Silk Road Atlas silk-road.xyz

Religion & Culture

Paper-making on the Silk Road

The transfer of paper-making from China to the Islamic world and, from there, to Europe is the single most consequential technology transfer in the history…

The transfer of paper-making from China to the Islamic world and, from there, to Europe is the single most consequential technology transfer in the history of the Silk Road. The development of paper, and later of printing, transformed the economics of literacy, the scale of government, the character of scholarship, and the structure of the early modern world. The transmission of paper-making westward, beginning in the eighth century CE and continuing over more than four centuries, is one of the central stories of Silk Road cultural exchange.

Cai Lun and the Chinese tradition

Paper had been developed in China by the first century CE, and the technology was refined and consolidated in the workshops of the Han and Tang empires. The traditional Chinese account attributes the systematisation of the industry to Cai Lun, a eunuch official of the Eastern Han court, who is said to have presented a refined paper to the emperor in 105 CE. The archaeological record shows that paper was in use in China from at least the early first century BCE — the earliest surviving example was found at Fangmatan in Gansu (a fragment dated to around 8 BCE, in the British Library’s collection of IDP material) — but Cai Lun’s contribution is generally treated as the turning point. I think the Cai Lun story is half history, half foundation myth, and the proportion shifts depending on which Chinese source you read.

Paper is, in modern terms, the most consequential of all Silk Road technologies.

Chinese paper was made from a wide range of plant fibres: hemp, mulberry bark, rags, and various grasses. The fibres were pulped, suspended in water, and lifted on a bamboo screen, the fubo, which retained the mat of fibres as the water drained through. The resulting sheet was dried and polished, and could then be cut or bound. The principal production centres were in the Yangtze valley, in Sichuan, and in the lower Han valley. The technique is not, on inspection, especially complicated — the screen-and-deckle method was reinvented independently several times — but the Chinese were the first to industrialise it, and the gap between Chinese production volumes and anything elsewhere before the eighth century is enormous.

By the Tang dynasty, paper was in widespread use in China for documents, for Buddhist sutras, and for the new practice of printing. The Dunhuang manuscript collection, sealed in the Library Cave (Cave 17, by Wang Yuanlu’s later account) in the early eleventh century and rediscovered by Wang in 1900, contains tens of thousands of paper manuscripts, including printed almanacs and Buddhist texts that are among the earliest surviving examples of woodblock printing. A point that is often missed: the sealing of the cave around 1000–1010 was not, as the older accounts had it, the work of a single monk — it was a slow accumulation over decades, and the latest datable manuscript in the cache is from around 1002.

The Battle of Talas and the capture of papermakers

The principal event in the transmission of paper-making westward was the Battle of Talas in 751 CE, in which an Abbasid Arab army defeated a Tang Chinese force near the Syr Darya in what is now Kyrgyzstan, on the border between the modern provinces of Talas and Chui. According to the principal Arab source, the historian al-Azdi (whose Conquest of Sīn is the standard reference), the victorious Arabs captured a number of Chinese artisans, including papermakers, who were settled in Samarkand and put to work. The archaeological record of the eighth and ninth centuries shows a sudden appearance of paper in the markets of Central Asia, in Iran, and in the Islamic Mediterranean, with datable fragments appearing first in Samarkand and then, within fifty years, in Damascus, Cairo, and Fustat.

The Battle of Talas is one of the great turning points in the history of technology. Within a generation of the battle, paper factories had been established in Samarkand, Bukhara, and other Central Asian cities, and the new product was being exported across the Islamic world. The paper trade was a direct consequence of the military confrontation between the Tang and the Abbasid caliphate. Popular accounts tend to overstate the cleanness of the transmission — paper did not, in fact, spread in a single wave from Samarkand; the Islamic Mediterranean had its own independent experiments with papyrus and parchment substitutes, and the Chinese-derived industry competed with these for at least a century. What the Battle of Talas gave the Abbasids was not a monopoly but a head start.

The Royal Museum of Central Africa at Tervuren, Belgium, has the world’s largest single collection of Arabic and Persian manuscripts on paper from the post-1100 period. The Bibliothèque nationale de France’s département des manuscrits (the Richelieu site, on rue de Richelieu) has the earlier Abbasid material. If you want to see the transmission in the flesh, both collections are open to researchers, and both have a few pieces on permanent display.

A personal note

A note on the Battle of Talas. The traditional account — that Arab forces captured Chinese papermakers at the Battle of Talas in 751 — is the foundation myth of paper-making transmission. It is supported by some Arabic sources (notably the later accounts of Ibn al-Nadim) but is not directly documented in the surviving Chinese sources. The capture may have happened; the heroic status of the story is more recent than the event.

Further reading

  • Hansen, The Silk Road: A New History (Oxford, 2012). The single best modern overview.
  • The relevant chapter of The Cambridge History of Inner Asia (Cambridge, multiple dates).
  • One article from a recent issue of Central Asiatic Journal or Journal of the Economic and Social History of the Orient (JESHO) for current scholarship.
  • Susan Whitfield, ed., The Silk Road: Trade, Travel, War and Faith (British Library, 2018) for the British Library’s own primary-source scholarship.