Please note: This is a preliminary draft prepared for the conference "Rethinking Science and Civilization: The Ideologies, Disciplines, and Rhetorics of World History," Stanford University, May 21-23, 1999. Not for citation or circulation beyond the purposes of this conference.

"Chinese Sciences" and the "Triumph" of "Modern Science" in China

Benjamin A. Elman, UCLA

For presentation at the Conference on "Rethinking Science and Civilization: The Ideologies, Disciplines, and Rhetorics of World History," organized by Roger Hart and Haun Saussy

Sponsored by the Stanford University Department of Asian Languages and the Program in History and Philosophy of Science

May 21-23, 1999

This paper is about the contested nature of the interaction since 1550 between Chinese and Europeans over the meaning and significance of natural studies. Unlike the colonial environment in India, where British imperial power after 1700 could dictate the terms of social, cultural, and political interaction between natives and Westerners, natural studies in late imperial China were until 1900 part of a nativist imperial project to master and control Western views on what constituted legitimate natural knowledge. Each side made a virtue out of the accomodation project, and each converted the other's forms of natural studies into acceptable local conventions of knowledge. Europeans sought the technological secrets for silk, porcelain, and tea production from the Chinese. Chinese literati borrowed algebra, geometry, trigonometry, and logarithms from the West. Indeed, the epistemological premises of modern Western science were not triumphant in China until the early twentieth century. Until 1900, then, the Chinese interpreted the transition in early modern Europe from new forms of scientific knowledge to new modes of industrial power on their own terms.

Unfortunately, one of the most common generalizations scholars today make concerning the role of "science" (= "natural studies") in late imperial China is that studies of astronomy and mathematics were in steady decline there until the arrival of Jesuit missionaries in the sixteenth century. When Matteo Ricci (1552-1610) described the scientific prowess of Chinese during the late Ming dynasty (1368-1644), he noted that they "have not only made considerable progress in moral philosophy but in astronomy and in many branches of mathematics as well. At one time they were quite proficient in arithmetic and geometry, but in the study and teaching of these branches of learning they labored with more or less confusion." Ricci concluded: "The study of mathematics and that of medicine are held in low esteem, because they are not fostered by honors as is the study of philosophy, to which students are attracted by the hope of the glory and the rewards attached to it."

Chinese mathematics and astronomy, according to this view, had reached their pinnacle of success during the Sung (960-1280) and Yuan (1280-1368) dynasties but had declined precipitously during the Ming (1368-1644). This longstanding perspective has been challenged by recent studies that indicate that mathematics and calendar reform were important concerns among Ming literati before the arrival of the Jesuits in China. Others have demonstrated that the Jesuits misrepresented their knowledge of contemporary European astronomy to suit their religious objectives during the late Ming and early Ch'ing (1644-1911) dynasty. Such self-serving tactics, which produced contradictory information about new trends in European astronomy, lessened their success in transmitting the European sciences to late Ming literati. From this new perspective, late Ming scholars were not lifted out of their scientific "decline" by contact via the Jesuits with European astronomy. Rather, they themselves reevaluated their astronomical legacy, successfully taking into account pertinent features of the European sciences introduced by the Jesuits.

Views that late imperial literati, unlike their Sung and Yuan predecessors, were participants in a strictly humanist civilization, whose elite participants were trapped in a literary ideal that eshewed interest in the natural world, have been common since the Jesuits. Historians have typically appealed for corroboration to the civil examination system. Matteo Ricci, for example, wrote: "The judges and the proctors of all examinations, whether they be in military science, in mathematics, or in medicine, and particularly so with examinations in philosophy, are always chosen from the senate of philosophy, nor is ever a military expert, a mathematician, or a medical doctor added to their number." Catholic scholars were aware of the role played by political and social institutions in Chinese cultural matters, and.the Jesuits realized that the civil recruitment system achieved for Chinese education a degree of standardization and importance unprecedented by early modern European standards.

We know that the examination ethos had carried over for a time into the domains of medicine, law, fiscal policy, and military affairs during the Northern Sung dynasty. For example, Shen Kua (1031-95) wrote that during the Huang-yu reign (1049-53) civil examination candidates were asked to prepare essays on astronomical instruments. The essays were so confused about the celestial sphere, and the examiners were themselves so ignorant of the subject, however, that all candidates were passed with distinction. In addition, we have assumed that the classical curriculum for Ming civil examinations had refocused elite attention on a Tao Learning (Tao-hsueh , i.e., "Neo-Confucian") orthodoxy stressing moral philosophy and literary values and away from earlier more specialized or technical studies. Conventional scholarship still contends that technical fields such as law, medicine, and mathematics, common in T'ang and Sung examinations, were not replicated in late imperial examinations.

When faced with foreign rule (first under the Mongols, 1240-1368, and later under the Manchus, 1644-1911) significant numbers of literati, in addition to the usual number of candidates who failed, turned to occupations outside the civil service such as medicine. In the eighteenth and nineteenth centuries, when demographic pressure meant that even provincial and metropolitan examination graduates were not likely to receive official appointments, many literati turned to teaching and scholarship as alternative careers. Moreover, examiners used policy questions on natural events and anomalies to gainsay the widespread penetration of popular religion and the mantic arts among examination candidates and to keep such beliefs out of politics.

1. Interest in Natural Studies During the Ming Dynasty

"Natural studies" in China had at times since the Yuan dynasty been classified under the phrase ko-chih (lit., "inquiring into and extending knowledge," ko-wu chih-chih ). At other times, particularly in the medieval period, and often simultaneously after the Yuan, such interests were expressed in terms of po-wu (lit., "broad learning concerning the nature of things"). The full mapping out of the asymmetrical conceptual categories associated with these two potential candidates in premodern Chinese for "natural studies" and "natural history" respectively remains incomplete. Moreover we are still unsure how the two terms usually were deployed vis-.-vis each other.

Nonetheless, it appears to me, tentatively, that among post-Sung literati elites ko-chih was the most common epistemological frame for the accumulation of knowledge per se, while po-wu carried with it a more common and popular notion of "curiosities." The Sung philosopher Chu Hsi (1130-1200), who became the core interpreter of the late imperial classical canon, argued for instance that "inquiring into and extending knowledge" presupposed that all things had their principle (wan-wu chih li ). Chu therefore concluded: "one should in three or four cases out of ten seek principles in the outside realm" (san-ssu fen ch'ü wai-mien li-hui fan-k'o ). In most cases, six to seven out of ten, however, moral principles should be sought within.

Since the Southern Sung dynasty (1127-1280), due to Chu Hsi's scholarly eminence, ko-chih became a popular Tao-hsueh (lit., "Tao Learning") term borrowed from the Great Learning (Ta-hsueh ) in the Record of Rites (Li-chi ) classic by literati to discuss the form and content of knowledge. In fact, however, there was much classical debate surrounding Chu Hsi's single-minded prioritizing of the ko-wu passage in the Great Learning to establish the epistemological boundaries for literati learning. Yü Ying-shih 's longstanding claim that the seventeenth century turn among literati elites toward precise philology in classical studies can be traced back to sixteenth century debates surrounding the Old Text version of the Great Learning (Ta-hsueh ku-pen ) deserves mention here. Wang Yang-ming (1472-1528), for instance, preferred the Old Text version of the Great Learning to gainsay Chu Hsi's "externalist" views of the "investigation of things" (ko-wu ) in the Four Books. Subsequently, the delicate issue of the late Ming appearance of an even more ancient "stone inscribed version of the Great Learning" (Ta-hsueh shih-pen ), which was later determined a forgery, reopened for many sixteenth and seventeenth century literati Wang Yang-ming's famous claim that Chu Hsi had in Sung times manipulated the original text of this key passage to validate and make canonical his personal interpretation of the "investigation of things."

In addition to its central epistemological place in literati classical learning since 1200, the term "ko-chih" was also chosen by Ming literati in the seventeenth century as the equivalent native designation for early modern European "science" until its replacement by k'o-hsueh as "modern science" early in the twentieth century among modern Chinese intellectuals. This change from ko-chih to k'o-hsueh as the Chinese equivalent for "science" immediately suggests that the way in which the Chinese themselves chose their native terms for Western "science" was contested at different times and in different ways, which we will explore further below. Early Jesuit translations of Aristotle's theory of the four elements (K'ung-chi ko-chih , lit., "investigation of space," 1633) and Agricola's De Re Metallica (K'un-yü ko-chih , lit., "investigation of the earth," 1640) into classical Chinese, for example, had used the term ko-chih as a general word for the Latin scientia in their titles. Such titles suggest our earlier image of literati intellectual life before the arrival of the Jesuits has been one-sided, and that Tao Learning doctrine and natural studies, particularly medical learning, were not mutally exclusive.

