Based on Toby Huff, The Rise of Early Modern Science (Cambridge University Press, 1993).
Science must be integrated into the fabric of society — “institutionalized” — if it is to succeed over the longer term. More specifically, scientists must be able to practice and refine their craft, train the next generation of scientists, and be recognized for their work. Science was successfully institutionalized in Europe during the twelfth century (here). China and the Islamic world were at different times the world leader in science, but neither society institutionalized it, and science eventually faltered in both societies.
Institutional Constraints on Science in the Islamic World
Islam’s “golden age of science” was a period of intense activity. It began during the caliphate of al-Mamun (r. 813-833), who supported the collection and translation of Greek philosophy and science into Arabic, and also had works from India and Central Asia collected and translated. These works stimulated Islamic scholarship. Scholars read and organized them, wrote commentaries on them, and ultimately used them as the foundation for their own research. Islamic scholars soon led the world in several fields, including astronomy, mathematics and medicine.
One of the Islamic world’s most significant innovations was the experiment. The Optics of al-Haytham, written in the early eleventh century, was based on simple, clearly defined experiments. It significantly influenced Roger Bacon and other contemporary Western scholars after its translation into Latin in the middle of the twelfth century. The power of medieval experimentation is shown by the independent and nearly simultaneous (about 1304) demonstrations, by Kamal al-Farisi in the East and Theodoric of Freiburg in the West, that the rainbow is the result of two refractions and one reflection inside drops of rain. Islamic scientists also employed experiments in medicine, where al-Razi (d. ca. 925) used controlled experiments to determine the efficacy of herbal treatments, and in astronomy, where predictions were routinely tested against new data.
The golden age ended with a whimper rather than a bang. Islamic science became less adventurous and less innovative after the twelfth century, although significant discoveries continued to be made, particularly in mathematics and astronomy. The astronomer Ibn al-Shatir (d. 1375), for example, constructed a geocentric model of the solar system that was mathematically equivalent to the heliocentric model of Copernicus (d. 1543). Nevertheless, scientific leadership had certainly passed from the Islamic world to the West by the sixteenth century.
Toby Huff argues that the slowdown in Islamic science occurred because scientists were unable to establish their autonomy. Islam’s claim is that it is a complete way of life: nothing is outside of its jurisdiction. This claim impacted scientists in a number of ways.
First, Islamic scientists, unlike their Western counterparts, could not establish any theological wiggle room. Western theologians and natural philosophers believed that the world was created by God according to a plan of his own design, and that he would not now deviate from that plan (here). Every event in the observed world therefore followed from natural (as opposed to supernatural) causes, and natural philosophers were free to make causal claims about these events. The God of Christianity was essentially banished to the margins of his own creation. By contrast, the God of Islam was always and everywhere present. Al-Ash’ari (d. 935) propounded the doctrine of occasionalism, under which the world was held together from moment to moment by the will of God. Effect followed from cause only if God willed it to be so. Occasionalism was still a part of Islamic theology when Al-Ghazali (d. 1111) wrote The Incoherence of the Philosophers, his powerful attack on Aristotelian philosophy as practiced in the Islamic world:
According to us the connection between what is usually believed to be a cause and what is believed to be an effect is not a necessary connection; each of the two things has its own individuality and is not the other, and neither the affirmation nor the negation, neither the existence nor the non-existence of the one is implied in the affirmation, negation, existence or non-existence of the other — for example, the satisfaction of thirst does not imply drinking, nor recovery the drinking of medicine, nor evacuation the taking of purgative, and so on for all the empirical connections existing in medicine, astronomy, the sciences, and the crafts. For the connection between things is based upon the power of God to create them in successive order, though not because this connection is necessary in itself and cannot be disjointed — on the contrary, it is in God’s power to create satiety without eating, and death without decapitation, and to let life persist despite decapitation, and so on with respect to all connection.1
Islam did not draw back to create intellectual room for the philosophers. Aristotle’s view that the world is eternal was an irritant in the West; for Islam, it was an assault on its integrity.
Second, Islamic law did not recognize the possibility of competing jurisdictions. The Quran and the hadiths (the sayings of Mohammed) were held to constitute a complete legal system: nothing could be added to it and nothing could be taken away. A legal scholar could interpret the law when its application in a particular instance was unclear, but unlike English common law, that interpretation did not establish a precedent upon which new law could be built. Islamic law did not include any concept equivalent to the European corporation, and therefore lacked the associated concept of jurisdiction.
Third, and unsurprisingly in light of the legal and theological constraints, the Islamic world did not establish institutions in which scientists could engage in open and free investigation. Its institution of higher learning was the madrasa, which was designed for the training of the religious and legal scholars who were the moral and ethical guides of Islam. Nothing that was inimical to Islam could be taught there. Aristotelian philosophy and the natural sciences were not part of the curriculum, which instead focussed on such subjects as law, Quranic studies, Arabic grammar, and practical arithmetic.
