The Chinese Civil Examinations

During the early twentieth century, the Chinese civil service examinations became in China, as well as elsewhere, a symbol of the failure of the Chinese dynastic state and the backwardness of Chinese society. Chinese intellectuals saw the examinations as an obstacle to modernization, to the spread of modern science and technology, and to democratization and the broadening of education to the population at large. The “eight-legged essay” (baguwen), one of the genres tested in the examinations, was then typically singled out as the embodiment of a system that had forced millions of students to mindlessly memorize classical texts and reproduce them in calligraphic writing in a strictly parallel format that left no space for individual creativity and curricular innovation. These modernist critiques have long dominated in textbook accounts and the public perception of the Chinese civil service examinations. They stood in sharp contrast to pre-twentieth century Asian and European assessments of the examinations that have in recent years gained renewed attention. Time then to reassess why the examinations attracted ever-increasing numbers of candidates between the late sixth-century and 1900 even as the odds of success kept declining, and to reconsider what has been their longue durée cultural legacy.

A Brief Institutional History

In the imagination of many who participated in the civil service examinations organized by successive dynasties that ruled the Chinese territories, the foundation for a system in which men—and not women—were selected for government office on the basis of written examinations was laid in the Zhou Dynasty (ca. 1100–256 BCE). Would-be officials were then reportedly tested before they were given assignments, and such tests could be seen as precursors of the examination system. This view was based on texts that many believed to have been written during the Zhou period but that others convincingly showed to have been reconstructions dating to the first centuries before the common era. These texts provided an idealization of the governmental organization of the dynasty that was both the longest reigning in Chinese recorded history and the last of the Three Dynasties of Antiquity. The Three Dynasties were conceived of as a golden age in which the key elements of civilization were created and that therefore also continued to serve as models for later Chinese polities. The institutional history of the civil service examinations is long, but not quite this long.

We can more confidently trace back the use of written examinations for the recruitment of men for office to the medieval period. From the end of the sixth century onwards the Sui (581–618) and Tang (618–907) dynasties used written examinations not only to promote or assign posts to men who had already been serving or who had been recommended for office—as their immediate predecessors since the Han Dynasty had done—but also to select and recruit newcomers. The examinations thus became a means to reduce the power of the aristocratic families who had been dominating the political scene in the preceding five centuries. The interest of medieval emperors to recruit new people who would be loyal to them is best illustrated in the case of Empress Wu, the wife of the third Tang ruler, Emperor Gaozong. Empress Wu had long been influential at court and following her husband’s death, she oversaw the reign of their two sons. However, she was unsatisfied to govern through puppets and in 690 she proclaimed herself emperor, the only woman in Chinese history to rule in her own name. Unsurprisingly, her actions caused a lot of resentment. To counter the formidable opposition to her rule, she relied on the examinations to recruit more new officials than any of her predecessors. Under her reign, a growing number of those entering the civil service did so through the examinations. Successful examination candidates attained the highest positions in the bureaucracy.

Empress Wu’s use of the examinations established a pattern that was to be repeated subsequently. New rulers opted to select officials based on a competitive process and not solely on family pedigree or wealth. The founders of the Song dynasty (960–1276), for example, pushed the use of the examinations further as part of their attempt to curb the power of regional military commanders and powerful families. The early Song rulers turned the examinations into the primary channel for recruitment into officialdom. Earlier, the number of those who entered officialdom through the examinations constituted a minority in comparison with those who were recommended by powerful patrons or inherited positions through family connections. From the Song onwards, the examinations became the primary and most well-regarded way to get into government. The highest ranked jobs went to those who ranked highest in the examinations. Nevertheless, equal access or upward mobility were not the main drivers in this process. Certificates of good family standing were required for entrants. And, as competition both in the examinations and in the subsequent assignment to an official post increased, the number of titles and offices awarded on the basis of hereditary privilege and the purchase of official titles outstripped those obtained by examination graduates. The latter tendency increased when state coffers were depleted, as was the case in the late Qing (1644–1911) period.