During the early Yuan, ko-chih as a Tao-hsueh term was already used by the medical writer Chu Chen-heng (1282-1358) to denote technical learning. In Chu's most famous work entitled Ko-chih yü-lun (Views on extending medical knowledge), which was included in the Ssu-k'u ch'üan-shu (Complete Collection of the Four Treasuries) in the late eighteenth century, Chu opposed Sung medical prescriptions, but he made a strong appeal to Yuan literati that they should include medical learning in their "studies of the Way." In his view, medical learning was one of the key fields of study that not only complemented the moral and theoretical teachings of Tao-hsueh, but it was also a key to the practical uses (shih-hsueh ) of the latter. The Ssu-k'u ch'üan-shu editors cited Chu's preface as arguing that medicine was one of the concrete fields that informed the "inquiry into and extension of knowledge" ().

Willard Peterson in his study of Fang I-chih (1611-71) has noted how late-Ming views of the Tao-hsueh doctrine of the "investigation of things" (ko-wu , lit., "approaches to phenomena") had changed from a type of moral endeavor to a stress on external things. Fang I-chih's magnum opus entitled Wu-li hsiao-chih (Notes on the principles of things) stressed material investigations to comprehend the seminal forces underlying patterns of natural change. Fang generally accepted Western explanations of natural phenomena , such as a spherical earth, limited heliocentrism, and human physiology, brought to China in the seventeenth century by the Jesuits, but he was critical of them for leaving behind material investigations and ending in unverified religious positions. Fang I-chih favored, instead, descriptive knowledge of the natural world, and he inscribed the Tao Learning interpretation of the "investigations of things" with a new view of the accumulation of knowledge, which gainsaid both the introspective focus of Wang Yang-ming (1472-1528) and the moralist focus of Chu Hsi (1130-1200).

Similarly, the Ming scholar Hu Wen-huan (fl. ca. 1596) compiled the Ko-chih ts'ung-shu (Collectanea of scientia, lit., "collection of works inquiring into and extending knowledge") as a late-Ming repository of classical, historical, institutional, and technical works from antiquity to the present in China that presented all areas of important knowledge to a literati audience in the seventeenth century. The three hundred and forty-six works collected by Hu, some of which he himself had written, were divided into thirty-seven categories (lei ), such as classical instruction, philology, phonology, historical studies, rituals and regulations, legal precedents, geography, mountains and streams, medicine, Taoism, Buddhism, agriculture, stars, physiognomy, poetry and literature, painting, and epigraphy, among others.

Overall, the collectanea emphasized a broad learning of phenomena (po-wu ), one of the thirty-seven categories, that encompassed natural studies within a humanist and institutional agenda. Within the collection, Chang Hua's (232-300) Po-wu chih (A treatise on curiosities), and Li Shih's Sung dynasty continuation, titled Hsu po-wu chih (Continuation to a treatise on curiosities) were subsumed under the general category of ko-chih here. Other works included in the Ko-chih ts'ung-shu were the Shih-wu chi-yuan (Record of the origins of things and affairs) compiled by Kao Ch'eng circa 1078-85, and the Ku-chin shih-wu k'ao (Examination of ancient and contemporary things and affairs) prepared by Wang San-p'in in the Ming dynasty.

In addition to Hu Wen-huan's Ming "Ko-chih studies," Tung Ssu-chang completed the Kuang po-wu chih (Expansion of a treatise on curiosities), which paid more attention to "natural history." Such works on po-wu as "natural history" suggests that as a term po-wu needs to be conceptually mapped asymmetrically with ko-chih. Sometimes the former was included under the latter, sometimes not. In both ko-chih oriented and po-wu framed late-Ming works, the transformation of objects into artifacts, antiquities, and art objects was attempted. From heaven and earth to birds, animals, insects, fish, grasses, foodstuffs, architecture, and tools, the inventory of "scientia" from a Chinese frame of reference represented a systematic collection of data from a wide variety of native sources about China's natural resources, the arts, and manufactures. In the interaction with Western scientia, Chinese literati were drawn into a moderate transformation of their own traditions of natural studies.

Careful scrutiny of Ming dynasty examination records also reveals that civil examinations also tested the candidates' knowledge of astronomy, the calendar, and other aspects of the natural world, which were referred to as "natural studies" (tzu-jan chih hsueh ). The preeminent position of the Four Books and Five Classics was left unchallenged in the orthodox curriculum, but Ming candidates for both the provincial and metropolitan examinations, unlike their Sung counterparts, were expected to grasp many of the technicalities in the calendar, astronomy, and music. Indeed, during the T'ang and Sung dynasties, works on mathematics and astrology had been banned from publication for security reasons. Only dynastic specialists working on the calendar in the astronomy bureau were allowed such knowledge, even though in practice popularly printed calendars and almanacs were widely available.

In the early Ming, for example, the Yung-lo emperor (r. 1402-24) put calendrical and practical studies near the top of what counted for official, literati scholarship. He ordered Hsieh Chin (1369-1415), the chief examiner for the 1404 metropolitan examination (on which 472 graduates drawn from over a thousand candidates were selected and appointed to high offices) to include questions that tested a candidate's "broad learning" (po-hsueh ). Hsieh selected policy questions (ts'e ) dealing with astronomy, law, medicine, ritual, music, and institutions, and the emperor was especially pleased with the top policy answer that year. More importantly, the emperor had legitimated "natural studies." Thereafter such questions regularly appeared on Ming civil examinations.

In the Ming dynasty, although the perennial relationship between classical and historical studies remained the most substantial consideration among literati, the importance of "natural studies" as part of the "broad learning" required of an official had also received imperial support. The demarcation between the universality of classical studies and specialization in practical studies was not called into question, however. Astronomy and calendrical studies, nevertheless, had successfully penetrated the Ming imperial civil service by being required periodically as policy questions on session three of the provincial and metropolitan examinations.

Table 1. Ming Dynasty Policy Questions Classified by Topic: Ying-t'ien Prefecture, 1474-1600, 230 questions, top 15 ranks only.

Rank Topic Pct. Of Total Selection Probability

1 Learning/Selection \ 9.6% 43.4%

2 Tao-hsueh 8.3 37.5

3 Ming rulers , 7.4 33.5

4 World-ordering 7.0 31.6

5 Economy/Statecraft 5.7 25.8

6 Ruler-official 5.2 23.5

7 National defense 4.3 19.4

7 Classical studies 4.3 19.4

9 Law \ 3.5 15.8

9 Military matters 3.5 15.8

11 Literature/Poetry 3.0 13.6

11 Natural studies 3.0 13.6

13 History 2.6 11.8

13 Agriculture 2.6 11.8

13 Customs/Values 2.6 11.8

SOURCE: Nan-kuo hsien-shu (Record of civil examination success in the Southern Capital Region). Compiled by Chang Ch'ao-jui . Ca. 1600 edition.

NOTE: The probability for each policy question is calculated based on the assumption that each of the five selections is mutually independent. If the selection of five questions were mutally dependent, then the probability for each type would be slightly higher. Most "topics" above and below are based on actual Chinese categories. I have added a few, such as "natural studies," which are based on combining categories, such as "astrology," "calendrical studies," and "mathematical harmonics." In the case of "classical studies" versus "philology," which of course are overlapping fields, I have separated them to show the increasing importance of the latter in Ch'ing times (see Table 2).

Table 1, for instance, reveals that candidates for the provincial examinations during the Ming dynasty could reasonably expect a required policy question on astronomy. During any cycle of triennial provincial examinations, a large number of the 50 to over 75 thousand candidates empire-wide likely were prepared to answer them, although we usually only have the best answers. In the Ch'ing, curiously, the likelihood of such policy questions was negligible (see Table 2), which means that candidates increasingly had to answer questions dealing with textual issues growing out of the evidential research studies that peaked in the late eighteenth century. We should quickly add, however, that the ability to deal with astronomical, medical, mathematical, and other technical questions was an essential tool of the new classical studies emerging in the late Ming and early Ch'ing. It just was not tested for the Ch'ing civil service before 1860.

Table 2. Ch'ing Dynasty Policy Questions Classified By Topic: Che-chiang Province, 1646-1859, 460 questions, top 15 ranks only.

Rank Topic Pct. Of Total Selection Probability

1 Classical studies 14.1% 63.7%

2 Learning/Selection \ 10.7 48.4

3 Economy/Statecraft 9.6 43.4

4 World-ordering 7.8 35.3

5 History 7.4 33.4

6 Tao-hsueh 6.1 27.6

7 Literature/Poetry 5.1 23.1

7 Local governance 5.1 23.1

9 Philology 4.2 18.9

10 National defense 3.8 17.2

11 Law \ 3.1 14.0

11 Literati training 3.1 14.0

13 Agriculture 2.7 12.2

13 Military matters 2.7 12.2

15 People's livelihood 2.2 9.9

SOURCE: Pen-ch'ao Che-wei san-ch'ang ch'üan-t'i pei-k'ao (Complete listing of all questions from the three sessions of the Che-chiang provincial civil examinations during the Ch'ing dynasty). Compiled ca. 1860.

