As one of the most influential intellectual traditions in Chinese and even East Asian history, the early history of Daoxue 道學 (Learning of the Way) is a topic that has fascinated historians in recent decades. Borrowing Peter Bol's phrase, Daoxue served as a special kind of Confucianism that “claims to be reclaiming what Confucius and Mencius really meant and that often speaks through its interpretations of the ancient texts”.Footnote 1 Daoxue attracted the attention of many scholars because of the crucial role it played in shaping the intellectual and state orthodoxy of late imperial China under the rubric of “Neo-Confucianism”.Footnote 2 Chan Wing-tsit utilized the rubric to explore the orthodox “Neo-Confucian” philosophical tradition since the eleventh century. With a more inclusive philosophical perspective, William Theodore de Bary used the rubric to highlight the dominance of this orthodox philosophical and religious tradition over all Confucians in late imperial China. Hoyt Tillman and Yu Yingshi's studies inquired into the formative stage of this rubric by calling for historical studies of the early evolution of Daoxue. In his recent work, The Making of Song Dynasty History, Charles Hartman summarizes his studies on Song Daoxue historiography and demonstrates how historiography contributed to prevalent understandings of Song Confucianism for their contemporaries and for later generations.Footnote 3 It seems that recent studies have shifted to the historical and historiographical construction of early Daoxue history. This article provides a persuasive case study of the making of the evil image of an early Daoxue follower, Xing Shu 邢恕, who became a villain and, in later narratives, was excised from the early Daoxue community.
The conventional narrative of early Daoxue history attributes its origin to the thought of several celebrated scholars in the late eleventh century. These scholars include Zhou Dunyi 周敦頤 (1017–73), Shao Yong 邵雍 (1011–77), Zhang Zai 張載 (1020–77) and, most importantly, the Cheng brothers: Cheng Hao 程顥 (1032–85) and Cheng Yi 程頤 (1033–1107). Among various traditional narratives of Chinese intellectual history, the “Song Yuan ruxue'an xulu” 宋元儒學案序錄, written by the Qing historian Quan Zuwang 全祖望 (1705–55), provides a comprehensive summary of early Daoxue scholars and their positions on the general landscape of Song and Yuan intelligentsia.Footnote 4 Following the Daoxue 道學 genealogy set by previous scholars, Quan Zuwang highlighted the intellectual lineage from Cheng Yi to Zhu Xi 朱熹 (1130–1200), the greatest synthesizer of Daoxue scholarship.
However, what distinguishes Quan's Daoxue narrative is that it supplemented the Cheng-Zhu lineage with extra textual materials about the scholars in between the Cheng brothers and Zhu Xi, most of whom were Cheng Yi's students.Footnote 5 In the “Song Yuan ruxue'an xulu”, Quan arranged these students in an order of significance, depending on their fidelity to the teachings of the Cheng brothers and their personal reputations. The first is Xie Liangzuo 謝良佐 (d. 1121), then comes Yang Shi 楊時 (1053–1135), followed by You Zuo 游酢 (1053–1123), Yin Tun 尹焞 (1071–1142), Guo Zhongxiao 郭忠孝, and Wang Pin 王蘋 (1082–1153).Footnote 6 In the closing, Quan introduced the Cheng brothers’ early students, including some with disappointing reputations.Footnote 7 Among these students, Quan considered Xing Shu as the most notorious, comparable to Gongbo Liao 公伯繚, a disloyal disciple of Confucius.Footnote 8 Like Quan, many traditional and even modern historians have considered Xing Shu as the black sheep among Cheng Yi's students. Why would historians consider Xing Shu as a disloyal disciple of the Cheng brothers and how was his historical image as the black sheep constructed? The answers to these questions constitute the focus of this article.
In particular, an investigation of the construction of Xing Shu's historical image echoes recent studies on the narratives, themes, and structure of Song historical writings. Hartman persuasively argues that the received version of Song history emerges from numerous Song historiographical interventions that serve political and ideological purposes.Footnote 9 At the heart of these historiographical interventions were Daoxue and Daoxue-influenced scholars, who developed a specific historiography that left clear imprints in the products of what Hartman has called the “grand allegory” of Song history – the final product was the Yuan-compiled official dynastic history of Song (Songshi 宋史). Regarding Song history as a “grand allegory”, Hartman's work has decoded this very allegory into three thematic clusters, respectively “The Empire of Benevolence”, “The Deification of Taizu”, and “The Lineage of Evil”.Footnote 10 The first and third clusters show a clear propensity for Daoxue moral values. Concerning the third cluster, Hartman's textual analysis of several Songshi biographies in the chapter of “nefarious officials” (jianchen zhuan 奸臣傳) thoroughly reveals the influence of Daoxue historiography.Footnote 11 Most of these “nefarious officials”, such as Cai Jing 蔡京 (1047–1126) and Qin Gui 秦檜 (1091–1155), were powerful ministers and officials who demonstrated clear hostility towards Daoxue and Daoxue scholars. Despite Xing Shu's Daoxue identity, he was also one of the “nefarious officials” in the Songshi. In this light, the decoding of Xing's constructed historical image reveals how the same historiographical interventions in “the making of Song history” could also effectively operate in the case of a Daoxue scholar.
Following Hartman's endeavour in decoding the Song “grand allegory”, this article traces the construction of Xing Shu's historical image to a series of Song historiographical interventions, especially three first-hand testimonies concerning Xing's betrayal of his teacher Cheng Yi in an incident in 1097. In addition to Hartman's approach, I also borrow the French philosopher Paul Ricoeur's conceptual framework of “historiographical operations” to explicate Xing Shu's case. Building upon the conventional understanding of “historiographical operations” in the compilation of historical writings,Footnote 12 Ricoeur classified the internal mechanism of these operations into three phases: (i) the documentary phase; (ii) the explanation/understanding phase; and (iii) the representative phase.Footnote 13 Ricoeur defined the documentary phase as “the one that runs from the declaration of eyewitnesses to the constituting of archives”.Footnote 14 According to him, the documentary phase manifests itself as a multilayered structure of inhabited space, historical time, testimonies, archives, and documentary proof. The first two concepts of inhabited space and historical time offering the transcendental infrastructure for the presence of historical events and memories exceed the scope of this article.Footnote 15 However, the last three concepts are useful to our analysis of Xing Shu. Borrowing Ricoeur's conceptualization of testimonies, documentary proof, and archives, the following sections examine the Song historiographical operations of the three first-hand testimonies concerning Xing Shu's relationship with Cheng Yi, as well as different documentary proofs of Xing's image as a nefarious official in a variety of Southern Song records. Before probing into the testimonies and records, it might be useful to introduce Xing Shu's historical significance in relation to Daoxue and his historical world.