A second trend that started during the Song period was the expansion of the number of candidates, those preparing for and those participating in the examinations. During the twelfth and thirteenth centuries, the number of candidates grew at exponential rates, and overall numbers continued to go up in the Ming (1368–1644) and Qing dynasties, roughly following demographic trends. The number of candidates attempting the Song qualifying prefectural examinations increased from an estimated 20,000 in the early eleventh century to 400,000 by the mid-thirteenth century—when the Song boundaries had been pushed back to the Huai and Yangzi Rivers and out of a total estimated population of about 50 million. By 1850, the number of candidates for the local examinations had increased to about 2 million out of an estimated population of 430 million.¹

Crafting Curricula

The institution of the civil service examinations as the pre-eminent avenue to office in the eleventh century was part of a broader package of centralizing policies aimed at strengthening court control over the military, the bureaucracy, and local powerholders. The implementation of centralizing policies coincided with the growing marketization of the economy and urbanization in large parts of the Song territories. These intertwined processes shaped the social, political, and cultural impact of the civil service examinations in unforeseen ways.

First, it is important to be clear about what the regularization and standardization of the examinations did and did not cover in the eleventh century and throughout most late imperial Chinese history. Successive courts and bureaucracies defined the basic structure and organization of the examinations. They set out the procedures that were to be observed during the examinations, the evaluation process, and announcement of the results. From the Song through the end of the Qing periods, examinations were held at regular intervals—usually one to three years with only short periods of interruption—and, in a hierarchical series, starting at local levels and moving up to the capital and the palace. Even though the complexity of the examinations grew over time, this basic hierarchical structure was retained, with graduates at lower levels traveling to the provincial and metropolitan centers of political power and eventually, albeit often theoretically, facing the emperor himself in the palace examinations.

Central governments also described, and from time to time revised, the written genres in which candidates would be tested at each level of the examinations, the number of essays in each genre, and the order in which they were to be submitted. In the last 900 years of the examinations the following genres were most common:

• essays on the meaning of the classics, in which candidates should explain and interpret passages from the Five Classics and, from the early thirteenth century, also the Four Books;
• regulated poems, in which candidates were asked to respond to a topic in parallel verse and with appropriate prosody;
• disputations, in which candidates responded to a question drawn from classical, historical, or philosophical texts; and
• policy essays, in which candidates responded to questions on current affairs that drew on a broad range of classical, historical, literary, and administrative texts.

Sometimes candidates could opt among some of these genres—e.g., poetry or essays on the meaning of the classics—and sometimes other genres were added; for example, legal judgments were occasionally added. The same genres and similar questions were tested at all levels. The order, first or last, was taken to be significant, with some arguing that the genres tested first were the ones that mattered most, and others objecting that those tested in the last session ultimately determined the candidate’s ranking and were therefore decisive.

Importantly, central governments in imperial times by and large did not prescribe a curriculum. The classics, or at least some of the classics, were obviously required reading. Some emperors, councilors, and local officials had editions and commentaries printed that they wished to see as the standard editions. There was, moreover, a repertoire of texts that most would have considered to be part of the formation of a scholar-official: the dynastic histories, the military classics, the “master texts,” or composite philosophical writings attributed to the founders of pre-imperial schools of thought, as well as the poetry and prose of great writers past and present. But most shied away from the kind of interventionist educational policies and examination reform that had been tried under Councilor Wang Anshi in the late eleventh century. Wang attempted to embed new commentaries on the classics and a dictionary that redefined the meaning of many characters in both the examinations and the school system, all in support of his encompassing reform package called the New Policies. This program was met with fierce resistance among peers and was later held up as a negative example. It is important to keep in mind that there was no standardized curriculum that outlined what topics or what titles ought to be taught at what stage. Furthermore, the examinations were not embedded in a comprehensive network of national or empire-wide government schools. These institutions existed and increased in number under the reign of some emperors and councilors, but they enrolled only a minority of the large numbers of students attempting the local examinations from the eleventh century onwards.

