For roughly thirteen centuries, educated men in imperial China could seek status and office through examinations. A man who passed might qualify for a government career, gain a reputation that helped him teach, or bring new prestige to his family. Preparing demanded years of study and money that many families did not have. Competition was severe, yet the chance to gain status through study extended beyond the sons of hereditary nobles.
The system was called keju. It took shape under the Sui dynasty, in the late sixth and early seventh centuries, and continued in changing forms until the Qing government abolished it in 1905. The examination sat by a Tang poet in the eighth century was unlike the one faced by a Qing candidate in the nineteenth. Despite those changes, emperors continued to use written examinations to assess the learning and judgement of prospective officials.
Passing an exam did not guarantee an appointment, and the system did not make the empire democratic. Women could not enter the regular civil examinations, while poverty prevented most families from supporting a student through lengthy preparation. Personal connections still affected careers. Even with these restrictions, examination credentials made education politically valuable. For provincial educated families, success could connect their local standing to the imperial government.
Before the imperial exams: the Han dynasty and powerful families
Long before the Sui, Chinese emperors had to find men to run their administration. Under the Han dynasty (206 BC to AD 220), recommendations from local officials and influential figures were a major means of recruitment. A man might be put forward for scholarship, conduct or apparent administrative ability. Influential families could use recommendations to advance their own members. Officials sometimes underwent tests, and knowledge of classical writing mattered increasingly, but there was no competition on the scale of the later imperial examinations.
The Han rulers established an Imperial Academy, the Taixue, where students studied texts linked to Confucian teaching: the Classic of Poetry, Book of Documents, Book of Rites, Book of Changes and Spring and Autumn Annals. These works provided examples of rulers, families and political decisions, along with arguments about ethical conduct and order. Study of the classics prepared prospective officials to discuss a ruler’s responsibilities and the duties between government and subjects.
After the Han empire broke apart, the standing of powerful aristocratic lineages became especially important to official careers. Families with established networks were well placed to secure appointments; an unknown scholar might be overlooked regardless of his ability. Later rulers sought capable men beyond these lineages to reduce their dependence on hereditary power. Columbia University’s Asia for Educators account of the Confucian curriculum explains the long connection between studying canonical texts and entering public service.
How the Sui dynasty began the examination system
The Sui reunified much of China, creating a government that needed administrators across regions recently ruled by different powers. Emperor Wen (581 to 604) and Emperor Yang (604 to 618) expanded ways of selecting men for their literary learning and their capacity to discuss policy. Recommendations remained part of recruitment. The rulers added literary qualifications to older methods of appointment, which continued alongside them.
The date 605 often appears as the beginning of the imperial examinations. The reforms developed over several years. The early jinshi, usually translated as “presented scholar,” is associated with Sui developments around 605 to 607, although the exact timing of its characteristic features is disputed. No surviving account establishes that Emperor Yang introduced a single, empire-wide standardized test open to every eligible man on that date.
For a long time, family standing and access to education still affected who could become a candidate. Under the Sui, literary qualifications recognized by the central court became more important. It could support a man’s claim to office beyond the reputation of his household. When the dynasty fell in 618, the new examination practices remained useful to the rulers who followed, because they helped identify officials sharing a common education.
Tang examinations: poetry, recommendations and Empress Wu
Under the Tang dynasty (618–907), candidates could enter different examination categories. Some studied for assessments on Confucian classical texts, while the increasingly respected jinshi route placed greater emphasis on writing. Examinees might compose verse under strict formal rules, demonstrate their prose style or answer questions about governing. The administration relied heavily on written communication, so these skills had practical value alongside their prestige.
A successful Tang examination did not automatically lead to a government post. Noble families retained influence, state schools offered other credentials, and candidates could seek patrons by circulating their writings or winning a recommendation from an established official. Personal reputation continued to influence recruitment. An examination result was one path to distinction in a system that continued to rely on appointments shaped by relationships.
Wu Zetian, who emerged from the Tang court and ruled as emperor of her own Zhou dynasty from 690 to 705, encouraged the examinations. They offered her one way to recruit men whose advancement depended on the sovereign, rather than entirely on aristocratic families. Recruiting men loyal to the central court mattered politically as well as educationally. Recruitment methods continued to evolve after her reign.
In the Tang period, literary achievement could bring recognition even when a candidate failed. Poems by aspiring officials circulated among other educated people, and reputations sometimes outlasted disappointing examination results. Du Fu, later celebrated as a poet, did not gain office through the examinations as a young man and eventually looked to other means of entering service. His career also shows that government service could be pursued without passing the examinations.
Why the Song dynasty made examinations central to government
The Song dynasty, founded in 960, made examinations a more important route into administration. Earlier decades had shown how regional military commanders could accumulate power while competing dynasties struggled to control them. The Song court sought a stronger body of civilian officials to govern the provinces and limit the independence of generals. These administrators oversaw revenue, supervised local government, prepared policy and sent written memorials to the emperor.