Xing Shu's political career and his relationship with Cheng Yi
The Songshi portrays Xing Shu as typical of the bureaucratic elite in the late Northern Song, but his political career was a rollercoaster ride. Xing was born in Zhengzhou (鄭州, modern Henan 河南), in proximity to the birthplace of the Cheng brothers, Luoyang 洛陽. He first studied with Cheng Hao in the late 1060s and earned the jinshi 進士 degree in 1067, when the young emperor Shenzong 神宗 (r. 1067–85) ascended the throne embracing a vision of great achievements that could only be realized with the assistance of ministers and officials for profound bureaucratic, political, and social reforms. In 1068/4,Footnote 16 the future chief minister Wang Anshi 王安石 (1021–86) entered central government and soon launched a series of reform campaigns in the name of New Policies (xinfa 新法).Footnote 17 The New Policies had immediately received criticism from some leading scholar-officials, among them Sima Guang 司馬光 (1019–86), Lü Gongzhu 呂公著 (1018–89), and Han Qi 韓琦 (1008–75).Footnote 18 Revolving around Wang Anshi and his opponents, coalitions of reformists and anti-reformists formed and set the stage for the elevation of factional conflicts in the next several decades.
Not long after Xing Shu had begun his political career, he found himself in the vortex of factional conflicts between the reformists and the anti-reformists. From the 1060s to the 1080s, Xing had been associated with the circle of anti-reformists. He maintained close relationships with Sima Guang and Lü Gongzhu, as well as with Sima Guang's son, Sima Kang 司馬康 (1050–90). Owing to Xing's lifetime friendship with Sima Kang, some Song literati thought that they were in the same examination cohort even though in fact they were not.Footnote 19 It was possible that Xing Shu became acquainted with Sima Guang in the 1070s through his connection with Sima Kang. In 1069, Xing's criticism of Wang Anshi's New Policies and his ensuing dismissal from the central government to the Prefecture Henan 河南府 also earned him a good reputation among anti-reformists, especially another political patron Lü Gongzhu.Footnote 20
Along with his exclusion from the centre of power in the late 1060s, Xing Shu gradually became one of the Cheng brothers’ principal disciples and played a central role in the early Daoxue community. The most solid evidence for Xing Shu's close affiliation with the Cheng brothers is his laudatory tribute to Cheng Hao in 1085, when Cheng died immediately after the enthronement of the new emperor Zhezong. According to Cheng Yi, he selected Xing Shu's “laudatory tribute” (xushu 敍述) from myriad relevant laudatory tributes, to supplement his own biographical sketch (xingzhuang 行狀) of Cheng Hao.Footnote 21 It is noteworthy that Cheng Yi selected only four laudatory tributes from Cheng Hao's disciples and friends as an addendum to his biographical sketch.Footnote 22 Clearly, Cheng Yi trusted and valued Xing Shu's record of his elder brother Cheng Hao. Reciprocally, Xing Shu was committed to the identity of being a Cheng brothers’ student. According to a handwritten letter by Cheng Yi, Xing Shu had some private conversations with Cheng Yi about his teacher's opinions on other disciples, including Yang Shi, who was commonly regarded by historians as the principal disciple of the Cheng brothers.Footnote 23 The fact that Cheng Yi and Xing Shu felt free to judge the former's other disciples in front of each other demonstrates the intimacy between them. Cheng considered Xing a close disciple and friend; reciprocally, Xing considered himself Cheng's principal disciple, responsible for overseeing other students in transmitting Cheng Yi's scholarship.
In 1077, Xing Shu returned to the central court after long service in several counties in Henan and Shanxi 陝西.Footnote 24 His political stance dramatically changed thereafter. Notwithstanding his close affiliation with anti-reformist politicians and their supporters the Cheng brothers, Xing Shu began to establish a rapport with the reformist leader Cai Que 蔡確 (1037–93) at the end of Emperor Shenzong's reign. Based on myriad historical sources, especially the 1138-compiled Veritable Records of Emperor Zhezong (Zhezong shilu 哲宗實錄, the Shaoxing 紹興 version), the Southern Song historian Li Tao rendered a vivid portrait of Xing Shu's engagement in an imagined “coup” about the imperial succession of Emperor Zhezong 哲宗 (r. 1085–1100) in 1085.
As Li Tao correctly pointed out, the 1085 “coup” was a fake coup, originating from the distrust between the young emperor Zhezong and his grandmother, Empress Dowager Xuanren 宣仁皇太后 (r. 1086–93). Xing Shu played an inglorious role in the construction of the coup. In brief, Xing Shu attempted to convince a few reformist ministers that Xuanren had prepared to recommend two other princes rather than Zhezong to ascend the throne after Shenzong's death, lest Zhezong should continue the New Policies initiated by his father Shenzong. In the spring of 1085, rumours regarding Xuanren's coup against Zhezong's succession were circulating around the capital Kaifeng 開封. Several key sources used by Li Tao reveal that Xing Shu was critical to the creation of these rumours.Footnote 25 Xing also implied that a major minister Wang Gui 王珪 was engaged in Xuanren's coup. However, because of his persuasion of the reformist ministers Cai Que and Zhang Dun 章惇 (1035–1105), Xing Shu successfully defeated the “coup” and thus secured Zhezong's succession.