The Islamic golden age produced a line of important scholars who were deeply influenced by Aristotle, including al-Kindi (d. ca. 870), al-Farabi (d. 950), al-Razi (d. ca. 925), Ibn Sina (d. 1037), al-Baghdadi (d. 1152), Ibn Rusd (d. 1198). They supported themselves by working as physicians or judges, or were supported by wealthy benefactors. The option of merging their scholarship with their professional life was not open to them, as it was in the West.
Sabra has argued that Islamic science went through three phases. In the first phase, exposure to foreign scholarship, particularly that of the Greeks, opened new horizons for Islamic scholars. They responded by producing a wide variety of innovative science and philosophy, but there was continuing friction with religious scholars. In the second phase, which Sabra calls naturalization, Islamic scientists became more conscious of the need to harmonize science with Islam. Two important works during this phase were Al-Ghazali’s The Incoherence of the Philosophers and Ibn Rusd’s The Incoherence of the Incoherence. The former work argued that Aristotelian philosophy was inconsistent with Islam, while the latter attempted to refute al-Ghazali’s arguments. Al-Ghazali prevailed in the East, while Ibn Rusd’s collective works strongly influenced Western science. The scope of Islamic science narrowed during this phase. As Sabra explains,
For the religiously committed Ghazali this means, not only that religious knowledge is higher in rank and more worthy of pursuit than all other forms of knowledge, but also that all other forms of knowledge are subordinate to it…Thus, among the non-revealed forms of knowledge: medicine is necessary only for the preservation of health; arithmetic for the conduct of daily affairs and for the execution of wills…There is only one principle that should be consulted whenever one has to decide whether or not a certain branch of learning is worthy of pursuit: it is the all-important consideration that “this world is the sowing ground for the next.”2
The third phase consisted of the long dénouement of Islamic science.
Al-Shatir is credited with having produced in the fourteenth century a geocentric model of the solar system that is mathematically equivalent to the Copernican model of the sixteenth century. Why did he not make the great leap to a heliocentric model? Perhaps the answer is that until the time of Galileo, the astronomer’s role was not to understand the nature of the solar system, but to observe and predict. A model that fit the data was all that was required. And yet this answer seems unsatisfactory. Did al-Shatir never notice that the retrograde motion of Mars would be more easily explained if the earth also moved? Or did he notice, and decide not to pursue an idea that would lead him into conflict with the religious scholars?
Institutional Constraints on Science in China
Francis Bacon (1561-1626) wrote, “Printing, gunpowder and the compass: these three have changed the whole face and state of things throughout the world.” All three of these inventions appeared first in China, along with a great many
others.3 China was an early leader in science as well as technology, but by the eleventh century, it lagged behind both the Arab world and the West in the “core” areas of astronomy, optics, physics and mathematics.4 It lacked knowledge of deductive geometry and trigonometry, despite being in almost constant contact with Indian and Arab scientists. China would not become current in the sciences until the seventeenth century, and even then would not contest the West’s leadership.5
Huff argues that this long period of stagnation occurred because China, like the Islamic world, failed to institutionalize science. Scientists in the Islamic world could not become autonomous because there was only one jurisdiction, that of Islam. Scientists in China could not become autonomous because there was only one jurisdiction, that of the emperor.
Both China and Europe were feudal states at the beginning of the Song dynasty (960-1279), but their political institutions strongly diverged from that point forward. In Europe the legal revolution of the twelfth century led to the development of autonomous institutions and hence to the decentralization of power within each country. The first Song emperor, Taizu, pushed China in the opposite direction, toward greater centralization of power. China remained rigidly centralized through both the Song dynasty and Ming dynasty (1368-1644). The Ming dynasty ended just a few decades before England’s Glorious Revolution, which protected citizens from the depredations of the king and entrenched their right to representative government. In China the emperor remained an absolute monarch.
The first Song emperor, Taizu, ruled a country that had collapsed at the end of the Tang dynasty (907) and had then been reunited through decades of war. He recognized that the regional warlords who had helped to unify the country were now the greatest threat to its unity. Each had an army and a geographic base, and each was a potential rebel. Taizu persuaded the warlords to retire, with generous rewards, and assigned their administrative functions to civil servants. He also ensured that the majority of the new civil servants would be selected on merit, as measured by their performance on civil service examinations.
The civil service subsequently evolved into a complex hierarchy in which information was passed upward and decisions were passed downward. At the bottom of the hierarchy was the district magistrate, who was at once judge, financial officer and sheriff. The next level up from the district was the prefecture. Officials at this level were appointed so that their jurisdictions overlapped. There were also roaming officials who wrote independent assessments of each prefecture’s situation, and censors who had broad powers to investigate the public and private activities of all officials. This web of authority was designed to ensure that no territorial official could threaten the stability of the country by developing an independent power base.