NOTE: The probability for each policy question is calculated based on the assumption that each of the five selections is mutually independent. If the selection of five policy questions were mutally dependent, then the probability for each type would be slightly higher. Most "topics" above and below are based on actual Chinese categories. I have added a few, such as "natural studies," which are based on combining categories, such as "astrology," "calendrical studies," and "mathematical harmonics." In the case of "classical studies" versus "philology," which of course are overlapping fields, I have separated them to show the increasing importance of the latter in Ch'ing times (see Table 1).

2. The Elimination of Natural Studies in Early Ch'ing Examinations

The previous discussion demonstrates that it is a mistake to read back into the Ming dynasty the view that Tao Learning moral philosophy and natural studies were opposed to technical learning. Specialized knowledge about astronomy, the calendar, and musical harmonics required in the civil examinations made some difference in the cultural prestige and social status of literati-officials vis-.-vis experts employed in the Astronomy Bureau or the Office of Music. As moral generalists versed in the classical orthodoxy that granted them the highest social, political, and cultural prestige, Ming civil officials were required to know how astronomy, mathematics, calendrical studies, and musical harmonics were part of the orthodox apparatus of ritual. They were not licensed to become "scientists," but neither were they hostile to understanding the role of natural phenomena in governance.

Moreover, the longstanding political raison d're for the literatus had been his official status as a moral paragon who made his classical degree, earned by examination, relevant to his bureaucratic position. Classical statecraft had always been premised on the linkage between classical learning and political competence. That competence was not measured by the literatus' status as an expert in natural studies. Part of it, however, involved using his knowledge of the Classics to understand the role of the calendar or music in governance. In the policy questions, technical learning was not the ultimate object of the question. Rather, the examiners expected candidates to place technical learning within the classical narrative of world-ordering bequeathed by the sage-kings.

Accordingly, the policy questions on natural studies were restricted to fields relevant to bureaucratic governance and discussed in the basic Classics, or at least read into them by the early commentaries. Other fields such as medicine and alchemy were not deemed appropriate for the examination curriculum. It was important that astronomy and mathematics were discussed in the early Classics, while medicine and alchemy were not. The "five phases" and "inauspicious calamities" were cosmological interpretations of the workings of nature within which political governance was rationalized according to the imperial synthesis of classical learning. The "wrong" answer to such policy questions would indicate that the candidate had failed to grasp the heterdox implications of any effort to observe phenomena in the heavens or on earth in ways that challenged the dynasty in power.

As a public event, the policy question and answer delivered in the precincts of an examination compound made "natural studies" part of the orthodox system by placing them, during the Ming, within the civil service examination curriculum. By promoting technical knowledge, the examiners successfully domesticated astronomy, music, and the calendar. Literati were chosen for officialdom in this way precisely because they knew that the moral terms of their success presupposed the subordination of expert knowledge to Tao Learning cultural knowledge that translated via civil examinations into bureaucratic power.

Looking at natural studies in late imperial China from the angle of this cultural hierarchy, which paralleled the social and political hierarchies, we see that such learning was justified as the proper concern of the moral generalist exactly because it could thereby be brought within the orthodox system. Experts, as long as they were subordinate to dynastic orthodoxy and its legal representatives, were necessary parts of the cultural, political, and social hierarchies. The literatus-official coexisted with the calendrical expert in the bureaucratic apparatus but at higher levels of political status, cultural prominence, and social prestige. The Ming civil examinations, therefore, were not remarkable because they included policy questions on natural studies. They were remarkable because they successfully encapsulated natural studies within a system of political, social, and cultural reproduction that guaranteed the long-term dominance of the dynasty, its literati, and the Ch'eng-Chu orthodoxy. Hence, literati did not become "scientists," but they did learn the terms of their reliance on experts.

We have some clues, however, about why in the Ch'ing period such policy questions on natural studies were so rare and uninformed when compared to the Ming dynasty. In the 1660s, for example, a policy question on astrological portents was given on the Shan-tung civil provincial examination. One of the candidates ignorant of the subject tried to disguise his answer by discussing geography instead. Thinking he had failed, the candidate was so surprised to see his name on the list of graduates that he read through the examiner's comments on his answer: "The question was on astrological portents and the answer combined this with discusssion of geography. The candidate can be praised as a learned and refined literatus." Geography and astrological studies had been overlapping fields in earlier dynasties, but during the early Ch'ing this linkage was broken when, as shown below, the court banned policy questions on the calendar and celestial studies. Thereafter, geography, particularly local geography, flourished as a source for provincial and metropolitan policy questions.

Given our present understanding of how the Jesuits used astronomy to take over the Bureau of Astronomy, and the interest early Manchu emperors had in astronomy, we would expect that such influence, as in the Ming, would have carried over to the civil examinations. It is likely that we are here captive of an assumption that natural studies ought to be progressive, and we therefore overlook the likelihood that the Manchu throne sought to monopolize this potentially volatile area of expertise within the confines of the court. The early Manchu promulgation of the civil examinations in the 1650s and 1660s, when, for example, the civil examinations were drastically reformed in 1664 and then unreformed in 1667, and the contemporary calendrical debates between Jesuits and literati-officials, which challenged the Yuan-Ming calendrical system during the Ming-Ch'ing transition, were probably connected and gave the imperial court pause about allowing possibly divisive questions on the calendar to appear in civil examinations.

The collapse of the Ming dynasty and its Ch'ing successor under non-Han rule created opportunities until 1685 for experts in astronomy and music to break out of their subordinate positions and to challenge a discredited Ming elite for political power under a new Manchu ruling elite. The increased cultural importance of astronomical expertise, when the new dynasty had to reformulate in expert terms its calendrical and musical raison d're as quickly as possible, probably outweighed, or at least challenged for a time, the cultural distinction accumulated by literati via mastery of classical studies.

Not until the 1680s, when the Manchu dynasty had mastered its political and military enemies, did the intellectual fluidity of the early decades of the Ch'ing begin to disappear, leaving Han literati and Manchu elites in a precarious balance at the top (and calendar specialists again in the middle or near the bottom) of the political and social hierarchies, which lasted into the eighteenth century. In the process, policy questions on the third session of the provincial and metropolitan examinations virtually ceased to include natural studies. Perhaps the hard-fought court victory of Tao Learning by the 1680s, manipulated by a shrewd Manchu emperor, precluded in civil examinations the successful literati accomodation with the natural studies that had marked Ming civil examinations.

What we do know is that by 1715, the K'ang-hsi emperor (r. 1662-1722) banned focus in the civil examinations on study of astronomical portents and the calendar because they pertained to Ch'ing dynastic legitimacy. The emperor, for example, decreed in 1715 that thereafter all examiners assigned to serve in provincial and metropolitan civil examinations were forbidden to prepare policy questions on astronomical portents, musical harmonics, or calculation methods ( ). The latest works in Ch'ing natural studies, court projects on which the K'ang-hsi emperor had employed Jesuit experts, were put off limits to examiners and examination candidates. The ban on natural studies was stipulated within a general effort by the court to keep the mantic arts and discussion of auspicious versus inauspicious portents out of public discussion.

This evolving Ch'ing ban on Chinese literati studying astronomy, astrology, and music openly was noted at the time in Shen Hsin-chou's 1712 preface to his study entitled Ti-hsueh (Studies of geography). Shen indicated that all discussions of astronomical portents (yen t'ien-wen ) were forbidden late in the K'ang-hsi reign. In this public acknowledgement of Ch'ing imperial policy, we see by way of contrast how important the Yung-lo emperor's early Ming decree had been in encouraging natural studies when he demanded such policy questions on examinations in 1404. Likewise, we can understand how the K'ang-hsi emperor's ban affected Ch'ing intellectual life. What seems counterintuitive, namely that the Ming dynasty encouraged while the Ch'ing discouraged natural studies on the civil examination curriculum, makes historical sense. In place of the banned natural studies, historical geography in particular prospered as an acceptable field of Ch'ing scholarship, although map-making was kept secret by the throne. The Yung-cheng emperor also changed the K'ang-hsi emperor's policy a bit by admitting imperial students with specializations in astrology (t'ien-wen-sheng ) into the dynastic schools. Both emperors and their literati officials were responsible for curricular changes in civil examinations, emperors through their private concerns and examiners through their scholarly interests.