Understandably, Xing Shu's construction of the imagined 1085 “coup” resulted in Empress Dowager Xuanren's personal hostility towards him and his dismissal from the Edict Drafter position (zhongshu sheren 中書舍人) on 1085/12/27.Footnote 26 The year 1085 played a crucial role in Xing Shu's political career. After 1085, Xing had opportunities to return to the court in the 1090s and the early 1100s. However, he never regained the reputation that he had previously earned from anti-reformists. Xing's inglorious role in the creation of the imagined 1085 “coup” fundamentally ruined his reputation as an anti-reformist and the principal disciple of the Cheng brothers. The edict of Xing Shu's dismissal, drafted by Zheng Yong 鄭雍 (1031–98), describes Xing as a man who studied ancient texts and dressed in Confucian garments, but finally betrayed his learning.Footnote 27 The edict's emphasis on Xing Shu's Confucian background refers possibly to Xing Shu's early training under the guidance of the Cheng brothers. However, in 1089, most political allies of the Cheng brothers in the central government had distanced themselves from Xing. Zhu Guangting and Wang Yansou 王巖叟 (1043–1093) as Cheng Yi's followers condemned Xing Shu for his collusion with Cai Que.Footnote 28 Zhu Guangting in his memorial drew particular attention to how Xing Shu's behaviour defied the learning of Confucius and Mencius, but imitated the tactical rhetoric of Su Qin 蘇秦 (382–284 bc) and Zhang Yi 張儀 (d. 309 bc) – two famous strategists of the Warring States period (475–221 bc).Footnote 29 By categorizing Xing Shu with Su Qin and Zhang Yi, Zhu Guangting excluded Xing from the rank of genuine Confucians (chunru 醇儒) who followed the Cheng brothers’ teachings.Footnote 30 A similar expression with slightly different wording appears in Zheng Yong's edict on Xing Shu's dismissal.Footnote 31 To Zhu Guangting, Wang Yansou, and Zheng Yong, Xing Shu had already betrayed Cheng Yi's teachings since he decided to collude with Cai Que in 1085.
Despite the general hostility to Xing Shu among Cheng Yi's admirers in the anti-reformist government of the Yuanyou 元祐 (1085–94) era, Cheng Yi himself never criticized Xing Shu, even for his misconduct in the creation of the 1085 “coup”. After Emperor Zhezong began his personal rule and swung back to the reformist side during the Shaosheng 紹聖 (1094–98) era, Xing Shu returned to the court following the recommendation of reformist leaders, especially Cai Bian 蔡卞 (1058–1117) and Zhang Dun.Footnote 32 On 1097/10/21, with the assistance of Zhang Dun, Xing Shu ascended to the post of the Vice-Censor-in-Chief 御史中丞, the head of the Censorate (yushitai 御史臺).Footnote 33 From 1097/10 to 1098/3, i.e. within half a year, Xing Shu as Vice-Censor-in-Chief had submitted over a dozen memorials on various policies.Footnote 34 Meanwhile, Xing Shu's intellectual mentor, Cheng Yi, was having the most difficult time of his life. The factional purges in the Shaosheng era involved systematic persecution of anti-reformists in both political and intellectual domains. Cheng Yi's previous affiliation with Sima Guang offered his political opponents an effective weapon to bring charges against Cheng Yi. However, Zhezong's personal hatred of Cheng Yi was the primary reason for Cheng's exile to the far south in Fuzhou 涪州 (modern Sichuan) in 1097/11.Footnote 35 Xing Shu's response to this incident eventually consolidated his historical image not only as a nefarious official but also as a vicious man. The consolidation process was rooted in the historiographical operations of three firsthand testimonies concerning Xing Shu's response to Cheng Yi's exile in 1097.
Testimonies concerning Xing Shu's betrayal of Cheng Yi
In later accounts of Xing Shu, Xing Shu's response to Cheng Yi's exile in 1097 served as critical evidence of his betrayal of his master. Most of these accounts relied upon Zeng Bu's 曾布 (1036–1107) personal diary (rilu 日錄). As one of the chief leaders of Zhezong's government, Zeng recorded many political anecdotes in his diary, including a record about Xing Shu's response to Cheng Yi's exile. The anecdotal record in Zeng's personal diary serves as the first testimony of Xing's betrayal of Cheng. It reads:
Cheng Yi was closely affiliated with Xing Shu. Xing disliked an official named Lin Xi and planned to co-opt some remonstrators to impeach Lin. Cheng Yi's exile in 1097 should be attributed to Lin's efforts. At the beginning, Lin thought that Xing Shu [as Cheng Yi's student] would definitely attempt to defend Cheng Yi. Xing's defence hence would make himself vulnerable to the political attack from Lin. In response to Cheng Yi's exile, Xing Shu told the others: “Even if the court cut Cheng Yi into ten thousand pieces, I would not save him”. Those who heard this response laughed.Footnote 36
頤素與邢恕善，而恕雅不樂林希，謀與諫官共攻之。頤編管，蓋希力。希意恕必救頤，則因以傾恕。恕語人曰：「便斬頤萬段，恕亦不救。」聞者笑之。
Zeng Bu's personal diary dated Xing Shu's statement to 1098/1/18. The exile of Cheng Yi, according to Li Tao, occurred on 1097/11/27.Footnote 37 Zeng's record of Xing Shu's betrayal hence makes perfect sense in terms of chronology. Lin Xi 林希 (1035–1101), who played a crucial role in causing Cheng Yi's exile, agreed with Zeng Bu in identifying Xing Shu with the Yuanyou group of anti-reformists. Xing's relationship with Cheng Yi made him a target for Lin Xi's political attack from late 1097 to early 1098; hence, Xing's relationship with Cheng Yi led to the exile not only of Cheng Yi but also of Xing Shu himself in 1098/4/14.Footnote 38
Zeng Bu's diary record of Xing Shu's response to Cheng Yi's exile reflected how Xing Shu was perceived by his political opponents. Notably, Zeng Bu clearly demonstrated a liking for Lin Xi during the Shaosheng years. Available evidence shows that Zeng strove to defend Lin in a private audience with Zhezong in 1098/1/23 – only five days after he had recorded Xing Shu's cruel language regarding Cheng Yi's exile.Footnote 39
The question is: did Xing Shu actually betray Cheng Yi when the latter experienced hard times? Cheng Yi himself had some reservations concerning the rumour of Xing Shu's betrayal. According to Yang Di's 楊迪 (1055–1104) (Yang Shi's son) version of Cheng Yi's dialogical records, Xie Liangzuo asked Cheng about Xing Shu's role in Cheng's 1097 exile. This dialogical record serves as the second testimony concerning Xing Shu's betrayal of Cheng Yi, but from a different angle from Zeng Bu's diary record. At the beginning of the dialogue between Xie Liangzuo and Cheng Yi, Xie blamed Cheng's exile on Xing Shu and Cheng's junior relative (zuzi 族子), Cheng Gongsun 程公孫. Cheng Yi replied that he and Xing Shu had a long friendship (qinghou 情厚) and he could not doubt Xing. Moreover, Cheng Yi compared himself with Mencius and argued that he would not be destroyed by hostility from others if he had really inherited the sagely Way.Footnote 40 Xie Liangzuo then made further comments about Xing Shu:
[Xie Liangzuo] hence commented: “Master, despite Xing Shu's bad behaviour, he surely would not be so bad as to harm you”.