The demand for instruction on how to prepare for the various genres tested in the examinations was mostly met by those engaged in the process of preparation itself, the so-called literati. The literati were the cultured elite; not merely literate but well-versed in the written and material cultures, the languages, and the social conventions of a class that—more so than any other social group in imperial Chinese history—networked both locally and on regional and empire-wide scales. Participation in the examinations was a defining characteristic of this group. The increased wealth produced, especially in the southeast of the Chinese territories, led to demographic growth and rising numbers of candidates from the eleventh century onwards. Because late imperial Chinese states opted for small government and chose not to expand their bureaucracies and set the overall quota for graduates low—about 1/100 on average—success rates soon became prohibitive; with some prefectures in the south seeing graduation rates of 1 in 200—enough to make any modern examination or grant competition look favorable. The examinations then also obtained a prominent place in Chinese religious life and fortune-telling. Stories about premonitions of examination success or failure were widely shared from Song to Qing, while students and their families turned to patron gods like Wenchang in temples distributed across large parts of the Chinese territories.

The decision to keep the number of officials more or less stable ensured that even successful candidates could not or could only intermittently count on income from an assignment in the bureaucracy. Under these conditions, the examinations became foremost a venue where social status was obtained. Those who participated as well as those who succeeded could turn to and, throughout the course of their lives, combine a diverse array of occupations. Many became teachers working individually or for government or private schools and academies. They also turned to compiling and editing textbooks and anthologies, publishing real and mock questions or fascicles with successful and unsuccessful examination essays. These were, alongside a wide variety of other literature, printed and marketed for “those engaged in examination preparation.” Others became doctors, notaries, litigation specialists, secretaries, merchants, managers of family estates or local welfare projects, or became engaged in scholarship. Some also joined or led uprisings, as was the case with Hong Xiuquan who, after repeated failure in the examinations picked up a protestant tract in 1843 and eventually established the Kingdom of Heavenly Peace in 1851. Hong’s theocratic state was a mix of traditional and new elements of which a remodeled examination system was a prominent example.

The standards by which examination learning and literati status ought to be evaluated were to be worked out in the examinations themselves and in the broader field of examination preparation and critiques thereof. These negotiations involved various agents, including the court and central government officials, lower-level officials, examiners who were temporarily dispatched on ad hoc assignments to supervise examinations, teachers in the official metropolitan and local schools, private teachers, authors and printers, and individual candidates. Intellectual and political networks bringing together officials, teachers, and students shaped and reshaped the curricular standards by which examination learning was defined over time. There are plenty of examples of how new intellectual and sociopolitical trends gradually emerged in the examination halls, ranging from the Return to Antiquity movement of the eleventh century, the utilitarian historical scholarship of the Yongjia school of thought and the moral philosophy of the Learning of Way in the twelfth and thirteenth centuries, Buddhist and Daoist thinking at various times during the Song and Ming, the Learning of the Mind in the sixteenth and seventeenth century, evidentiary scholarship in the eighteenth century, and, in the nineteenth and early twentieth century, liberal economics.

Perhaps the most memorable example of the mediating role played by the examinations in Chinese political and intellectual history relates to the spread and eventual official endorsement of Neo-Confucianism. This was the reorientation of Confucianism toward the general meaning of classical texts and the prioritization of modes of self-discipline, grounded in a new metaphysics formulated in response to Buddhist philosophy, social practices, and organization. While acting as local magistrate in the early 1150s Zhu Xi, a central figure in Neo-Confucianism, presented local students with the following examination question:

Recently, you, students, have been devoting yourselves to the text of The Analects. You have pondered all that is said in its twenty chapters. Even though you still have not been able to investigate the meaning like you have investigated the text, still, one cannot say that you have not applied yourselves to this [investigating the meaning]. You have already firmly absorbed the abstract and general parts in your minds, but can you also let the examiner know how you have applied this in practice? If not, you would be studying here not because you have your minds set on the path of morality, but only because of self-interest; this is not what I expect from you.²

In case the point had been lost, he reformulated the question as follows in what may be the shortest policy question ever asked in the history of the civil service examinations:

When people are young they learn things, after they have grown up, they want to put these things into practice. Can we hear from you, gentlemen, what you are learning today and what you will put into practice in the future?