3. Evidential Studies and Ko-chih-hsueh

Such bans, however effective in the examination system, did not carry over to literati learning, where a decisive sea change in classical learning was occurring. In contrast to their Tao-hsueh predecessors, Ch'ing "evidential research" (k'ao-cheng) scholars stressed exacting research, rigorous analysis, and the collection of impartial evidence drawn from ancient artifacts and historical documents and texts. Evidential scholars made verification a central concern for the emerging empirical theory of knowledge they advocated, namely "to search truth from facts" (shih-shih ch'iu-shih). This program involved the placing of proof and verification at the center of the organization and analysis of the classical tradition in its complete, multidimentional proportions, which now included aspects of natural studies and mathematics.

The seventeenth-century research contributions made by Ku Yen-wu (1613-82) and Yen Jo-chü (1636-1704) in the formation of evidential studies were continued in the eighteenth- and nineteenth-century philological research of Tai Chen , Ch'ien Ta-hsin , Tuan Yü-ts'ai (1735-1815) , Wang Nien-sun (1744-1832), Wang Yin-chih (1766-1834), Chiao Hsun (1763-1820), and Juan Yuan (1764-1849). Philological studies developed and evolved during the eighteenth and nineteenth centuries because their published works were parts of a dynamic classical research enterprise whose goals were not "scientific" or "objective" per se but instead were tied to a new literati commitment to use the language of the ancient Classics as an impartial means to recapture the ideas and intentions of the sage-kings of antiquity. Even if they were scholarly iconoclasts in their own time, they still were firmly conservative in their social beliefs and commitments.

By the late eighteenth century, reflecting the scholarly trends of the Ch'ien-lung era (1736-95), the policy questions for civil examinations began to exhibit a common five-way division of topics, usually in the following order: 1) Classics; 2) Histories; 3) Literature; 4) Statecraft; and 5) Local geography. The primacy of classical learning in the policy questions was due to the impact of Han Learning and evidential research among literati scholars, first in the Yangtzu delta, and then empire-wide via examiners from the delta provinces of Chiang-su, Che-chiang, and An-hui. What was fueling the popularity of the revival of first a poetry question on session one and then philology in the policy questions in session three of the civil examinations was the close tie between the rules for rhyming in regulated verse and the field of phonology, which became the queen of philology during the Ch'ien-lung reign. The role of phonology in evidential research studies was paying dividends by improving literati knowledge of classical sounds and rhymes.

As Hamaguchi FujiÙ shows, Ch'ing dynasty evidential scholars such as Tai Chen had in mind a systematic research agenda that built on paleography and phonology to reconstruct the meaning (i yin ch'iu i ) of Chinese words. Later Wang Nien-sun, and his son Yin-chih, extended Tai's approach and attempted to use the "meanings" of Chinese words as a method to reconstruct the "intentions" of the sages, the farsighted authors of those words. Moreover, technical phonology when applied to the study of the history of the classical language reached unprecedented precision and exactness. To achieve this end, evidential scholars chose philological means, principally the application of phonology, paleography, and etymology, to study the Classics.

One byproduct of these philological trends was the full realization of how important poetry, particularly regulated verse, was for the reconstruction of antiquity via phonology, paleography, and etymology. For example, Liang Chang-chü (1775-1849), who assembled one of the first cultural studies of the examination regime entitled Collected Comments on the Crafting of 8-legged Civil Examination Essays (Chih-i ts'ung-hua ) in the early nineteenth century, also compiled a collection on poetry studies in which he outlined the study of poetry and the rules of regulated verse. In the conclusion, Liang traced how Qing classical scholars had finally unraveled the rhyme system of the Poetry Classic. They had thereby illuminated the technical rules in regulated verse and made major advances in the study of phonology.

A full-blown scientific revolution as in Europe did not ensue, but k'ao-cheng scholars made astronomy, mathematics, and geography high priorities in their research programs, another by-product of the sea change in classical studies then underway. Animated by a concern to restore native traditions in the precise sciences to their proper place of eminence, after less overt attention during the Ming dynasty until the coming of the Jesuits in the sixteenth century, evidential scholars such as Tai Chen, Ch'ien Ta-hsin, and Juan Yuan successfully incorporated technical aspects of Western astronomy and mathematics into the literati framework for classical learning . Ch'ien Ta-hsin, in particular, acknowledged this broadening of the literati tradition, which he saw as the reversal of centuries of focus on moral and philosophic problems: "In ancient times, no one could be a literatus (Ju ) who did not know mathematics . . . Chinese methods [now] lag behind Europe's because Ju do not know mathematics."

The impact of evidential research made itself felt in the attention k'ao-cheng scholars gave to the Western sciences of mathematics and astronomy first introduced by the Jesuits in the seventeenth century. Such interest had built upon the early and mid-Ch'ing findings of Mei Wen-ting (1633-1721), who was sponsored by the Manchu court once his expertise in mathematics and calendrical studies was recognized. Mei had contended that study of physical nature gave scholars access to the "principles" (li ) undergirding reality. In essence, Mei saw Jesuit learning as a way to enhance the mathematizing of the Tao-hsueh notion of moral and metaphysical principle. At the same time, however, the imperial court and Mei Wen-ting prepared preliminary accounts stressing the native Chinese origins (chung-yuan ) of Western natural studies. Such origins made it imperative, Mei and his highly placed follower in the early Ch'ing court Li Kuang-ti (1642-1718) thought, to restore and rehabilitate the native traditions in the mathematical sciences to their former glory. Under imperial patronage during the K'ang-hsi reign mathematical studies were upgraded from an insignificant skill to an important domain of knowledge for literati that complemented classical studies.

For example, Ch'en Yuan-lung's (1652-1736) Ko-chih ching-yuan (Mirror origins of scientia), was published in 1735, and in the 1780s it was included in the Imperial Library. A repository of detailed information divided into thirty categories culled from a wide variety of sources, the Ko-chih ching-yuan represented a post-Jesuit collection of practical knowledge that narrowed the focus of Hu Wen-huan's late-Ming Ko-chih ts'ung-shu to cover almost exclusively the arts and natural studies. Special attention was given to the origins and evolution of printing and stone rubbings, in addition to topics dealing with geography, anatomy, flora and fauna, tools, vehicles, weapons and tools for writing, as well as clothing and architecture.

The seventeenth century impact of Jesuit knowledge in China was not always so easily domesticated in the eighteenth, however. Literati scholars took a range of positions concerning natural studies. A private scholar, Chiang Yung (1681-1762), for instance, combined a classical loyalty to Chu Hsi's Tao-hsueh teachings with knowledge of Western Jesuit studies obtained through evidential studies. Conservative as a classical scholar, Chiang was quite radical in his critique of both Han Learning and Mei Wen-ting in natural studies for exalting nativist ancient studies in all cases. Chiang Yung recognized the advantages Western astronomy had over native traditions, while at the same time he continued to uphold the cultural superiority of the Tao-hsueh view of morality. Although Chiang preferred Western learning for understanding the "principles" of nature because they were more precise and consistent than native traditions,he maintained a clear distinction between astonomical methods and cultural values.

Overall, Juan Yuan's compilation of the Ch'ou-jen chuan (Biographies of astronomers and mathematicians) while serving as governor of Che-chiang province in Hang-chou from 1797 to 1799, reprinted in 1849 and later enlarged, marked the climax of the celebration of natural studies within the Yangtzu delta literati world of the eighteenth century, which had been increasing since the late seventeenth century. Containing biographies and summaries of the works of 280 ch'ou-ren, including thirty-seven Europeans, this work was followed by four supplements in the nineteenth century. Limin Bai has noted how the mathematical sciences had begun to grow in importance among literati beyond the reach of the imperial court in the late eighteenth century. They were now linked to classical studies via evidential research. Because Juan Yuan was a well-placed literati patron of natural studies in the provincial and court bureaucracy, his influential Ch'ou-jen chuan represented the integration of the mathematical sciences with evidential studies. Mathematical study was no longer independent of classical studies.

Such evidence from the eighteenth century gainsays the usual conclusions that Western and Chinese scholars have drawn concerning the "failure" of the Macartney mission to open China to European science and trade in 1793. Joanna Waley-Cohen has reevaluated the Ch'ing dynasty's so-called "blindness" to world developments in the eighteenth century and revealed how this erroneous assessment grew out of Western technological superiority after the nineteenth-century industrial revolution, which was then read back into the 1793 Macartney mission to China by later historians and diplomats. This misassessment in Western attitudes toward China, Waley-Cohen argues, was also due in part to the Ch'ing court's need under the Ch'ien-lung emperor to reassert the "public Chinese attitude of superiority toward foreigners" in the factionalized court politics of 1793, even though the emperor at the same time avidly employed Jesuit experts in the arts of warfare for late-eighteenth-century military campaigns against rebels within the empire.