Master [Cheng Yi] responded: “Yes. Xing Shu sent me a letter arguing that he had defended me in front of the powerful ministers. I do not understand why Xing as the head of the Censorate spoke like this. In fact, I am not concerned about whether or not he would harm me. My concern is whether or not he would make an effort to save me. That is the key point”.
[Xie Liangzuo] further asked: “Xing was your student for a long time. But it seems that he was ignorant and therefore later became extremely unscrupulous”.
The Master responded: “It is difficult to say that Xing is totally ignorant. But since his moral principle could not overcome his desire for power, he ended up becoming like this”.Footnote 41
因問：「邢七雖為惡，然必不到更傾先生也。」
先生曰：「然。邢七亦有書到某云：『屢於權宰處言之』。不知身為言官， 卻說此話。未知傾與不傾，只合救與不救，便在其間。」
又問：「邢七久從先生，想都無知識，後來極狼狽。」
先生曰：「謂之全無知則不可，只是義理不能勝利欲之心，便至如此也。」
Cheng Yi's conversation with Xie Liangzuo occurred after Zeng Bu's personal diary. Possibly, Zeng Bu first recorded the rumour about Xing Shu's betrayal of Cheng Yi in his personal diary. After Xie Liangzuo heard similar rumours, he approached Cheng Yi and sought his master's clarification. Xie's action was understandable because he had known Xing Shu since at least the 1080s, when Cheng Hao was still alive. Xie was surely concerned about the rumour of Xing Shu's role in Cheng's exile. Despite Cheng Yi's dissatisfaction with Xing Shu's hesitation to save him, his positive reply to Xie Liangzuo's judgement against Xing Shu clearly demonstrated that Cheng was still inclined to believe Xing's good intentions.
Furthermore, Cheng Yi mentioned that Xing Shu's private letter explained his plan to rescue Cheng. Xing Shu's letter was lost, but its basic tone coincides with his memorial submitted to the Historiography and Veritable Records Institute (shilu yuan 實錄院) under Huizong's 徽宗 (r. 1100–26) reign in 1101. Xing composed this memorial, namely the “Shen shilu yuan zhuang” 申實錄院狀, in 1101/8 to explain to Huizong his relationships with major anti-reformists, especially Sima Guang, Lü Gongzhu, and his master Cheng Yi. This memorial serves as the third testimony concerning Xing Shu's betrayal of Cheng Yi. In the memorial, Xing Shu recalled his explanation in an imperial audience with Zhezong three years earlier:
Because Lin Xi and Zhang Dun considered Shu [Xing Shu's self-designation] as a former student of Cheng Yi, they framed a case, in front of Zhezong, against Cheng, that led to his exile to Fuzhou. Before Cheng Yi's exile was actually decreed, Lin and Zhang slandered me in the meeting of ministers and said, “If the teacher is [evil] like this, what can we expect from the student?” Upon hearing this, Shu petitioned to Zhezong: “I have no idea what kind of accusation was made in relation to Cheng Yi's exile. Indeed, I regarded myself as Cheng's student in the past. However, after Cheng Yi and I met in the capital in 1080, we never met again in these twenty years. Moreover, Cheng Yi's and my official careers did not overlap during the Yuanyou era. How could I be a target of Zhang Dun's charge against Cheng Yi? Because of his affiliation with Lin Xi, Zhang charged Cheng Yi and me out of personal bias. How could the people of the empire accept this as just? Thus, I think it is necessary for me to report all these details”.Footnote 42
程頤貶涪州，亦是林希與章惇以為恕素師事頤，故遂於哲宗前陷成頤罪。未從貶頤，乃於執政大臣聚會處見詆云：「師既如此，為弟子者當如何？」恕尋聞其說，亦曾對哲宗皇帝開陳云：「程頤之貶，臣不知以何罪。臣於頤，昔者實以師友之間處之，但自元豐三年，頤曾到京師與之相見後，至今二十年不曾相會。元祐間與頤又不同進退。然則惇雖罪頤，焉能中臣？但惇緣希故，挾情用刑，則天下安得心服？理當奏知。」
Xing Shu's “Shen shilu yuan zhuang” reiterated what he mentioned in his letter to Cheng Yi: he had defended Cheng Yi in front of Emperor Zhezong and other powerful ministers. Moreover, a comparison between the “Shen shilu yuan zhuang” and Zeng Bu's diary reveals a clear inconsistency. Zeng's diary, with all its simplicity and vivid portrait, depicts Xing Shu as a cold-blooded opportunist who firmly determined to sacrifice Cheng Yi in order to save himself. By contrast, Xing Shu's own testimony in the “Shen shilu yuan zhuang” reflects a more complicated picture of that opportunistic mentality. Admittedly, Xing's memorial was primarily based on self-interest. His testimony that he had not met Cheng Yi after 1080 was false, because both he and Cheng Yi attended Cheng Hao's funeral in 1085. Moreover, what concerned Xing the most was that he would not be implicated in Cheng Yi's case. Despite his denial of involvement in Cheng Yi's case, Xing still used the word “framed” or “trapped” (xian 陷) in referring to the case and questioned Cheng's exile.
Some may argue that Xing fabricated his defence of Cheng Yi in his memorial to maintain his reputation among anti-reformists. But there was no reason for Xing Shu to do so in the pro-reformist era from the late 1090s to the early 1110s, especially if he was as cold-blooded and selfish as he was portrayed in Zeng Bu's diary. In my opinion, the testimony provided by Xing Shu in the “Shen shilu yuan zhuang” conveys an individualized testimony indicating Xing Shu's psychological struggle underlying his opportunistic acts in Cheng Yi's exile. On the one hand, Xing had distanced himself from Cheng Yi since 1097, as he admitted in his own testimony three years later. On the other hand, he also unveiled the unfair charges against Cheng Yi; moreover, he indirectly defended Cheng in the same testimony by pointing out the factional struggles behind Cheng Yi's exile. In the absence of other evidence, Xing Shu's testimony still reflects his sympathetic attitude towards Cheng Yi in 1101, when the new reign of Emperor Huizong was swinging to the reformist side.Footnote 43
In sum, direct evidence for neither Xing Shu's defence of Cheng Yi nor Xing Shu's betrayal to Cheng Yi can be found. It is thus impossible to reconstruct the full picture of the betrayal based on existing sources. Nevertheless, by juxtaposing the three testimonies concerning Xing Shu and Cheng Yi, it is still possible to study the historiographical operations in their documentary phase, through which later memories of Xing were constructed, distorted, and manipulated by different individuals for their respective purposes.
