Why China Invented Everything and Then Stopped

In 1088 CE, a Chinese civil servant named Shen Kuo published a book called Dream Pool Essays that casually documented, among hundreds of other observations, the existence of movable type printing (invented by Bi Sheng roughly forty years earlier), the magnetic compass, the use of petroleum and natural gas as fuels, the principles of erosion and sedimentation that would not be formalized in Western geology for another seven centuries, and mathematical techniques for interpolating astronomical data. Shen Kuo was not primarily a scientist or inventor; he was a government official and polymath writing down interesting things he had noticed. The casual density of technological sophistication in a single eleventh-century Chinese administrative text is arresting when you sit with it.

Europe at this moment was deep in the medieval period, roughly two to three centuries away from developing some of the technologies Shen Kuo was cataloguing as established facts. China had the compass, Europe would not develop it for another century. China had paper money in circulation; Europe would not attempt paper currency for another six centuries. China had gunpowder weapons deployed in warfare; Europe’s gunpowder technology would not catch up for another two centuries. The technological gap between Song Dynasty China and contemporary Europe was, by most measures, enormous — arguably larger than any technology gap between nations in the modern era.

Four centuries later, when Portuguese ships arrived in China in 1513, the balance had shifted completely. European naval technology was superior to Chinese. European firearms were more advanced. And more significantly, Europe was in the midst of an accelerating technological revolution — the Scientific Revolution, the printing press’s consequences, the beginnings of systematic experimental methodology — that would leave China behind for the next four centuries.

What happened? This is not a trivial question. It may be the most important question in economic history, because the answer tells us something fundamental about what conditions produce sustained technological innovation and what conditions kill it.

The Song Dynasty as Peak Innovation

To appreciate the magnitude of the subsequent stagnation, it helps to understand how extraordinary the Song Dynasty (960–1279 CE) was as an innovative environment. The Song period produced or substantially advanced paper money, movable type printing, the magnetic compass, advanced ironmaking and coking techniques, gunpowder weapons, water-powered mechanical clocks, sophisticated hydraulic engineering, and agricultural innovations including new rice varieties and sophisticated irrigation systems that supported the densest and most productive agriculture in the world.

The Song economy was, by the standards of its time, extraordinarily commercialized. China’s internal market was probably the largest single market in the world, connected by a sophisticated canal and river network that made internal trade cheap. Paper money facilitated commerce. There was a substantial merchant class. Urban populations were high: Hangzhou, the Southern Song capital, had roughly one to two million inhabitants by the twelfth century, making it the largest city in the world by a significant margin.

Song China also had the world’s most sophisticated examination-based civil service — the keju system — which in theory opened the government to talented men regardless of birth. This is often cited as one of China’s great institutional achievements, a meritocratic alternative to the hereditary aristocracy that structured European governance. And in certain respects it was. But the keju’s consequences for technological innovation were deeply ambiguous, and understanding this ambiguity is central to understanding what went wrong.

The keju examined candidates on their mastery of the Confucian classics: the Four Books and Five Classics, literary composition, and policy essays grounded in classical precedent. Technical knowledge — mathematics, engineering, natural philosophy — was not examined and conferred no status. The Chinese state produced an enormous annual cohort of highly educated men competing intensely for government positions, and it selected them entirely on humanistic literary criteria. The smartest young Chinese men spent their formative years memorizing classical texts rather than doing mathematics or experimenting with natural phenomena. The opportunity cost of the examination system for technical development was real and large.

The Needle Trades: Why Technology Stopped Diffusing

One of the clearest illustrations of what went wrong is the history of printing technology. Bi Sheng invented movable type printing in China around 1040 CE. Johannes Gutenberg invented movable type printing in Europe around 1440 CE — roughly four centuries later. In Europe, the printing press triggered an information revolution within decades: standardized texts, vernacular literacy, the Protestant Reformation, the wide diffusion of scientific knowledge, the development of scientific journals, and the gradual construction of the networked scientific community that produced the Scientific Revolution.

In China, movable type printing spread slowly and unevenly, and its consequence for intellectual life was far more modest. Why the difference?

Part of the answer is linguistic. Chinese writing uses thousands of distinct characters, making movable type far more expensive to set than an alphabet of twenty-six letters. The economic case for movable type printing was much weaker in Chinese than in alphabetic languages. Woodblock printing — where a skilled carver cuts the entire page of text into a wooden block — remained economically competitive in China far longer than in Europe, because the marginal cost of setting type in Chinese was so much higher relative to the baseline.

But the linguistic explanation is incomplete. The deeper answer involves the structure of intellectual life and the rewards to knowledge diffusion. In Europe, the printing press arrived in the context of competing universities, competing religious authorities, competing city-states, and a merchant class with commercial interests in information. There were multiple, competing buyers for printed knowledge, which drove both demand for the technology and the diffusion of ideas through the new medium. In China, the state maintained a much tighter monopoly on legitimate intellectual production. The examination system defined what counted as valuable knowledge. Commercial publishing existed and flourished in certain genres — novels, popular entertainment, practical handbooks — but the diffusion of innovative technical and scientific knowledge through print faced structural barriers that European knowledge did not face.

