摘要
An assessment system is an important prerequisite for the formation of bureaucracy, in ancient regimes and modern states alike. Our purpose here is to present a complete and thorough reconstruction of the T'ang assessment system, and to explain the relationship of this particular system to the whole bureaucracy.
The T'ang assessment system was basically a modification of the institutional heritage that previous dynastic governments had handed down. But the system itself was refined with regulations and practices more complicated than before. Through a well-defined system of evaluation, every T'ang official, civil or military, within or outside the current of promotion, was to a great degree under its influence. Depending on his rank, a T'ang official had to pass through repeated assessments conducted by his superiors in office. The result of assessment, most important, the official ratings, was written into an "assessment record" which was eventually sent to the central government. The latter then gave its final approval and decided whether the official deserved promotion and transfer, or demotion and banishment.
The assessment system underwent great change during the T'ang dynasty. This was more or less a reflection of broader change within the bureaucracy. When the equilibrium between the members of candidates for office and of available posts was broken after the second half of the seventh century, the excessively severe competition among candidates was not only noticeable in aspects of recruitment and Civil Service Examination, but also conspicuous in the field of assessment. In order to smooth such struggles, the issue of "seniority" was raised and made to outweigh merit first in 669, and then once again in 730. Subsequently, the T'ang assessment system was fixed in terms of merits and seniority which are also the basis of the "merit system" in modern bureaucracies.
Other changes such as the numbers of assessment and "jump in promotion" were directly related to the speed of promotion concerned by every official within the T'ang bureaucracy. The 4th grade and higher official ratings were eagerly sought because they offerred the official assessed an opportunity to increase both his emoluments and his rank.
To prevent illegal offenses by officials, T'ang law set forth intricate regulations on their demotion, exile and banishment. Punishments of such kinds were pertinent results for those officials with bad "assessments records."
However, the T'ang assessment was neither perfectly designed to the extent that its official rating could be formulated by an arithmetic method as Nemoto Makoto 根本誠 had once assumed (see Appendix), nor was it merely paper sham as one modern scholar has concluded. Although the assessment official was given unchallengeable power, one must see the whole system within the context of bureaucratic efficiency. A carefully designed system of checking procedures reflected a balance of powers which extended over most parts of the T'ang administrative machinery.
Of course, the system had its inherent problems. Almost every extant T'ang contemporary discussion on the recruitment and assessment systems shows that crises and tensions within the bureaucracy were caused by the improper conduct of these policies. Their criticisms of abuses and suggested solutions remind us that the reality of the T'ang assessment system has to be examined not only through the static regulations, but also in the context of dynamic changes within the bureaucracy. We will evaluate the assessment system from its dynamic perspective in the future.