Zeng Bu's diary from early 1098 serves as the earliest testimony of Xing Shu's betrayal of Cheng Yi. Zeng's testimony, which was based on rumours rather than reliable records, was coloured by factionalism, especially the fierce conflict between Xing Shu and Lin Xi. The dialogical record of Cheng Yi's conversation with Xie Liangzuo, which was documented by Yang Di before 1104 (Yang died that year), offers another piece of testimony that counters Zeng Bu's diary. Xing Shu's personal testimony in 1101 contradicts Zeng Bu's testimony in a similar way. The existence of several testimonies opens space for a critique of the testimonies.Footnote 44
The three testimonies competed with one another soon after their emergence. Xing Shu's own testimony in 1101 was possibly cited in five written records composed by himself and his descendants, including: 1) the family biography of Xing Shu (jiazhuan 家傳); 2) the record of Xing Shu's words and deeds (yanxing lu 言行錄); 3) the biographical sketch of Xing Shu (xingxhuang 行狀); 4) the authentic record of Xing Shu's career (xingshi 行實); and 5) Xing Shu's self-defence record (zibian lu 自辨錄).Footnote 45 All the five records were lost and only their excerpts were preserved by Li Tao in his XCB. Item 5, judging by its name, should be an elaborated version of Xing Shu's own records of his political career, including the 1101 testimony. Items 1, 2, 3, and 4 were composed by Xing Shu's descendants, mostly his grandson Xing Yi 邢繹.Footnote 46 Southern Song historians cast doubt on the credibility of these written records and argued that they were deliberately composed to highlight Xing Shu's good deeds.Footnote 47 In other words, these records served to camouflage Xing Shu's evil intent. Southern Song historians had the same doubts about Xing's 1101 testimony and did not regard it as a reliable source. Since Xing Shu's writings were never formally published, the 1101 testimony was consigned to oblivion in the late twelfth century.Footnote 48
Compared with Xing Shu's own testimony, Cheng Yi's conversation with Xie Liangzuo as a firsthand testimony gained more popularity among scholar-officials, especially Daoxue scholars. Thanks to its recorder, Yang Di, the son of the celebrated Daoxue pioneer Yang Shi, this conversation was collected and preserved in various early versions of the Cheng brothers’ surviving writings (yishu 遺書). It was circulated among Daoxue-interested scholars from the late Northern Song and was eventually edited into the Jinsi lu 近思錄 in 1175. The Jinsi lu text reads:
[Xie Liangzuo] asked: “Xing Shu has been your student for a long period. Lacking in necessary wisdom, Xing put himself in an extremely embarrassing situation”.
Yichuan answered: “It is not appropriate to say that Xing Shu is totally ignorant. But Xing's morality was overpowered by his desires and personal interests. Thus, he has come to this end”.Footnote 49
問：「邢恕久從先生，想都無知識，後來極狼狽。」
伊川曰：「謂之全無知則不可，只是義理不能勝其利欲之心，便至如此。」
The editor of the 1175 edition of Jinsi lu trimmed Yang Di's record of Cheng Yi's conversation with Xie Liangzuo down to the texts above. In Yang Di's original record, Cheng Yi expressed confidence in his friendship with Xing Shu and Xing's good intentions to save him. After hearing Cheng Yi's explanation, Xie Liangzuo commented that Xing Shu would not be so bad as to harm Cheng Yi.Footnote 50 However, the 1175 edition of Jinsi lu deletes two key phrases from Yang's record that absolved Xing Shu of direct complicity in Cheng Yi's exile: it does not mention Xie Liangzuo's positive comment about Xing Shu, i.e. Xing “surely would not be so bad as to harm master Cheng Yi” 必不到更傾先生也; it also says nothing about Xing Shu's private letter to Cheng Yi that explained Xing's plan to rescue his teacher. Instead, the Jinsi lu text correlates Xie Liangzuo's suspicion about Xing Shu with Cheng Yi's final response to Xie. Therefore, the abridged version in the Jinsi lu reads like Cheng's criticism not only of Xing's moral fallacy but also of his betrayal.
Zhu Xi was predominantly responsible for the abridged version in the 1175 edition of Jinsi lu.Footnote 51 Interestingly, in his draft of the Yiluo yuanyuan lu, Zhu documented the entire conversation between Xie Liangzuo and Cheng Yi about Xing Shu.Footnote 52 Indeed, the Yiluo yuanyuan lu record duplicates Yang Di's record.Footnote 53 The textual differences between the Yiluo yuanyuan lu and the Jinsi lu could be partly attributed to the nature of each text. The Yiluo yuanyuan lu, which was first drafted by Zhu Xi in 1173, serves as a collection of historical records about the Cheng brothers and their students. Therefore, it documents more details than the Jinsi lu, which was designed by Zhu Xi and Lü Zuqian 呂祖謙 (1137–81) as Daoxue teaching material. Notably, the compilation of the Jinsi lu in 1175 was accompanied by the famous intellectual debate between Zhu Xi and the Lu brothers (Lu Jiuyuan 陸九淵 (1139–93) and Lu Jiuling 陸九齡 (1132–80)) at Ehu 鵝湖 (literally, the Goose Lake, modern Jiangxi). In the Ehu debate, Zhu Xi and the Lu brothers failed to achieve a consensus. Although their criticisms of each other softened in the late 1170s,Footnote 54 the year 1175 still witnessed Zhu Xi's shift towards a more exclusive attitude in defining the Daoxue community. After the Ehu meeting, Zhu's final revision of the Jinsi lu in 1175/8 was characterized by a strict selection in the words of Daoxue masters.Footnote 55 Cheng Yi's positive comments on his “treacherous” student Xing Shu might be seen as a stain on Cheng's reputation. For this reason, Zhu Xi apparently trimmed down Cheng Yi's comments to what they look like in the extant Jinsi lu.