The earlier "modernization narrative" that described British imperial expansion colliding with a sinocentric Ch'ing state unsympathetic with the needs of scientific knowledge should be amended. Furthermore, the Ch'ien-lung emperor's famous letter to George III gainsaying Western gadgets should not be read as the statement of a Manchu empire completely out of touch with historical reality. James Hevia has recently shown how misreadings in European circles of the Ch'ien-lung emperor's letter to the English king resulted when later scholars essentialized Chinese diplomacy using a European lens to view the so-called "stagnant" and "out-of-date tribute" system. When contextualized, the emperor's reaction to the mission can be understood in light of the mutual misunderstandings that swelled from the overstated claims Macartney made about the pre-industrial revolution gadgets cum gifts--a replica of the solar system, for example--he had brought for the emperor (who did not think the planetarium so fabulous). Later emperors who found English military firepower irresistible in the 1839-42 Opium War and thereafter were dealing with a different set of technological circumstances.

As Waley-Cohen also shows, the emperor's letter was not a categorical rejection of Western technology, which has become the standard interpretation. Coming before the industrial revolution, the scientific trinkets the Macartney mission brought to China were contested by the court. Moreover, the Qing court welcomed the advice of Jesuits in their midst concerning cannon-building and empire-wide cartography. That literati scholars had incorporated mathematical study into evidential research and made natural studies a part of classical studies is another piece to the puzzle concerning the fate of the sciences and technology in late imperial China since the Jesuits first made their presence felt in the seventeenth century.

4. The High Tide of "Ko-chih-hsueh" as Science in the Nineteenth Century

Philology and natural studies were wedded together when Ch'ing literati scholars such as Mei Wen-ting and his grandson Mei Chueh-ch'eng (d. 1763) evaluated early modern European findings in astronomy and searched through the classical canon for evidence that this new knowledge was likely based on ancient Chinese knowledge, which had been transmitted to the Western regions in antiquity. Mei Chueh-ch'eng contended, for instance, that the Sung-Yuan T'ien-yuan (heavenly origins) method for graphically representing algebraic equations was the equivalent of the algebraic formulas later introduced by the Jesuits. This more sophisticated "Chinese origins" (Chung-yuan ) argument, when compared to that of the K'ang-hsi emperor, legitimated renewed Ch'ing literati interest in the sciences, and philology became one of the key tools later evidential research scholars employed.

The mathematics associated with evidential research in the eighteenth century had been algorithmic, i.e., focusing on getting the right results, and thus was less concerned to justify methods and formulas. Wang Lai (1768-1813) and Chiao Hsun, for example, each tried to build on traditional Chinese algebraic equations, known as T'ien-yuan, rather than just automatically accept the Indic-Arabic forms of algebra that the Jesuits and later the Protestants taught when they came to China. Wang in particular derived more than one positive root for a T'ien-yuan equation, which by following Western views of positive and negative roots contributed something new to the traditional focus on a single, positive solution for any algebraic equation.

Wang Lai, who was appointed to the dynastic observatory in Peking, employed Western methods long accepted in the calendrical office since the K'ang-hsi reign, in his calculations of T'ien-yuan. As a result of his professional ties to the Jesuit "new studies" harbored in the observatory, Wang was criticized by more conservative k'ao-cheng scholars interested in traditional mathematics for going too far in his emulation of Western methods. Because he was a literatus outside the court and thus tied to the Yangtzu delta academic community, Li Jui (1773-1817), who devised a theory of T'ien-yuan equations strictly in terms of Sung T'ien-yuan mathematics, received more support from literati, many of whom still revered Yang Kuang-hsien (1597-1669) for his prosecutions of the Jesuits in the K'ang-hsi court in the 1660s. Before 1850, then, classical learning still took precedence over Western learning, and the antiquarian interests of evidential scholars stimulated them to study the textual history of native mathematics rather than build on the findings of Western mathematics, as Wang Lai had.

In the aftermath of the bloody defeat of the Taipings, however, a weakened Ch'ing dynasty and its literati-officials began to face up to the new educational requirements the civil service would have to fulfill to survive in a world increasingly filled with menacing industrializing nations. While the Opium War (1839-42) had not provoked any important calls for introduction of "Western learning" into the civil service curriculum, the situation after the fall of the Taipings was remarkably different. Literati such as Hsu Shou (1818-82) and Li Shan-lan (1810-82) began to build conceptual bridges between post-industrial revolution Western learning and the traditional Chinese sciences in the middle of the nineteenth century. In the process, post-industrial revolution Western science, now called "modern science," was initially introduced in the mid-nineteenth century as compatible with but no longer subordinate to native classical learning.

Both Feng Kuei-fen (1809-74), a Hanlin academican, and Hsueh Fu-ch'eng (1838-94), who was prevented by the Taiping wars from taking civil examinations, became administrative experts and advisors to many of the chief ministers of the late Ch'ing, including Tseng Kuo-fan (1811-72) and Li Hung-chang (1823-1901), the leaders of the post-Taiping turn toward foreign studies (yang-wu yun-tung ). The classical curriculum needed to adapt more Western learning and science subjects to be viable, they claimed. Western models became a legitimate object of concern and debate to reform the civil examinations. Li Hung-chang, for example, followed Feng Kuei-fen's recommendation and in 1863 established a school of Western languages and science in Shanghai, which was added to the Chiang-nan Arsenal in 1869. Li also proposed establishing eight categories for civil examination (pa-k'o ch'ü-shih ) in 1867, which included "mathematical science" (suan-shu ko-chih ) and "technical science" (chi-ch'i chih-tso ) as a single category.

To attain the goal of "wealth and power" (fu-ch'iang ), which Ch'ing literati and officials became obsessed with in the last decades of the dynasty, Feng Kuei-fen, for example, while living in the treaty port of Shanghai to avoid the Taipings had prepared an essay around 1861 entitled "Kai k'o-chü i" (Proposal for reforming the civil examinations) in which he attempted to balance the strengths of the selection process with the needs of the future. What was now needed was a revamped civil examination that would test more challenging knowledge than poetry or prose writing. Feng proposed three new sessions for the provincial and metropolitan examinations: 1) explication of the classics; 2) policy and discourse questions; and 3) ancient studies.

At first sight, these recommendations seem traditional, but Feng was aware that he had to sell the changes to opponents who would oppose any blatant effort to introduce Western learning into the examination curriculum. Instead, Feng altered the content of native traditional fields. What Feng meant by classical studies, for instance, included evidential research (k'ao-chü ) and philology (hsiao-hsueh ), subjects of learning already included in provincial and metropolitan policy questions. In addition, he added mathematics to the field of classical studies and quietly relegated the literary essay and poetry question to the last session.

Feng also called for widening the selection process for officials to include recommendation and the promotion of clerks who demonstrated their administrative abilities to their superiors. One way to do this, according to Feng, was to divide the civil examination system in two, with one group required to master machinery and physics (chih-ch'i shang hsiang ). Based on such reforms, "our China [Chung-hua ] can begin to arise in the world." Otherwise, Feng presciently predicted, she will be a victim of native militarists hiding behind the slogan of "self-strengthening" (tzu-ch'iang ).

In other essays composed circa 1861, Feng also called for increased specialization of knowledge among officials and more attention to Western studies. Feng made clear that the world had changed so drastically that models from antiquity were no longer appropriate to present conditions. And one of the strengths of Western learning, he noted, was its mastery of mathematics, which Feng did wish to incorporate into the civil examinations. Geography and calendrical studies, the latter banned in dynastic schools and civil examinations since the K'ang-hsi reign, were also essential fields for literati, Feng contended. Not until 1887, however, were candidates specializing in mathematics allowed to pass the provincial examinations under a special quota, although they also had to fulfill the same classical requirements.

Literati learning in the nineteenth century developed rhetorical strategies to legitimate elite interest in natural studies. In the process, by building on eighteenth century classicism, which had incorporated mathematics as a part of evidential studies, literati associated with Han Learning after the Taiping Rebellion created the intellectual space needed for literati to indulge in natural studies. For instance, Li Shan-lan went to Shanghai in 1852 and for eight years there worked for the London Missionary Society to translate Western science works into classical Chinese. Later, Li was recommended to the newly established T'ung-wen-kuan (Translators bureau) in 1864, but he took up the appointment in 1866 only after the T'ung-wen-kuan was upgraded to a college and a department of mathematics and astronomy was added. There, Li Shan-lan worked with W. A. P. Martin (1827-1916), who served as president of the college from 1869 to 1882, to teach mathematics and prepare scientific translations.