During the early Song, the palace examination developed into the final stage associated with the emperor himself. The procedure associated success with the emperor himself and reduced the appearance that high-ranking graduates owed their careers to a local patron. The emperor was not personally responsible for reading every script. The ritual helped bind successful candidates to the throne they would serve.
Officials devised procedures to limit favouritism in marking papers. A candidate’s name might be concealed and the answers recopied so examiners would not recognise the writer’s hand. These procedures took time and money, but an unfair result could damage the credibility of a competition on which families staked so much. Columbia’s account of Song scholar-officials describes the court’s political interest in such protections and the students’ concern that their work be judged impartially.
The number entering the competition rose sharply. Columbia’s teaching materials give fewer than 30,000 candidates near the beginning of the Song and roughly 400,000 by its end. Government vacancies did not keep pace. Well-educated men might spend years preparing and taking examinations without ever securing office. Elite families increasingly pursued examination success, although failure was common among candidates.
The expansion took place amid growing cities, commercial exchange and the spread of printing. Books became easier to obtain, and prosperous households outside the oldest aristocratic clans could devote resources to education. Civilian government needed trained officials, while families valued the standing and opportunities attached to examination credentials. A 2022 study by James Kai-sing Kung links the rise of commercial society and private education with the Song rulers’ wish to keep military leaders from monopolising power.
What candidates studied and what an exam actually tested
Candidates needed far more than basic literacy. Students learned classical Chinese, the written language shared by educated readers across regions where people spoke different varieties of Chinese. They memorised passages, studied traditional commentaries and practised writing within accepted forms. Much of the curriculum drew on Confucius, traditionally dated to 551–479 BC, and the teachings about government, family obligation and the cultivation of proper conduct associated with him.
The syllabus was never identical across dynasties. Tang students might write poetry and other literary pieces, while Song examinations could include policy questions and interpretations of the classics. Under the Ming (1368–1644) and Qing (1644–1912), the Four Books and Five Classics became central to preparation. The Four Books were the Analects, Mencius, Great Learning and Doctrine of the Mean; the latter two began as sections of the Book of Rites. The interpretations of Zhu Xi, a Song philosopher, carried particular authority in this later curriculum.
Many Ming and Qing candidates were expected to master the eight-legged essay, or baguwen. The composition followed a tightly defined sequence of sections and demanded careful control of language and classical references. Producing a successful essay was difficult, but candidates worked within officially favoured readings of the texts. The same training rewarded conformity, making unconventional arguments risky even as candidates honed their literary skills.
The tests required more than memorisation. Across different periods, questions covered policy, classical commentary, poetry and other forms of writing. Emperors needed administrators capable of settling disputes and responding to practical crises, while tests measured performance under literary rules. A scholar able to produce an excellent script might still be ill prepared to administer a county suffering from famine. No change in the syllabus entirely removed that tension.
The examination ladder: from local student to jinshi
By the Ming and Qing periods, candidates normally worked their way through an established sequence of qualifications. Local examinations could grant the status of shengyuan, often called xiucai. That qualification brought standing among local educated families but did not itself make the holder a senior official. Men at this level might teach, assist with local affairs, look after family property or continue studying for the next competition.
The provincial examination, or xiangshi, was generally held every three years. Men who passed were called juren, “recommended men,” and received a qualification that could create further opportunities for office. They could then sit the metropolitan examination in the capital, the huishi, and proceed to the palace examination, the dianshi. Those awarded the highest qualification were jinshi. This ordered ladder describes the later imperial system most accurately. It should not be projected in full onto the Sui or early Tang examinations.
The palace examination established a ranking among the final candidates. First place carried the distinction zhuangyuan, bringing attention to the graduate and often to his home district. Even those awarded the highest degree entered a government with finite vacancies and several routes to appointment. The degree improved their prospects, although even top graduates were not assured a well-paid post.
The examination system required its own administrative apparatus. Officials chose examination dates, appointed graders, set quotas, controlled papers, supervised entrants and announced results. An imperially prescribed Qing examination rulebook preserved by Academia Sinica sets out procedures for composing questions, processing scripts and arranging official ceremonies. It shows the practical administration required to make written competition part of government recruitment across a large empire.
Inside the examination hall: isolation, handwriting and cheating
In the later dynasties, provincial and metropolitan examinations could require candidates to remain for long stretches inside compounds of small individual cells. They wrote and slept in confined spaces, bringing their own necessities and working under close supervision. Isolation limited access to helpers or books, and the harsh conditions demanded physical endurance alongside scholarship. These detailed arrangements are best documented for the mature examination system, not its earliest Sui form.
Surviving papers show how those controls worked. Academia Sinica holds a 1685 black-ink metropolitan examination script associated with Zhao Hao, preserving identifying details and room for written answers. It also holds a 1685 vermilion copy of an examination paper by Wang Wei. A candidate originally wrote in black ink, but clerks could recopy scripts in red so that examiners did not recognise a person’s handwriting. The original and copied papers show how officials tried to judge writing without revealing the writer’s identity.