In terms of textual transmission, the 1175 edition of the Jinsi lu had a direct impact on Southern Song writings about the relationship between Xing Shu and Cheng Yi. The earliest extant annotation on the Jinsi lu, compiled by a Fujian scholar named Ye Cai 葉采 in 1248, supplements the Jinsi lu records with other textual materials, especially dialogical records.Footnote 56 In his supplementary notes on the Jinsi lu records of Xing Shu, Ye stated that he used other dialogical records and official history (guoshi 國史) based on court archives.Footnote 57 Ye definitely cited some information from Yang Di's dialogical record. For example, he mentioned how Xie Liangzuo also considered Cheng Gongsun a culprit in Cheng's exile. Indeed, Ye Cai's record adds a historical edge to the original Jinsi lu text. However, like Zhu Xi, Ye Cai ignored the two key passages in Yang Di's record that absolved Xing Shu of direct complicity in Cheng Yi's exile. With more historical background, the 1248 edition of the Jinsi lu ironically reinforces its readers’ impression of Xing Shu's betrayal of Cheng Yi.Footnote 58
The most influential testimony concerning Xing Shu's betrayal of Cheng Yi is Zeng Bu's personal diary. The Southern Song bibliographer Chen Zhensun 陳振孫 (1179–1262) recorded Zeng Bu's diary in two juan 卷, respectively referring to the diaries of 1094 and 1100.Footnote 59 A later reconstruction of Zeng's diary was based on its excerpts left in the voluminous Ming collection of Yongle dadian 永樂大典, which only covers the period from 1099/3 to 1100/7.Footnote 60 Compared with Chen Zhensun and others, Li Tao had an opportunity to access the complete copies of Zeng Bu's diary, since in his XCB he quoted Zeng's 1098 diary in Xing Shu's case. More importantly, Li Tao's quotation corroborated Xing Shu's betrayal when Li transformed the anecdotal record in Zeng's diary into part of his annalistic records of Song history. The oral testimony about Xing Shu's betrayal, first recorded by Zeng Bu, was documented by the eminent historian Li Tao and thus shaped later writings of early Daoxue history.Footnote 61
After Li Tao's XCB, another Southern Song historical record also cites testimonies about Xing Shu's betrayal. It is the Daoming lu 道命錄, which was compiled by the pro-Daoxue historian Li Xinchuan 李心傳 (1166–1243) in 1239 and was re-compiled by Cheng Rongxiu 程榮秀 (1263–1333) (a Yuan local scholar and a descendant of Cheng Yi) in 1333.Footnote 62 As Charles Hartman persuasively argued, a large portion of the extant version of the Daoming lu should be regarded as the work of Cheng, instead of Li.Footnote 63 In this light, the inclusion of the 1097 imperial edict that promulgated Cheng Yi's exile to Peizhou in the extant Daoming lu may also be attributed to Cheng Rongxiu's work.Footnote 64 However, the commentary on the edict also includes some of Li Xinchuan's original comments. The commentary on the 1097 edict in the Daoming lu, which cites Cheng Yi's conversation with Xie Liangzuo, was possibly Li Xinchuan's original words.Footnote 65 Regarding Cheng Yi's exile, Li's citation deleted several key phrases in support of Xing Shu's innocence, including Cheng's statement about Xing's good intention to save him, as well as Xie Liangzuo's positive comment about Xing. Moreover, Li quoted Zeng Bu's record of Xing Shu's cruel reflection on Cheng Yi's exile, including Xing's notorious statement: “Even if the court cut Cheng Yi into ten thousand pieces, I would not save him”.Footnote 66 Through a skilful collage of Cheng Yi's conversation and Zeng Bu's personal diary, Li Xinchuan created a seamless historical narrative of Xing Shu's betrayal.
The Daoming lu record also echoes Zhu Xi's conversation with a student. According to the student's record, Zhu Xi said: “Xing Shu once told Zhezong: ‘I previously regarded Cheng Yi as a teacher and friend. But now even if the court cut Cheng Yi into thousands of pieces, I would not save him’” 恕言於哲宗: “臣於程某嘗事之以師友，今便以程某斬作千段，臣亦不救.Footnote 67 The statement consisted of two parts which have been cited from the two earlier testimonies of Xing Shu's self-declaration memorial and Zeng Bu's personal diary.Footnote 68 Zhu cited Xing Shu's self-declaration out of its original context of the 1101 memorial and connected it with the oral testimony in Zeng Bu's 1098 diary. Together with the Daoming lu, Zhu Xi's conversation with his student set the tone for Xing Shu's betrayal in later historical sources.
“Documentary proof” of Xing Shu's evil in the Southern Song written records
In Southern Song, the historical works of Li Tao, Li Xinchuan, and Zhu Xi consolidated the memory of Xing Shu's betrayal of Cheng Yi based on the Northern Song testimonies. Intentionally or not, these Southern Song works stigmatize Xing Shu's historical image. Borrowing the term from Paul Ricoeur again, these works serve as “documentary proof” for later historians’ cognition of Xing Shu's nefarious character.Footnote 69 In modern historiography, the transition from testimonies to documentary proof involves historiographical operations such as textual criticism. Southern Song historians constructed their own documentary proof about Xing Shu by selecting and modifying related testimonies.
In addition to historians, some Southern Song scholars also participated in the construction of Xing Shu's image as the traitor of the early Daoxue community. For example, Chen Changfang 陳長方 (1108–48) documented that the Daoxue master Shao Yong had once criticized Xing Shu as an insincere person. Therefore, Shao refused to teach his metaphysical concepts to Xing.Footnote 70 However, in Xing Shu's postscript to Shao Yong's collected works, he stated that he had learned from Shao Yong, but he was unable to complete the learning because of his wandering life.Footnote 71 Shao Yong's poems to Xing Shu also demonstrate how Shao treasured his friendship with Xing.Footnote 72 Because Chen Changfang's anecdotal record was corroborated by Zhu Xi in the latter's conversation with students, it gained popularity among Southern Song literati.Footnote 73 In contrast, Shao Yong and Xing Shu's own testimonies were forgotten in later generations.
