses, fodder and labour from the populations near the stations rather than out of a central treasury. What would you expect that to do over time, and how would you check?
Answer
You would expect the burden to fall hardest where traffic is heaviest, to grow as officials and their friends abuse the right of requisition, and eventually to depopulate the station districts as people move away from a tax they cannot escape, degrading the very network the empire depends on. That is a standard failure mode of any corvee obligation tied to a fixed place. The check is documentary, and the Yuan sources oblige: there are repeated edicts restricting who may use the stations, complaints about unauthorised holders of paiza, and attempts to limit the number issued, which is the signature of an abuse the centre could see and could not stop. There is also an economic tell worth looking for: if the burden were being borne willingly you would not need to keep legislating about it. The general lesson is that infrastructure funded by requisition rather than by taxation or tolls has no feedback loop connecting what it costs to what it earns, and tends to be over-consumed by exactly the people with the authority to do so.
What the merchants said
Two kinds of evidence show that the traffic responded. The first is institutional: Mongol elites invested in trade directly through the ortogh, partnerships in which a prince or a khatun supplied capital to an association of Muslim or Uighur merchants who traded it and returned an agreed share. That converts the ruling class from predators on trade into stakeholders in it, which is a rare and valuable alignment.
The second is a handbook. Francesco Balducci Pegolotti, an agent of the Bardi company of Florence, compiled La pratica della mercatura around 1340, a merchant's reference of weights, currencies, tariffs and routes. On the road from Tana on the Sea of Azov to Cathay, he reports that it is perfectly safe by day or by night, according to the merchants who have used it. He lists the stages, from ox wagon to camel wagon to pack ass, with the days each takes, and the total runs to something like eight or nine months of travel. He advises taking a dragoman, servants who know the language, and provisions, and he notes what will and will not sell.
The last clause of his safety claim is the important one: according to the merchants who have used it. Pegolotti never went. He was compiling from informants for a commercial house, in a manual whose purpose was to encourage and equip trade, and by the time the book was in circulation the political conditions it described were already dissolving. It is excellent evidence that Florentine merchants believed the road across Asia was open in the 1330s, and no evidence at all about how many took it.
Example. How much weight can Pegolotti's statement that the road is perfectly safe by day or by night bear?
Less than it is usually made to carry, and more than nothing. Weigh it as you would any testimony: the author never travelled it, he names no informant, he is writing a promotional and practical manual for a banking house with money to place, and the manuscript circulated after the conditions it describes had begun to fail. Against that, he had no reason to lie to his own colleagues, who would lose money if he were wrong, and the detail elsewhere in the book, currencies, weights, tariffs, what sells where, is accurate enough to be used as a source for Mediterranean commerce. So the sentence is good evidence that Florentine merchants in the 1330s believed the Asian route was open and usable, which is itself a fact about the Mongol peace worth having, and it is no evidence about how many merchants went, what they carried, or whether the road really was safe. The distinction between evidence of belief and evidence of fact is one to keep, because a great deal of what survives from this period is the first kind and gets quoted as the second.
Now you. European accounts of the Mongols shift from horror in the 1240s to something close to enthusiasm by the 1290s. What changed, and what should you infer from the shift?
Answer
Two things changed, and only one of them is about the Mongols. The invasions of Hungary and Poland in 1241 had been a direct and terrifying threat to Latin Christendom, and by the later century that threat had receded while the Ilkhanate, at war with the Mamluks of Egypt, looked like a possible ally against a common enemy. There was also a religious hope, encouraged by the presence of Christians at the Mongol courts and by embassies like Rabban Bar Sauma's, that the khans might convert, which fed a long European fantasy of a great eastern Christian ally. What you should infer is that these sources are reporting European strategic and religious hopes at least as much as Asian realities, and that their tone is a poor guide to conditions on the ground. The correct use of them is what this course has done throughout: extract the checkable details, the routes, the prices, the institutions, the things a writer had no motive to invent, and treat the evaluations as evidence about the writer.
The travellers
The Mongol century produced the only substantial body of eyewitness travel writing across the whole route. John of Plano Carpini was sent by the pope in 1245 and William of Rubruck by Louis IX in 1253, and both wrote sober, observant reports on Mongol society. Ibn Battuta, who left Tangier in 1325 and travelled for nearly thirty years, covered the Islamic world and beyond. Rabban Bar Sauma, a Uighur Christian monk born near modern Beijing, went the other way: sent west by the Ilkhan in 1287 as an envoy, he met the Byzantine emperor, the king of France, and Edward I of England at Bordeaux, where he gave the English king communion.