Hsu Shou initially collaborated with John Fryer (1839-1928) at the Chiang-nan Arsenal in Shanghai to translate Western scientific literature into classical Chinese, an enterprise that combined a narrow vision of science, brought by Protestant missionaries to attract Chinese converts, with the skewed k'ao-cheng view of the sciences as a domain of classical studies appropriate only for literati. Hsu, like Fryer, in effect remained for the most part a cultivator not a researcher of science. Together they founded the Ko-chih shu-yuan in Shanghai in 1874, which was curiously translated into English as the "Shanghai Polytechnic Institute." From different sides, Chinese literati and Western modernizers saw in ko-chih what they wanted to see, a nativist trope or modern science. The Institute had a reading room and library of scientific works. Hsu and Fryer also created the first science journal in China entitled Ko-chih hui-pien , known in English as "The Chinese Scientific Magazine," which ran weekly issues in Shanghai from 1862 to 1883 and reached some 2000 readers in the treaty ports. Such conceptual compromises were based on maintaining the post-Jesuit term for scientia (= ko-chih-hsueh), but this time using it to refer to modern, not early modern, science, and successfully made mathematics and the other more industrial sciences such as chemistry into proper, if still less popular than the civil service, activities for literati.

5. The Denigration of Traditional Chinese Science By Champions of K'o-hsueh/Kagaku

Despite the relative success of natural studies and Western science in developing together from the seventeenth to the late nineteenth century among literati elites in China under the rubric of ko-chih-hsueh, there was little attention by those same elites to European science as a form of practice requiring laboratories to replicate experiments and for such experiments to confirm or reject past scientific findings. For Catholic or Protestant missionaries and literati mathematicians, natural studies was little more than translation of technical knowledge, mastery of those technical texts, and the reproduction via memory of technical learning. Moreover, those who were drawn after the Taiping Rebellion to scholarly work in the new arsenals in Fu-chou, Shanghai, and elsewhere, or translation positions in the T'ung-wen-kuan, tended to be marginal literati such as Hsu Shou and Li Shan-lan, men who had failed the more prestigious civil examinations several times and saw Western learning and the sciences as an alternative, but still secondary, route to fame and fortune. Yen Fu (1853-1921) and Lu Hsun (1881-1936) were also famous examples of this alienated group of outcasts from the civil examinations that initially served as the pool of highly educated men who filled the more limiting world of late-Ch'ing instititions oriented toward ko-chih-hsueh.

Recent research indicates, however, that the various arsenals, shipyards, and factories in the treaty ports were important technological venues for experimental practice where, in addition to the production of weapons, ammunition, and navies, a union of scientific knowledge and experimental practice among literati and artisans was first forged. Indeed, it is likely the case that the "techno-science" of late-Ch'ing China was an important building block for the rise of both dynastic and private industry in the late nineteenth century treaty ports where most of the arsenals were established. The Chiang-nan Arsenal and Fu-chou Shipyards, for instance, were acknowledged by Europeans and Japanese to be more advanced than their competitor in Meiji Japan, the Yokosuka Dockyard, until the 1880s. Accordingly, outside the civil examination regime and its precincts of licentiates, chü-jen (lit., "raised candidates," i.e., provincial graduates), and chin-shih (literatus presented to the emperor for appointment), where millions competed for few places in the bureaucracy, a notable group of doctors, nurses and medical assistants were trained in missionary schools, and an even larger group of military technicians and technical specialists were instructed in the arsenals and shipyards.

It was not until the Sino-Japanese War of 1894-95, when the Japanese navy, which was tied to Yokosuka technology, decisively defeated the Ch'ing navy, which was tied to Fu-chou and Shanghai technology, that the alleged superiority of Japan in military technology, or so it was interpreted, became common knowledge to Chinese and Japanese patriots. Although most thought the Kiangnan Arsenal was superior in science and technology to Yokosuka until the 1880s, after 1895 each side then read their different fates in 1895 teleologically back to the early Meiji period (later even back further to Rangaku "Dutch Learning"), in the case of triumphant Japan, or back to the failures of the self-strengthening movement after 1865, in the case of the defeated Ch'ing. Another sea change in elite and popular opinion in China now determined how the Chinese refraction of Western science and technology through the lens of ko-chih-hsueh would be interpreted after 1895. Literati radicals such as Yen Fu declared that accomodation between Chinese ways and Western institutions had failed. The Sino-Japanese War thus altered the frame of reference for the 1860-1895 period for both Chinese and Japanese. The beginnings of the "failure narrative" for Chinese science, which parallelled the story of political and economic decline of the late empire.

Yen Fu, whose poor prospects in the civil examinations led him to enter the School of Navigation of the Fu-chou Shipyard in 1866, expressed long pent up bitterness toward the civil examinations, when he became a publicist and prepared articles for the reformist press that emerged after 1895. Since 1885, Yen had failed the provincial examinations four times. Many like Yen Fu began in the 1890s to link the weakness of the Ch'ing dynasty to the classical education required in civil examinations, which allegedly had wasted the minds of generations. Moreover, Yen and other reformist voices associated the power of the West with modern schools where students were trained in modern subjects requiring practical training in the sciences and technology.

For Yen Fu and the reformers, Western schools and Westernized Japanese education were examples that the Ch'ing dynasty should emulate. The extension of mass schooling within a standardized classroom system stressing science courses and homogenious or equalized groupings of students seemed to promise a way out of the quagmire of the imperial examination regime, whose educational efficiency was now, in the 1890s, suspect. Uncritical presentations of Western schools and Japanese education as success stories were widely accepted. Those involved with the 1898 Reform Movement (Wu-hsu pien-fa ) contended that political reform required fundamental educational change, and educational change was possible only if the civil examinations were reformed.

One of the products of the iconoclasm of the 1898 reforms that survived the Empress Dowager's coup was the Imperial University of Peking, which was established to be at the pinnacle of an empire-wide network of schools that would expand on the T'ung-wen-kuan. The new university was designed like the Translation College to train civil degree-holders, i.e., literati, in Western subjects suitable for government service. W.A.P. Martin, who had earlier worked with Li Shan-lan, was chosen as the dean of the Western faculty. Science courses at the Imperial University, interestingly, were still referred to as ko-chih, and the facilities included modern laboratories equipped with the latest instruments for physics, geometry, and chemistry. This promising development was short-lived, however, because rebels associated with what was called the "Boxer Rebellion" smashed everything in sight at the university in the summer of 1900.

During the chaos of the Boxer Rebellion in Peking, the Ch'ing court forgot the lessons that earlier emperors and literati had understood very well. The delegitimation of classical learning, once complete, eventually had consequences that went beyond what the court and literati expected. The race to establish new institutions of higher learning that would stress modern science accelerated after the occupation of the capital by Western and Japanese troops in 1900. The Boxer popular rebellion and the response of the Western powers and Japan to it unbalanced the power structure in the capital so much that foreigners were able to put considerable pressure on provincial and national leaders. Foreign support of reform and Western education thus strengthened the political fortunes of provincial reformers such as Yuan Shih-k'ai (1859-1916) and Chang Chih-tung (1837-1909), who had opposed the Boxers.

The story of the demise of traditional natural studies and the rise of modern science in China was more complicated than just the demise of classical learning and the rise of modern education, which would subordinate the classics to science. A social, political, and cultural nexus of classical literati values (within which natural studies were embedded), dynastic imperial power, and elite gentry status was unraveling . The Ch'ing dynasty became a party to the delegitimation of classical studies and the accompanying rethinking of the nature and scope of ko-chih-hsueh vis-.-vis modern science. By first decanonizing the classical canon, late nineteenth century literati hoped to free themselves from the moral and classical imperatives of the "examination life," but they also began to distance themselves from traditional views of and approaches to natural studies, medicine, and technology.

The delegitimation of classical learning after 1900 initially did not challenge the use of ko-chih as a term from the Four Books to translate modern science into classical Chinese, however. Session one of the reformed reformed civil examinations, for example, expected candidates to answer five questions dealing with Chinese institutions and politics. Session two included five policy questions on Western institutions and politics. The last session required three classical essays, two on quotations from the Four Books and one from the Five Classics. In theory, all three sessions were expected to count equally for the final rankings, but how this would work out in practice remained unknown. Would examiners really relegate classical essays and give priority to contemporary issues? And what role would the sciences play in this reform?

Format of Provincial and Metropolitan Civil Service Examinations during the Late Ch'ing Dynasty, after the 1901 Reform (abolished in 1905)

Session No. No. of Questions

One

1. Discourses on the history of 5 essays

Chinese politics

Two

1. Policy questions on world politics 5 essays

Three

1. Four Books 2 essays

2. Five Classics 1 essay

NOTE: The requirement that all candidates specialized (chuan-ching ) on one of the Five Classics was dropped in 1787. In addition, such essays no longer had to follow the "eight-legged" style.