Other Southern Song literati tended to record Xing Shu's betrayal of Cheng Yi as a cautionary tale of betrayal. Wang Mai 王邁 (1184–1248), an outspoken critic of Emperor Lizong's 宋理宗 (r. 1224–64) appointment of ministers such as Shi Songzhi 史嵩之 (1189–1257), cited Xing Shu in a short poem as an example of people who failed to follow the righteous way.Footnote 74 Wang wrote the poem as a way of consoling himself about his demotion by the court.Footnote 75 In the poem, Wang categorized himself into the party of Fan Zhongyan 范仲淹 (989–1052), who represented true worthies (xianren 賢人). On the opposite side, Wang quoted Xing Shu as the bad example of someone who betrayed his own teacher because of ignorance.Footnote 76 Likewise, Wang Yinglin 王應麟 (1123–1296), in his analysis of Sima Guang's 司馬光 (1019–86) Zizhi tongjian 資治通鑑, also cited Xing Shu as a typical example of betrayal, but from a different perspective. In Wang's view, the betrayal from Xing Shu would not diminish Cheng Yi's reputation as a genuine Confucian.Footnote 77 It seems that both Wang Mai and Wang Yinglin considered Xing Shu's betrayal of Cheng Yi as a fact.
Other Southern Song documentation of Xing Shu's political career contributed to the stigmatization of his image. Shao Bowen's detailed records of Xing Shu's collusion with Cai Que in the 1085 succession crisis, entitled the Bianwu lu 辯誣錄, depicted Xing as an ambitious opportunist. However, as Li Tao had already noted, most records in the Bianwu lu were unreliable because they were established upon rumours, anecdotes, and Shao's incorrect dating of key events.Footnote 78 A textual comparison of Shao Bowen's records with other private and official records of the 1085 succession crisis is beyond the scope of this research. But Shao's Bianwu lu did successfully categorize Xing Shu with the group of villainous reformists. Given Shao Bowen's identity as Shao Yong's grandson, his indirect testimony against Xing Shu – mostly based on rumour – corroborated Xing Shu's “imagined betrayal” of Cheng Yi and even the whole Daoxue community.
Following Shao Bowen's Bianwu lu, Southern Song scholars constructed more documentary proof of Xing Shu's nefarious character. In his essay on Qin Gui's postscript to Chen Shidao's 陳師道 (1053–1101) collected works, the early Southern Song literati Sun Di 孫覿 (1081–1169) portrayed Xing Shu as a formidable icon of evil.Footnote 79 Specifically, Sun doubted Qin Gui's record of a story about Xing Shu and Zeng Gong 曾鞏 (1019–83). According to Qin's record, Zeng recruited Xing to draft the official chronicle of Emperor Yingzong, but was dissatisfied with Xing's draft. Therefore, Zeng made direct corrections to Xing's draft and criticized it as “bringing disorder to the Way” (luandao 亂道).Footnote 80 Nevertheless, Sun Di questioned the credibility of Qin's record, because Xing's evil plot in the 1085 succession crisis left an unfading shadow in the minds of most Song literati. In Sun's words, “even his contemporaries cringed with fear when they heard Xing Shu's name” 至今人聞其名, 往往縮頸.Footnote 81 If so, it is difficult to imagine how Zeng Gong dared to offend Xing Shu in such a straightforward way.
Sun Di's comments on Xing Shu seem commonsensical, but in fact they are not. The obvious reason for this is that it was Zeng Gong who invited Xing Shu to assist him in compiling the official chronicle of Yingzong in 1081.Footnote 82 If Zeng was afraid of Xing, he would not have invited Xing at the very beginning. Moreover, in 1084/11, the court promoted Xing to the Vice Director of Bureau of Operation (zhifang yuanwai lang 職方員外郎).Footnote 83 Afterwards, Xing was released from the compilation project of Yingzong's official chronicle. In other words, the collaboration between Zeng Gong and Xing Shu regarding the compilation project could only have been possible before 1084/11, when Xing Shu was still a reputable scholar among anti-reformists. Zeng had no reason to be afraid of Xing Shu and his evil plot in the 1085 coup because the coup had not yet happened at the time of their collaboration in the compilation project.
Indeed, there is another version of the same story between Xing Shu and Zeng Gong. Wang Mingqing 王明清 (1127–?), whose career started slightly later than Sun Di's, documented how Zeng Gong (based on another piece of Qin Gui's writing) had corrected and criticized Xing Shu's draft of Yingzong's official chronicle.Footnote 84 According to Wang's quotation of Qin's words, Xing Shu was at first reluctant to accept Zeng Gong's criticism and questioned Zeng's corrections. However, after re-reading Zeng's impromptu corrections on his draft, Xing understood the merit of Zeng's writing skill. Eventually, Xing became an excellent writer whose writings could rival the celebrated literati of his time.Footnote 85
Like Sun Di, Wang Mingqing questioned the credibility of Qin Gui's record. Regardless of the difference in their quoted sources – perhaps Qin Gui had presented the story in different ways – Wang's comment did not mention anything about Xing Shu's personal character. Instead, Wang argued that in the early 1080s, Xing Shu was not responsible for the compilation project of official chronicles simply because it was not part of his official duty.Footnote 86 Without any preconception, readers of Wang Mingqing's comment would not have a bad impression of Xing Shu. Wang's comment contrasts sharply with Sun Di's remark that attributed the unreliability of Qin Gui's record to Xing's notoriety. Despite Wang Mingqing's relatively objective documentation, it was Shao Bowen and Sun Di's “documentary proof” that set the dominant tone for later memories of Xing Shu. This “documentary proof” dovetailed with the rise of pro-conservative rhetoric during the Shaoxing era, in which reformists in the late Northern Song were seen as responsible for the fall of the dynasty during the Jurchen invasion.
Notwithstanding Xing Shu's close affiliation with pro-conservative thinkers like Cheng Yi and Shao Yong, his possible engagement in the 1085 coup and his post-1085 connection with reformists rendered him an easy target to blame for the deteriorating politics in the late Northern Song. Wang Cheng 王稱, a Southern Song historian of the twelfth century, commented on Xing Shu in his private history of Northern Song. Specifically, Wang Cheng emphasized how Xing Shu shared with other reformists (Cai Que and Zhang Dun) the same opportunistic character of pursuing self-interest. In Wang's opinion, Xing was the Song counterpart of Jiang Cong 江充 (d. 91) and Xi Fugong 息夫躬 of the Han dynasty, and Li Xun 李訓 (789–835) and Zheng Zhu 鄭注 (d. 825) of the Tang dynasty – four typical examples of opportunists in the traditional understanding of nefarious officials.Footnote 87
Thus, the “documentary proof” offered by Shao Bowen, Sun Di, and Wang Cheng consolidated the Southern Song literati's impression of Xing Shu's nefarious character. Zhu Xi's conversations with his students best illustrate this kind of impression. These conversations were collected by Zhu's students and admirers beginning in the Jiading 嘉定 era (1208–24). Among the early collections of Zhu Xi's conversations, the one compiled by Huang Shiyi 黃士毅 was conventionally designated as the Shulei 蜀類 because it was printed in the Mei Prefecture of Sichuan 四川眉州 in 1219.Footnote 88 An extant Korean edition preserves most of the contents of the Shulei. In the Shulei edition, Zhu Xi's most severe criticism on Xing Shu falls under the category, “Disciples of Cheng Yi” 程子門人. It says:
[The student]: Huang Lü and Xing Shu lived in the Imperial University when they were young. Xing was definitely smart and outstanding. Huang was honest and diligent. Yet both men became so unscrupulous later on.