The examination system ultimately became the only way to enter the civil service. Passing the district exams and then the prefectural exams made a man a “cultivated talent.” Passing the provincial exams (offered every three years) made him a “recommended talent.” Finally, passing the exams in the capital made him a “presented scholar” — and only then eligible for a civil service appointment.
The exams were based on the works of Confucius. Preparation for the exams began in childhood, with lessons in calligraphy and writing classical poetry. Children began to memorize the Confucian classics long before they were able to understand them. (Preparation for the exams after 1787 required the memorization of more than 500,000 characters of text.) As children grew older, they began to learn the meaning of the classics and prepare to answer examination questions based upon them. Most candidates received instruction in private academies that followed an officially prescribed curriculum.
The exams were highly stylized and rigidly graded. Since it was not uncommon for a candidate to fail the provincial exams a dozen times, candidates could devote decades of their lives to passing the exams. There was, however, no shortage of candidates because the civil service was one of the few opportunities for upward mobility.
The examination system had some beneficial effects on Chinese society. Its rigour meant that only “the best and the brightest” were recruited into the civil service. It also ensured that all well-educated people — whether or not they attained their goal of becoming “presented scholars” — were imbued with Confucian ethics. Confucianism’s emphasis on traits like loyalty, integrity and benevolence contributed to the stability and harmony of Chinese society.
On the other hand, the examination system probably squandered a significant part of China’s human talent. The system winnowed down the number of candidates until it roughly matched the number of new civil servants required. The winnowing was accomplished by testing the candidates on such things as their ability to compose highly stylized “eight-legged essays.” These skills were essentially uncorrelated with the skills needed to be, say, a successful district magistrate. If the examinations had ended when there were two or three times as many candidates remaining, and if the civil servants had been chosen from this pool by lottery, the quality of the successful candidates would not have been significantly different, and the unsuccessful candidates would have been spared years of unproductive study.6
The dominance of the civil service within Chinese society, coupled with the upward mobility that it offered, made it an attractive option for young and capable men. Candidates for the civil service had a clear path to follow, and there were government-sanctioned academies that existed solely to push them along that path. It must have been difficult for any of “the best and the brightest” to escape the civil service funnel. On the other hand, a career in science would have been difficult to achieve. Exams were occasionally held to fill government positions in mathematics and astronomy, but these exams were not regularly scheduled. There was no established curriculum in the sciences, and there were no institutions of higher learning that could prepare a student for a career in science. The absence of such institutions also meant that there were few opportunities for a career in science outside of the government bureaucracy.
Chinese philosophy posited a connection between heavenly and earthly events. The link between the heavens and the emperor was imagined to be particularly strong, with disorder in the heavens signalling a fault in the emperor’s conduct. Astronomy and timekeeping were therefore matters of state, requiring a high degree of circumspection. The free exchange of ideas that characterized European scholarship did not occur, and unconventional thinking was discouraged.
Scientists in China were simply unable to establish the autonomy required for science to be institutionalized. As in the Islamic world, its bright beginnings gave way to a long period of stagnation.
Unintended Consequences
There was one jurisdiction in the Islamic world because Islam was expected to govern every part of its adherents’ lives. There was one jurisdiction in China because its rulers wished to prevent the rebellions that had previously splintered the country. These choices made sense when they were made. Both choices had, to a large degree, the consequences that they were intended to have. But they also had unintended consequences: both societies would be left on the sidelines when modern science became a dominant force in the world.
- Al-Ghazali, The Incoherence of the Philosophers, quoted in Huff, p. 113. ↩
- I. Sabra, “The Appropriation and Subsequent Naturalization of Greek Science: A Preliminary Statement,” History of Science 25 (1987), p. 239. Quoted in Huff, p. 87. ↩
- Many Chinese technologies, including paper and the heavy plow, diffused into the West, while others were independently developed. Movable type appears first in China, but was independently developed in Europe. Gutenberg’s printing press (which automatically inks the pages) and his moulds for casting type have no eastern antecedent. The compass also appears to have been independently developed in the West. The first records of its use in navigation in the West predate all Arab and Persian references to its existence, so the compass in unlikely to have reached the West through diffusion. ↩
- Huff, p. 242. ↩
- Huff, p. 248. ↩
- The examination system is essentially an example of the rent-seeking problem, in which agents compete for the possession of certain rents, and are able to improve their chances of gaining the rents by outspending their competitors. (For example, the rents could be the benefits associated with some public office, and the candidates could be able to improve their chance of being elected to that office by spending more on advertising.) Under certain circumstances the total amount spent by the agents will be equal to the value of the rents, so that the agents as a group are neither better nor worse off. (In the election, the winner gets the benefits of the office while the losers are stuck with their campaign debts. The total campaign debts are equal to the value of the office.) ↩
Merchants and Mechanics 