In 1902 the first civil examinations since enactment of the post-Boxer reforms took place in K'ai-feng , the capital of Ho-nan province. Because the provincial examination halls in Shun-t'ien, where the metropolitan examinations in Peking had also been held, had been burned down by the foreign troops sent in to relieve the Boxer siege of the international legations, the metropolitan examination could not be held in Peking. The 1902 examination reforms failed in the short run to accomplish their goals because of the tenacity of examiners. Nevertheless the overall scope of the examinations became decidedly more institutional and international in focus. A catalog of policy questions used in the examinations after the reforms, which was compiled in 1903, identified the thirty-two categories that were used:

1. Way of ordering 17. Mathematics

2. Scholarship 18. Sciences (I) ()

3. Domestic government 19. "" (II) ()

4. Foreign relations 20. State finance

5. Current affairs 21. Monetary system

6. Civil examinations 22. Military system (I) ()

7. Schools 23. "" (II) ()

8. Official institutions 24. Defense matters

9. Assemblies 25. Agriculture system (I) ()

10. State organization 26. "" (II) ()

11. Public laws 27. Public works

12. Penal laws 28. Commercial system

13. Education affairs 29. Roads & mines

14. Astronomy 30. Topography

15. Geography 31. History

16. Calendrical studies 32. Foreign history

Although the examiners' biases toward "Chinese learning" pervaded many of these fields, the impact of the Western sciences on the civil service curriculum was quite noticeable. For example, five of the eight questions on the natural sciences, which was still called ko-chih-hsueh, were phrased as follows:

1. Much of European science originates from China (Chung-kuo ); we need to stress what became a lost learning as the basis for wealth and power.

2. In the sciences, China and the West (T'ai-hsi ) are different; use Chinese learning (Chung-hsueh ) to critique Western learning (Hsi-hsueh ).

3. Substantiate in detail the theory that Western methods all originate from China.

6. Prove in detail that Western science studies mainly were based on the theories of China's pre-Han masters.

7. Itemize and demonstrate using scholia that theories from the Mohist Canon preceded Western theories of calendrical studies, light, and pressure.

Such views revealed that in official terms, the wedding between the traditional Chinese sciences and Western science, worked out beginning in the eighteenth century, was still in effect. Publicly at least, the officials of the late-Ch'ing dynasty maintained the fiction that "the Western sciences for the most part derived from the teachings of the pre-Han masters" ().

After 1905, however, when the civil examinations had been abolished, the ever increasing number of overseas Chinese students in Japan, Europe, and the United States perceived that outside of China the proper language for science included a new set of concepts and terms that superseded traditionalist literati notions of natural studies associated with ko-chih. For example, Japanese scholars during the Meiji period had already in the 1870s demarcated the new sciences by referring to wissenschaft as kagaku (k'o-hsueh ; lit., "classified learning based on technical training") and natural studies as kyûri (ch'iung-li , lit. "exhaustively study the principles of things"). The latter term, long associated with Tao Learning in China since the Sung dynasty, was reinterpreted in Japan based on the Dutch Learning tradition of the late eighteenth century, when Japanese scholars interested in Western science still used terms from Chinese learning (Kangaku ) to assimilate European natural studies and medicine.

Chinese students and scholars initially adopted the Japanese bifurcation between technical learning and natural studies. Yen Fu, for instance, rendered the terms "science" or "sciences" as k'o-hsueh in his 1900-02 translation of John Stuart Mill's System of Logic, while translating "natural philosophy" as ko-wu , or the "investigation of things." Similarly, when regulations for modern schools were promulgated in 1903, the term ko-chih was used to refer collectively to the sciences in general, while the sciences as individual, technical disciplines were designated as k'o-hsueh. This two-track terminology for science lasted through the end of the Ch'ing dynasty and was continued during the early years of the Republic of China, but Chinese students who returned from abroad increasingly saw the need to develop a single Chinese term for the Western sciences that would leave behind the earlier assimilation of traditonal Chinese natural studies into modern science.

Many overseas students were as radical in their political and cultural views, which carried over to their scientific iconoclasm. Traditional natural studies became part of the "failed" history of traditional China to become "modern," and this view now included the claim that the Chinese had never had any science. The "Chinese origins" of Western science position, so prominent before 1900, was now deemed "superstition" . What had come before was magic and the supernatural, not science. How premodern Chinese demarcated the natural and supernatural was lost, when both "modernists" and "socialists" in China accepted the West as source of all science

Linkage between political revolution and the perception by many radicals that a scientific revolution was also required influenced the changes that occurred after 1911. Those Chinese who thought a revolution in knowledge based on Western learning was required not only challenged what they called "Confucianism" (K'ung-chiao ), but they also unstitched the interwoven patterns of traditional Chinese science, medicine, and classical learning long accepted as components of an ideological tapestry butressing imperial orthodoxy. Those educated abroad at Western universities such as Cornell University or sponsored by the Rockefeller Foundation after 1914 for medical study in the United States, as well as those trained locally at higher-level missionary schools, regarded modern science as k'o-hsueh , not ko-chih-hsueh, because they believed the latter term was derived from the language of the discredited past and inappropriate for modern science. The belief that Western science represented a revolutionary application of scientific methods and objective learning to all modern problems was increasingly articulated in the journals associated with the New Culture Movement. The journal K'o-hsueh (Science), which was published by the newly founded Science Society of China (Chung-kuo k'o-hsueh-she ) and first issued in 1915, assumed that an educational system based on k'o-hsueh was the panacea for all of China's ills because its knowledge system was superior. By 1920, the Science Society, which had been founded by overseas Chinese students at Cornell in 1914, had some 500 members in China and grew to 1000 in 1930.

Such "scientism" on the part of Chinese scientists trained abroad, many from Cornell, was iconoclastic in its implications for traditional natural studies in China and influenced post-imperial literati such as Ch'en Tu-hsiu (1879-1942), who argued in the issues of the journal Hsin ch'ing-nien (New youth), which he helped found in 1915, that science and democracy were the twin pillars of a modern China that must dethrone the imperial past. In the process, post-imperial scholars initiated an assault on ko-chih-hsueh as a haven of superstition and backwardness. During the early Republic, the elite view of popular customs (feng-su ) was also reconfigured in modernist terms, a trend that culminated with Hsu K'o's Ch'ing-pai lei-ch'ao (Classified jottings on Ch'ing dynasty unofficial history). In Hsu's collection, popular lore was divided up and reclassified into the categories of "magicians and shamans" (fang-chi ) and "confused beliefs" (mi-hsin ), for example. Hsu K'o intended his collection of lore, published in 1917, as a sequel to the Northern Sung dynasty T'ai-p'ing kuang-chi (Expanded records of the T'ai-p'ing reign, 976-83), but the new cultural context ensured that such lore was publicly acceptable among modernist literati only if it could be pigeonholed as superstition.

Traditional Chinese medicine, which was the strongest field of the Chinese sciences during the transition from the late Ch'ing to the Republican era, was also subjected to such derisions, although it was more successful in retaining its prestige than Chinese astrology, geomancy, and alchemy, which were dismissed by modern scholars as purely superstitious forms of knowledge. When the Kuomintang-sponsored Health Commission proposed to abolish Chinese medicine (Chung-i ) in February 1929, for example, traditional Chinese doctors immediately responded by calling for a national convention in Shanghai on March 17, 1929, which was supported by a strike of pharmacies and surgeries nationwide on that day. The protest succeeded in having the proposed abolition withdrawn, and the Institute for National Medicine (Kuo-i kuan ) was subsequently established. One of its objectives, however, was to reform Chinese medicine along Western lines.

Bridie Andrews has documented the remarkable odyssey of Western medicine in early Republican China. She notes that the practice of Western medicine in China was assimilated by individual Chinese doctors in a number of different ways. Some defended traditional Chinese medicine, but they sought to update it with Western findings. Others tried to equate Chinese practices with Western knowledge and equalized their statuses as medical learning. The sinicization of Western pharmacy by Chang Hsi-ch'un (1860-1933), for example, was based on the rich tradition of pharmacopia in the Chinese medical tradition. Another influential group associated with the Chinese Medical Association, which stressed Western medicine, criticized traditional Chinese medical theories as erroneous because they were not scientifically based.

Andrews also documents how in this cultural encounter, techniques such as acupuncture were modernized by Chinese practitioners such as Ch'eng Tan-an (1899-1957), whose research on acupuncture enabled him to follow Japanese reforms by using Western anatomy to redefine the location of the acupuncture points. Ch'eng's redefinitions of acupuncture thus revived what had become from his perspective a moribund field that was rarely practiced in China and, when used, mainly served as a procedure for "primitive" blood-letting. This reform of acupuncture, which included replacing traditional coarse needles with the filiform metal needles in use today, ensured that the body points for inserting needles were no longer placed near major blood vessels. Instead, Ch'eng associated the points with the Western mapping of the nervous system. According to Andrews, a new "scientific accupuncture" sponsored by Chinese research societies thus emerged.