[Zhu Xi]: They were indeed good learners. But they could not behave like gentlemen. Huang was still a straightforward person. He was only ignorant, lacking in intelligence, and excessively career minded. Therefore, he was lured by Xing Shu to the evil side. Xing, on the other hand, had an intent to do evil and was more talented, hence his fault was greater.Footnote 89
問：「黃履邢恕少居太學，邢固俊拔，黃亦謹厚力學，後來二人卻如此狼狽。」
曰：「它固會讀書，只是自做人不好。然黃卻是箇白直底人，只是昏愚無見識，又愛官職，故為邢所誘壞。邢則有意於為惡，又濟之以才，故罪過多。」
Zhu Xi's comment here, again, stigmatizes Xing Shu's political career and his personal character. In 1270, when Li Jingde integrated various editions of Zhu Xi's conversations into a compendium edition, he rearranged a lot of entries and categories in the original editions, including the Shulei. In Li Jingde's edition, Zhu Xi's comment on Xing Shu does not fall into the “Disciples of Cheng Yi” category; instead, it was put under the category of the “Appointment of officials from Xining to Jingkang” 自熙寧至靖康用人.Footnote 90 Intentionally or not, the Li edition excludes Xing Shu from the community of Cheng Yi's students. Coupled with the circulation of the Li edition, later scholars and literati perceived Xing Shu as a political villain rather than Cheng Yi's student. When Zhen Dexiu 真德秀 (1178–1235) compiled brief biographies of the disciples of the Cheng brothers in 1259, he too excluded Xing Shu from the list of disciples.Footnote 91 In fact, Daoxue scholars’ writings after the early thirteenth century rarely mentioned Xing's intellectual connections with Cheng Yi and the early Daoxue community. The memory of Xing Shu as the early student of the Cheng school gradually receded into obscurity.
Concluding remarks
The Southern Song scholar-official Ni Si 倪思 (1147–1220) said that Xing Shu's contemporaries initially considered him to be a “good man” (shanshi 善士) because of his affiliations with Cheng Yi and Sima Guang; however, after Xing joined the reformist party, people considered him as an evil man. In contrast to Xing, Zhang Shangying 張商英 (1043–1121), who began in the reformist party, gained a good reputation during Huizong's reign. In one sentence, Ni Si summarized the different fates of Xing Shu and Zhang Shangying: “Good and evil are not constant. They all depend on people's individual choices” 善惡非有定也，在人而已.Footnote 92 Certainly, people can choose their own way in conducting good or evil deeds. However, the judgement about “good or evil” is more dependent on later perceptions.
Since his involvement in the 1085 coup, Xing Shu was portrayed by anti-reformists as an evil man. Zeng Bu's 1098 diary on Xing Shu's betrayal of Cheng Yi served as a perfect testimony to corroborate this image, despite its anecdotal nature and its obvious contradiction with Xing and Cheng's own testimonies. Through sophisticated historiographical operations, especially collaging testimonies, Southern Song historians constructed an internally consistent narrative of Xing Shu's betrayal of Cheng Yi. Li Tao, Li Xinchuan, and Zhu Xi exemplified this construction. Other scholars and literati, such as Shao Bowen, Sun Di, and Wang Cheng, created additional “documentary proof” for this narrative about Xing, in which some selected and modified records had assumed authority over others.
In the Yuan-compiled Songshi, completed in 1345, the compilers incorporated particular blocks of text from Song official and private sources to compose Xing Shu's biography. To a substantial extent, Xing's biography reproduces the text of his official biographies in the State History of the Four Courts of Shenzong, Zhezong, Huizong, and Qinzong (Sichao guoshi 四朝國史) and the new versions of the Veritable Records of these four courts. As many historians have persuasively argued, anti-reformist scholars manipulated the compilation and writing of the State History of the Four Courts and the new versions of the Veritable Records during the early Southern Song.Footnote 93 Despite the loss of these official records, it is imaginable that texts in these records generally demonstrated hostility towards reformists like Xing Shu. Along with the State History of the Four Courts and the new versions of the Veritable Records, Yuan compilers of the Songshi also integrated private records such as Wang Cheng's Dongdou shilüe and Shao Bowen's Bianwu lu to reinforce Xing Shu's nefarious image.Footnote 94 Although the Yuan compilers acknowledged that Xing had been an active member within the circle of the Cheng brothers before 1085, they highlighted Xing's political ambition and especially his betrayal of Cheng Yi as natural proof of his nefarious character.Footnote 95
As Paul Ricoeur has said, “The testimony is asked to prove itself. Thus, it is testimony that brings aid and assistance to the orator or the historians who invoke it”.Footnote 96 Nevertheless, some testimonies were never given the opportunity to prove themselves, because historians rewrote or simply omitted these testimonies while constructing related memories. The case of Xing Shu demonstrates the effectiveness of Song historiographical operations, through which contemporary historical testimonies were manipulated, documented, and archived for later recall. The deconstruction of such archived memories not only reveals the original testimony but may also shed new light on a more comprehensive understanding of early Daoxue history if more historians would join the unpacking of archived memories.
Acknowledgements
I would like to express my gratitude to Professors Charles Hartman, Ari Levine, Ming-kin Chu, Jaeyoon Song, and Zoe Lin for their insightful suggestions on the early draft of this article that has been presented in the 2021 Association for Asian Studies Conference.
Funding information
This article is part of my research project “Re-thinking the Song Yuan xue'an: The Formation of the Cheng School as a Disciplinary Matrix and Its Development in the Early Twelfth Century”, funded by the Research Grants Council of Hong Kong (General Research Funding, 2020–2022, Project No. 14608519).
Open access
