For researchers studying Chinese politics, the concern is not whether factions are important, but rather how to identify them. In the Chinese context, factional affiliations are often concealed, requiring researchers to devote extensive efforts to parse them. Faction detection methods have transformed over time, from the “rumours-have-it” approach of first-generation scholars, to the “backgrounds-in-common” framing of the second generation, and the “practices-of-patronage” focus of the current generation. This article offers a systematic review of these approaches and finds Junyan Jiang’s patronage-focused, “within-tenure promotion” approach to be the most accessible and justifiable. Building on Jiang’s work, we propose two additional criteria to this identification method, “double promotion” and “promotional grooming.” Finally, we test all the verifiable approaches against the odds of China’s prefectural-level leaders crossing career thresholds between 2000 and 2020. The test results show that the background-based approach has limited validity and Jiang’s patronage-based approach thus requires further refinement. In contrast, our revised identification methods prove to be effective in clarifying the factional factor. This study thus proposes an improved, verified approach to identifying factions in Chinese politics and provides researchers with a reliable tool for identifying the “people factor” in the comparative study of political elites.

对于中国政治的研究者而言,关心的不是派系是否重要,而是派系如何识别?但在现实政治中,政治人物的派系归属经常隐而不显,需要学者耐心尝试解析。对中国派系的厘清也因而经历三个世代的发展:分别以“传闻轶事”、“关系基础”与“扈从联系”为标准。本文对它们进行了系统的评述,并在蒋俊彦“任内拔擢”标准的基础上,提出“不次拔擢”和“拔擢培养”的改进识别方式。继而利用2000-2020年中国地市市委书记的数据进行经验检测,发现:(1)“关系基础”的标准未必可靠;(2)“任内拔擢”的标准仍可改进;(3)“不次拔擢”和“拔擢培养”最有助厘清派系归属。换言之,本文在回顾、考察既有方案的基础上,提炼出更适用于派系归属的识别标准。本研究因此不仅有助破解中国政治中的派系之谜,还为比较政治精英研究找到识别“人的因素”的可能办法