The contrast is really between a unified imperial knowledge system and a fragmented competitive knowledge system. European fragmentation — which contemporaries often saw as a weakness — turns out to have been a profound strength for innovation. No single authority could suppress an idea across all of Europe. If the Catholic Church condemned Galileo in Rome, his works circulated in Protestant Amsterdam. If one guild in one city refused to license a new technique, a competitor in another city might adopt it. Competition between political units created demand for innovation.

The Needham Question and Its Answers

The British biochemist and historian Joseph Needham spent most of his adult life documenting the history of Chinese science and technology in his monumental Science and Civilisation in China series. His central puzzle, which became known as the Needham Question, was this: given China’s clear technological lead over Europe for most of the first millennium CE, why did the Scientific Revolution occur in Europe rather than China?

Needham’s own answer emphasized the absence of a Chinese merchant bourgeoisie with independent interests in technological innovation, and the dominance of a bureaucratic Confucian culture that was, he argued, fundamentally resistant to the kind of mathematical-mechanical thinking that underlay European natural philosophy. These answers have merit but are incomplete. The Song Dynasty had merchants. The Song period had sophisticated mathematics. Yet it did not produce the Scientific Revolution.

The most compelling modern analysis focuses on incentive structures. Economist Joel Mokyr’s account of the European Enlightenment emphasizes the role of what he calls a “Republic of Letters” — a pan-European community of scholars who competed for reputation by contributing new knowledge, and who had material incentives to publish and diffuse innovations because the prestige from priority claims was real and rewarded. This community operated across national borders and was partly insulated from state and church control by its dispersed, networked character.

China had nothing equivalent. Chinese scholars operated within a state-defined prestige system that rewarded mastery of canonical texts and proximity to state power. There was no independent intellectual community competing for status through novel contributions to natural knowledge, because natural knowledge did not confer status within the system that mattered. The person who discovered a new mathematical technique could not leverage that discovery into examination success, political influence, or court favor. The incentive to invest in original technical research was correspondingly low.

The Mongol conquest of Northern China in 1234 and of the full Song Dynasty in 1279 accelerated the stagnation by destroying the advanced commercial and intellectual infrastructure of the Song period. The Yuan Dynasty and subsequent Ming Dynasty reconstructed a more centralized, less commercially dynamic, and more culturally conservative version of Chinese governance. The inward turn of the Ming Dynasty — exemplified by the abrupt cancellation of the famous Zheng He treasure voyages in the 1430s, when a new imperial faction simply stopped the program and destroyed the records — was an institutional choice to prioritize central control over commercial and technological dynamism.

The Printing Press Asymmetry

The contrast between what the printing press did in Europe and what it failed to do in China is the single most informative data point in the Needham Question. It reveals that technology does not have deterministic consequences; its consequences depend entirely on the institutional and incentive environment in which it operates.

The printing press in Europe amplified competition: between ideas, between religious authorities, between commercial publishers, between cities seeking to attract learned men. It drove down the cost of knowledge diffusion and therefore increased the returns to producing new knowledge, because a novel contribution could now reach an audience of thousands rather than hundreds. It created the first genuine knowledge markets — markets where ideas competed for attention and reputation was the currency.

The printing press in China amplified the existing system: it made it cheaper to reproduce classical examination texts, which were the dominant product of the Chinese publishing industry. It supported the spread of popular fiction, practical agriculture manuals, and commercial almanacs. But it did not create a competitive market for natural knowledge, because no such competitive market was incentivized by the social structure.

The lesson is not about Chinese culture being somehow incompatible with innovation — the extraordinary technological achievements of the Song period demonstrate conclusively that Chinese culture is not the constraint. The lesson is about the relationship between political structure, incentive design, and innovation output. Unified states with monopoly control over prestige and knowledge validation systematically underinvest in original research relative to competitive, fragmented systems where multiple patrons compete to attract innovators.

This is a lesson with clear contemporary relevance. The countries and institutions that maintain independent centers of intellectual authority, that compete for talent through multiple pathways, and that reward original contribution rather than classical mastery are the ones that produce technological revolutions. China’s post-1978 economic rise has been accomplished partly through extraordinary institutional changes that reintroduced competition into Chinese economic life. Whether similar changes to the intellectual and institutional environment for original research will produce a second Chinese technological revolution is the most interesting open question in the economics of innovation.

The Song Dynasty built the closest thing the premodern world had to a modern knowledge economy, and it was extraordinary. The institutions that replaced it chose control over competition, and paid for that choice in centuries of relative technological stagnation. The choice between dynamism and control is never one-time and never final, and every civilization that has faced it has eventually discovered that control is the more expensive option.