During the transition from the Ch'ing dynasty to the Republic of China, then, new political, institutional, and cultural forms emerged that challenged the creedal system of the late empire and refracted the latter's cultural forms of knowledge, such as traditional Chinese medicine. Just as the emperor, his bureaucracy, and literati cultural forms quickly became symbols of political and intellectual backwardness, so too traditional forms of knowledge about the natural world, were uncritically labelled as "superstition" (mi-hsin, lit. "confused belief "), while "modern science" in its European and American forms was championed by new intellectuals as the path to objective knowledge, enlightenment, and national power. Even those who sought to maintain Chinese traditional medicine by modernizing it according to Western standards of rigor, however, also played a part in the denigration of past medical practices.

The dismantling of the traditions of ko-chih-hsueh and po-wu-hsueh that had linked natural studies, natural history, and medicine to classical learning from 1370 to 1905 climaxed during the cultural and intellectual changes of the New Culture Movement. When their iconoclasm against classical learning and its traditions of natural studies climaxed after 1915, New Culture advocates helped replace the imperial tradition of ko-chih-hsueh with modern science and medicine. The fall of ko-chih-hsueh concluded a millennium of elite belief in literati values and five hundred years of an empire-wide classical orthodoxy that had encompassed the Chinese sciences. The legacy of destroying that cultural cum creedal system and the centering frames for human experience that it enforced should not be underestimated. What fell between 1905 and 1915 was an educational regime based on classical learning, which had been reproduced by the civil examination system from 1370 to 1904. Socially, classical credentials no longer confirmed gentry status or technical expertise, so sons of gentry turned to other avenues of learning and careers outside officialdom, particularly the sciences, modern medicine, and engineering. Literati increasingly traveled to Shanghai, Fu-chou, and other treaty ports to seek their fortunes in arsenals and shipyards as members of a new gentry-based post-imperial Chinese intelligentsia that would become the seeds for modern Chinese intellectuals.

Culturally, the longstanding affinity between literati learning and natural studies was also severed between 1905 and 1915. In other words, the linguistic monopoly of that official, classical knowledge by cultural elites no longer mattered as much socially or politically. As elites turned to Western studies and modern science, fewer remained to continue the traditions of classical learning (Han Learning), Ch'eng-Chu moral philosophy (Sung Learning), or ko-chih-hsueh that had been the basis for imperial orthodoxy and literati statuses before 1900. Thereafter, the "traditional Chinese sciences," classical studies, "Confucianism," and "Neo-Confucianism" survived as vestigal learning in the public schools established by the Ministry of Education after 1905 and have endured as contested scholarly fields taught in the vernacular in universities since 1911. The millennial hierarchy of literati learning, based on the Four Books and Five Classics, study of the Dynastic Histories, mastery of poetry, and traditional natural studies was demolished in favor of modern science and its impact via Darwinism on social and historical studies.

What then ensued after 1911 was a remarkable intellectual consensus among Chinese and Western scholars that imperial China had failed to develop science before the Western impact. Even the Chinese protagonists involved in the 1923 "Debate on Science and Philosophy of Life" accepted the West as the repository of all scientific knowledge and only sought to complement such knowledge with moral and philosophical purpose. We tend to forget the degree of skepticism that Joseph Needham's remarkable twentieth century "collectanea" Science and Civilization in China, initially provoked in the 1950s and 1960s. Many in Europe and America dismissed the great embryologist's foray into the history of Chinese science as a dead-end, a project they felt revealed Needham's wishful thinking about premodern China. The consensus then drew on heroic accounts of the rise of Western science to demonstrate that China had had no science worthy of the name. Both Western scholars and Westernized Chinese scholars and scientists had so essentialized European natural studies into a universalist ideal that when Chinese studies of the natural world, her rich medieval traditions of alchemy, or pre-Jesuit mathematical and astronomical achievements in China were discussed, they were usually treated dismissively and tagged with such epithets as "superstitious," "prescientific," or "irrational" to contrast them with the triumphant objectivity and rationality of the modern sciences.

Many scholars were so convinced that because China had had no industrial revolution and had never produced capitalism, therefore the Chinese could never have produced science. Few besides Needham, his collaborators, and Nathan Sivin stopped to consider what the rich archives in Taiwan, China, and Japan might yield if someone bothered to go through them. If we recall the situation as it was then, something our current graduate students are fortunately spared, then we can see how significant the contributions of Joseph Needham in England and Nathan Sivin in the U.S. were to beginning the unraveling of a consensus of poorly informed scholarly opinion about not only the "failed" history of Chinese science but also about the "victorious" history of European science. Both histories were pieces in the larger global narrative of Western success and Chinese failure that was taken for granted. The "triumph" of modern science in China should not be separated from the account of the "failure" of traditional natural studes. What Needham had inadvertantly rediscovered and revised was the "native origins" of science in China position promiment among classical scholars before 1900. In fact, Science and Civilization in China should be read as a modern, English version of its predecessors: the Ko-chih ts'ung-shu and Ko-chih ching-yuan.

As the evidence of a rich tradition of natural studies accrued in volume after unrelenting volume of the Science and Civilization in China project after 1954, it became harder and harder to gainsay it all as superstition or irrationality or just plain inductive luck, though some still tried. Nathan Sivin and others slowly persuaded us in the 1970s that as humanists and social scientists we were foolishly overlooking Needham's discovery of China as one of the great non-Western domains of human wisdom (along with the Islamic and Indian traditions) in all aspects of science, technology, and medicine. Sivin's pioneering studies of ancient Mohist optics (with A. C. Graham) and its penchant for geometry (when the field assumed that the Chinese had no geometry), early imperial Chinese mathematical astronomy and the demise of the cosmos (where we learned that Chinese calendrical specialists were more than just ideological hacks and political opportunists), medieval Chinese alchemy and its religious meaning (where we learned that Taoism was more than just the arm-chair philosophy of Lao-tzu and Chuang-tzu), Jesuit tactics in late imperial China and their manipulation of Copernicus (when we had thought the Jesuit agenda was premised on scientific interchange and not Catholic dogma), and the role of ritual practice in curing by doctors and healers in Chinese popular culture (when we thought that medical efficacy was simply the victory of the right drugs or the most advanced medical technology) all move across the technical boundaries of modern mathematics, chemistry, medicine, and astronomy to the social, political, and cultural systems of meaning that informed those fields in China before they broke free in the early twentieth century, for good and bad, from their premodern conceptual moorings.

As Chinese science has grown in respectability, the romanticized story of European science, whether capitalist or socialist in genre, has slowly unraveled under the onslaught of Thomas Kuhn and his historicist successors such as Bruno Latour. No longer reified and automatically placed on a priestly/scientific pedestal as the successor to our pre-1900 religions that had all supposedly "failed," the story of Western science became more complicated, more tragic, and more tense. The boyhood dreams of invention among Europeans, and their shrieks of "eureka" after new discoveries were revisited. Newton's mystical fascination with alchemy, Copernicus' devout religiosity that informed his astronomy, and the deadly serious but ungentlemanly competitive nature of "priority" in science and invention all challenged earlier politically neutral and culturally neutered fictions about the "scientific revolution" so long taken for granted. Scientists had been mostly men, after all, each caught up in a world of social, political, economic, and cultural concerns, which they could never completely leave out of their equations. Those equations, moreover, had yielded political, social, economic, and cultural consequences that few of them had foreseen while toiling in the laboratory.

With the exception of a reformed version of traditional Chinese medicine that has survived and is now thriving as one version of "holistic" medicine, the traditional fields of ko-chih-hsueh in imperial China were destroyed by the impact of modern science. Ironically, the history of Western science has also been decisively refracted when viewed through the lenses of Chinese natural studies, technology, and medicine at the end of the twentieth century. I suspect that the history of modern science can never again be what it once was, the history of a select group of early modern Europeans who supposedly discovered the natural laws of the universe and then passed them on with good will and cheer in the name of progress to the rest of the world.

In closing, let me stress a more positive legacy of the demystification of the traditional Chinese sciences at the turn of the twentieth century. Through their study of Chinese science, technology, and medicine, Needham and Sivin restored a measure of respect to the traditional Chinese sciences and thereby granted priority to research in areas where the received wisdom concerning ko-chih-hsueh was suspect, based as it was on careless speculation about banal generalities. Scholars now build on those contributions to search beneath the surface of self-satisfied discourses about "Western science" and the self-serving appeals to Greek deductive logic upon which they were conveniently based. The largest archive of premodern records for the study of nature most likely remains in China. Rather than asking if ko-chih-hsueh was "science," we should ask what the Chinese meant by the term and map its own categories. By better understanding premodern Chinese science, technology, and medicine, and the mystifications that undergirded it, we are more perceptive about ourselves and the mystifications that undergird our contemporary versions of modern science.