And Marco Polo, in Asia from 1271 to 1295, dictated in a Genoese prison the book that shaped the European imagination of the east for two centuries. Whether he went at all has been argued: Frances Wood's Did Marco Polo Go to China? (1995) pointed to his silence on tea, chopsticks, foot binding and the Great Wall, and to his absence from Chinese records. The strongest response, Hans Ulrich Vogel's Marco Polo Was in China (2013), works from the opposite end, showing that his descriptions of paper money denominations, salt production and the salt revenues are accurate in detail, sometimes in ways no other western source records and that match Chinese fiscal documents. The current balance of opinion is that he was there, that he served the Mongol administration in some capacity, and that his book is padded with material from other people's accounts.
Notice the asymmetry, because it is more revealing than any individual traveller. Many Europeans went east and wrote about it; very few Asians went west and did. That is not a comment on curiosity. Europe was the poor, peripheral, fragmented end of the system with everything to gain from contact, and the flow of travellers ran up the gradient of wealth, as it usually does.
Specialists, moved at swordpoint
The previous lesson concluded that skill moves only inside a person, and slowly. The Mongols found the exception: move the person by force.
Conquered cities were systematically stripped of craftsmen, who were spared the general massacre and deported to work for the khan. Persian and Central Asian artisans were resettled in Mongolia and China; Chinese and Khitan specialists went west. In 1267 an astronomer named Jamal al-Din presented Kublai with Persian instruments and a calendar, and an Islamic astronomical bureau operated alongside the Chinese one at the Yuan court. Rashid al-Din, the Ilkhanid vizier, assembled at Tabriz a world history that included a section on China and a compilation of Chinese medicine in Persian, working with Chinese informants at his court. Most vividly, when the Mongols were stuck before the fortress city of Xiangyang, they sent for engineers from the Ilkhanate who built counterweight trebuchets of Islamic design, and the fall of the city in 1273 opened the Yangzi and finished the Song.
That is a technique crossing 5,000 km in a couple of years, against the ten kilometres a year of the previous lesson. The mechanism is the one identified there, deliberate recruitment rather than diffusion, applied by a state that did not need to persuade anybody.
The cost
None of this should soften what the conquest was. Merv surrendered in 1221 and was destroyed; Nishapur likewise. The chronicles give figures: Juvayni, writing a generation later under Mongol employment, says 1.3 million died at Merv, and Ibn al-Athir gives 700,000.
Example. The walled area of Sultan Kala, the Seljuk city at Merv, is about 400 hectares. Test Juvayni's figure against it.
At 1.3 million people in 400 hectares the density would be per hectare, which is 32,500 per km². Pre-modern cities of mud brick and courtyard houses run to perhaps 100 to 200 people per hectare, and even the densest quarters of the most crowded pre-industrial cities do not approach a tenth of Juvayni's implied figure. Taking 200 per hectare as a generous ceiling gives about 80,000 inhabitants, so the chronicle number is too large by a factor of roughly sixteen. What the arithmetic does not show is that nothing happened. Merv was a great city and afterwards it was not: the excavations show contraction, the irrigation system that made the oasis possible failed, and the site of the medieval city was never rebuilt to the same scale. The correct statement is that a catastrophe occurred whose magnitude the sources cannot measure, and that the chroniclers' numbers are literary devices for enormity rather than counts.
Now you. Why would a historian in Mongol service like Juvayni inflate the number of people his employers had killed?
Answer
Because terror was the policy, and a chronicler who exaggerates it is advertising it rather than criticising it. The Mongol method depended on the credible threat that resistance meant annihilation while surrender meant survival, and that threat is only as good as its reputation, so enormous massacre figures circulating in every court and caravanserai were an instrument of war that saved the cost of later sieges. Juvayni also had literary reasons: he was writing in a Persian historiographical tradition where vast round numbers signal the scale of a calamity and are not expected to be audited, and a religious frame in which the conquest is a divine punishment whose scale demonstrates the sin. And he was a Persian official serving the conquerors of Persia, for whom emphasising the horror while working for its authors was a way of grieving in the only register available. None of this makes him a liar in his own terms. It makes his numbers a genre convention, which is how they should be read, while his account of events, chronology and personalities remains extremely valuable.
Why it stopped
The unity dissolved fast. The Ilkhanate broke up after 1335, the Chagatai khanate split, the Yuan were driven out of China in 1368 and the Ming that replaced them regarded the steppe as an enemy rather than a partner. The successor states fought over the same routes their predecessors had guaranteed.
And in the middle of that, between 1346 and 1353, the plague arrived at the western end of the system and killed something between a third and a half of everyone in Europe, the Middle East and North Africa. The two facts are related, though not in the crude way often implied, and the relationship has been transformed in the last few years by evidence that has nothing to do with chronicles at all.
That is the next lesson: the best documented cargo the routes ever carried, and the one nobody chose to send.