Impersonation, bribery and concealed notes all threatened a system whose authority depended on credible results. Some candidates nevertheless tried to carry forbidden material into the hall. The small Qing-era crib books and silk notes displayed in a National Tsing Hua University museum exhibition demonstrate how compact such aids could be. These objects document attempts to cheat, but cannot tell us how often the fraud succeeded.
The papers of successful students were useful long after results were announced. In 1702 the Qing Board of Rites circulated selected answers from the Huguang provincial examination, offering other students examples of highly rated work. A surviving printed selection held by Academia Sinica records this practice. Circulating successful scripts gave later candidates models of the writing examiners rewarded.
Could a poor farmer really become an imperial official?
A poor farmer’s son might advance through the examinations, but preparation imposed costs many households could not meet. Families needed to feed a student during years of study, provide books and possibly teachers, and pay for journeys to examinations away from home. A boy preparing for office was also unavailable for much household work. The poorest families often could not sustain the investment. At different times, certain occupations faced further legal or customary barriers.
Neither wealth nor family connections guaranteed a pass. Some relatives pooled their means to support a promising student, while local schools and lineage organisations helped men outside the highest aristocratic families continue their education. A degree might increase a family’s influence throughout its community. Candidates who never passed could still earn a living through teaching, record keeping, dispute mediation or service to local elites.
The regular civil examinations were closed to women, despite women’s participation in literary culture and the substantial education received by some. Claims about selection by merit apply only to the restricted group of men allowed to compete. Historian Benjamin Elman’s study of late imperial examinations argues that late imperial rulers were not trying to turn peasants in general into officials. The examinations circulated opportunities among educated families while strengthening the state’s ties to scholarly elites.
A single passing rate cannot describe thirteen centuries of examinations. Chances differed between dynasties and between local tests and the final degree, so a figure for one level cannot describe everyone who began studying. Elman’s research on the Ming and Qing examinations discusses millions of candidates in Ming and Qing China, with only a small proportion completing the entire sequence. The large number who failed shaped the institution’s social effects as surely as the small number who entered the bureaucracy.
How the Yuan, Ming and Qing changed the system
The Yuan dynasty (1271–1368), ruled by the Mongols, interrupted the examination tradition rather than simply adopting the Song system intact. The tests were suspended for an extended period and later reinstated under revised rules. Quotas and categories reflected the political hierarchy of the Yuan state, meaning candidates did not always compete on equal terms. The Yuan interruption shows how an examination tradition could persist for centuries despite periods of suspension and major changes in its rules.
Ming emperors restored examinations as a major channel of official recruitment, increasingly grounding the syllabus in Neo-Confucian readings of the classics. Qing rulers kept and adapted that framework. The Manchu-led court used examinations to draw Chinese scholars into government, but banner populations retained certain privileges and other routes to office continued. Hereditary privileges, patronage and purchased ranks or posts continued alongside examination degrees.
Late examination procedures are known in unusual detail through Qing regulations and surviving papers. Those records cannot simply be used to reconstruct how examinations worked a thousand years before. Elman’s study of examination records from 1315 to 1904 examines their later history while stressing differences between Ming and Qing practices and the literary competitions of the Tang and Song. The system’s long life depended partly on its capacity to change.
Why China abolished the civil service examinations in 1905
By the nineteenth century, the Qing government faced rebellions, financial strain and defeats in war. Reformers argued that years spent mastering the traditional curriculum did too little to prepare officials for challenges involving science, military organisation and foreign affairs. Literary education still had defenders, but reformers and defenders of literary education disagreed about which skills officials needed to govern and strengthen China.
Calls for reform grew after China’s defeat by Japan in 1894–95 and foreign intervention during the Boxer uprising of 1900. Qing authorities began developing new schools and forms of training. In September 1905, the court ordered the traditional provincial and metropolitan examinations to cease from the following year. Columbia’s account of late Qing reforms places this decision among wider attempts to reform the state and education before the dynasty’s fall in 1911.
Ending the examinations disrupted plans made by families who had spent years supporting candidates. Classical credentials carried social standing without a senior appointment, and repeated attempts consumed years of many candidates’ lives. New schools provided alternative prospects, but they could not immediately recreate the networks of learning and prestige surrounding the old tests. Abolishing keju therefore severed a familiar connection between education, local standing and imperial careers.
For approximately 1,300 years, imperial examinations served both the rulers recruiting officials and the educated families seeking advancement. Rulers obtained officials trained in a common written tradition. Households invested in learning because an examination credential could raise their standing. Access remained unequal, places were scarce, and success at literary composition did not ensure skill in government. Even with these restrictions, examinations shaped the social and political lives of educated families throughout imperial China.
Discussion
Comments