Zhu Xi’s questions formed part of his broader critique of the examinations, but they also showed how he incorporated alternative pedagogies into the examination regimen in order to address the pedagogical problems he associated with standardized examinations. His short questions violated most of the norms then governing the genre of the policy question. Examiners typically described a broader question in any field of governance—e.g., agriculture, taxation, military organization, military strategy, interstate relations, water conservancy, monetary policy, personnel selection and administration, literati learning, astronomy, etc.—and outlined various approaches taken in the past based on classical and historical texts and more recent precedent. Zhu Xi drafted his questions to have students link the conventionally required exegesis of canonical texts with self-reflection on individual practice.

Following fierce critiques from other examiners and peers, and following a ban on the distinct discourse of early Neo-Confucianism that was engulfing provincial and metropolitan examination halls in the late twelfth century, Zhu Xi’s interpretations of the classical tradition were eventually officially endorsed during the 1210s and 1220s, and then again in the early 1300s. His examination questions and the examination preparation materials that his students developed in the thirteenth century equally show how in order to gain a widespread following, intellectual movements had to make accommodations to the conventions of the examinations.

Modern Prose Aesthetics and Individuality

From the history of examination preparation materials we can thus conclude that the examinations were a platform where curricular standards and standards for literati status and statesmanship were negotiated and renegotiated. Do we see this nexus of state and literati interests reflected in the examination essays themselves? Given that twentieth-century examination critiques have focused in particular on the eight-legged essay (baguwen), we can narrow the question down as follows: how could examiners select the best on the basis of a format that required candidates to present their interpretation of one or more brief passages from the classics in eight strictly parallel verses? To early twentieth-century reformers this apparently ossified format had enslaved candidates for five hundred years and symbolized all the ills of the politics of the monarchy and the imperial system. These included a focus on rote learning and the memorization of a set of ancient texts rather than innovative argumentation, on form and sound rather than content, on artfulness rather than authenticity, on useless feudal knowledge rather than science for the betterment of the nation, and the qualities of dilettantes rather than democratic access to education and free speech. These critiques have had a lasting influence on how the examinations have been perceived ever since.

Based on a meticulous reading of eight-legged essays and critiques thereof between the 1470s—the conventional start date of the eight-legged essay—and the 1910s, Alexander Des Forges has recently shown that for most of its history “modern prose” (shiwen, the prose of our time), another term that was used to refer to this genre, was considered to test the opposite of the qualities that modern critics attributed to it.³ Take, for example, the notion of the eight legs itself. Successful and model essays varied in length to any number of legs—to be understood as parallel sections in the essay—and parallelism was something to be detected and appreciated by attentive readers as it could come in widely varying configurations and be abruptly and intentionally interrupted with free prose. Modifying conventions in meaningful ways constituted creativity and caught the eye of examiners and critics.

Similarly, the assumption that the requirement to speak in the voice of the original author or speaker in the quoted passage silenced the student’s individual voice and stifled originality reverses the aesthetics that informed the writing and appreciation of the genre. It ignores the high regard in which pre-twentieth century literati generally held it. In so doing, the modern mythologization of baguwen also misses the influence of modern prose on other literary genres throughout late imperial Chinese history. There was some debate as to whether speaking in the voice of someone from the past and imagining what they could have thought or said in a given situation turned examination candidates into playwrights or actors. But the great masters of Ming theatre and the vernacular novel agreed that the best modern prose made great literature. Moreover, the authors of successful essays tended to switch voice and perspective in their work. Candidates were by definition required to develop their own voice along with that of notable historical figures. From the Song through the Qing, examiners and critics alike insisted that candidates needed to demonstrate “their own intentions” (jiyi) after having articulated the voices of others. Shiwen was understood as a genre in which examiners could evaluate the candidate’s process of self–development, as well as the lack thereof.

Originality, creativity, and subjectivity were thus key concepts in the critical appraisal of the eight–legged essay throughout imperial Chinese history. Some central figures in modern China also recognized its significance in the longue durée Chinese literary and intellectual history. Zhou Zuoren, an activist translator of European literature and historian of Chinese literature, wrote that eight-legged essays were “the crystallization of Chinese literature and Chinese culture” and accorded them a critical place in the development of modern Chinese literature.⁴ The co-creator of pinyin, the now standard transliteration system for Chinese, Zhou Youguang, was similarly not a man stuck in tradition. In 2003, he demonstrated the critical potential of the genre in an eight-legged essay on a famous saying attributed to former Chinese President Jiang Zemin, “Moving with the Times.” The concluding leg reads in translation:

Truth also changes over time, it is not immutable. “Practice is the sole criterion for the investigation of truth.” Truth does not fear criticism; criticism nurtures truth. Whatever fears criticism is not truth; it is religion or dogma that has not yet adjusted to the times. The Age of Superstition is coming to an end, the Age of Slavish Obedience is coming to an end. Now is the age of thinking independently, of selecting and sticking with the good, of not being constrained to eagerly pursue “moving with the times.”⁵

Meritocracy and Social Reproduction

Leaving aside the broader cultural and political critiques of the civil service examinations that pre-occupied early twentieth-century critics, post-World War II social historians turned to the question of whether the examinations had served to promote social mobility. This is a question that has in recent years attracted renewed attention, in part due to the emphasis on meritocracy as a hallmark of Chinese culture in contemporary Chinese political discourse and in the political theory of New Confucianism. It is worth recalling that in many ways the post-war and contemporary debates about meritocracy and social mobility hark back to the seventeenth- and eighteenth-century European interpretations of the social and political significance of the Chinese civil service examinations. To many early modern French, German, English, Italian, Spanish, Dutch, and Russian readers of reports on Ming and Qing society and politics, the literate examinations were a model for European polities, a means to restructure the privileges of the nobility and to develop education in order to instill patriotism and civic duty in all. So, did the examinations promote social mobility?

Initially, the analysis of the backgrounds of examination graduates appeared to yield conclusive evidence for an extraordinary high degree of upward mobility in Chinese society, especially when compared to other pre-modern polities. Quantitative analyses comparing the social background of the graduates’ paternal forefathers across three generations led to the conclusion that staggering numbers of successful examination candidates—over 50% in the Southern Song, 49.5 % in the Ming, and 37.6% in the Qing—came from families with no immediate patrilineal forefathers in the bureaucracy. Others were quick to point out that significant family ties were not limited to immediate forebears in the patriline, and that many social groups were excluded from taking the examinations—such as monks and priests, beggars, sons of prostitutes, craftsmen, clerks, and, at times, merchants—not to mention all women. More recent work on the Qing list of graduates furthermore shows that the probability for succeeding in the examinations rose significantly when family background is measured by the highest status achieved by immediate paternal ancestors going back three generations and that the father’s official rank was highly significant in determining a graduate’s appointment and career. At a time when the percentage of students from underprivileged backgrounds in higher education is declining, and for those familiar with social reproduction among politicians and academics, this need not come as a surprise. Despite the wide arsenal of stories describing how hard work might win a poor student clan or community support or bring him under the patronage of local notables and result in great success at the capital, social reproduction was undoubtedly common and social equality across class boundaries was never articulated as an explicit goal.⁶

Nevertheless, the expansion of the civil service examinations in the eleventh century marked a significant turning point in Chinese and global history. Learning and writing gained an unprecedented role in recruitment and appointment to office; family pedigree, wealth, landholding, and marriage connections remained crucial determinants of elite status, but these latter elements no longer sufficed to secure an examination degree and obtain high official status, or indeed to acquire or retain literati status. The emergence of an empire-wide class of literati closely connected to each other and tied to court and empire ultimately proved to be key to the success and longevity of the examinations. This was a class defined not only by the consumption of classical and historical texts and examination preparation literature, but also by the individual production of literature to be examined by peers, or, in Des Forges’s terms, early modern literate industriousness.