This volume presents an English translation of and a comprehensive introduction to the Shangshu dazhuan 尚書大傳, a body of textual fragments inherited from a commentarial tradition on the canonical Shang shu 尚書 that originated in the Western Han period. The affiliation of the Shangshu dazhuan with the canonical text has been rather loose, which makes it an exceptional case in the history of Chinese commentarial texts: one can still observe important disagreements between the Shang shu and the Shangshu dazhuan fragments today, which betoken their separate transmission paths since antiquity. Although the Shangshu dazhuan was an important object of study during the early empires, by the Southern Song period, it survived only in an incomplete form. By the eighteenth century, when the Shangshu dazhuan began to attract serious attention from philologists, its text existed only in such a sorry state that no meaningful structure could be discerned in it. Accordingly, Sun Zhilu 孫之騄 (fl. 1722–1735), the author of the first modern edition, preferred to reconstitute (ji 輯) it anew, relying on quotations from the Shangshu dazhuan preserved in ancient and medieval texts. Different editions of the Shangshu dazhuan available to scholars today originate from such Qing-period reconstitution efforts. In recent years, however, the Shangshu dazhuan has been receiving less attention than it deserves, and even specialized scholarship devoted to the formation and transmission of the Shang shu frequently disregards its evidence. The book by Lin and Vankeerberghen brings the Shangshu dazhuan back into the spotlight. The importance of this publication is not confined to Western scholarship alone, for no recent studies of comparable depth and breadth exist in the Chinese or Japanese languages.

The eighty-six-page introduction presents the remarkably complex history, structure, and contents of the Shangshu dazhuan in a lucid and readable way. Understandably, the central figure in the introduction is Master Fu 伏生 (mid-third–second centuries BCE), a scholar to whom tradition assigns both the recovery of the text of the Shang shu after the brief and tumultuous reign of the Qin emperors and the authorship of the Shangshu dazhuan. The introduction critically challenges the simplistic assumptions of the tradition. Lin and Vankeerberghen duly note, first, that the relationship between the Qin-period and Han-period scholarship of the Shang shu was characterized more by continuity than rupture—meaning that the text of the Shang shu was not necessarily lost and may not have been re-recorded from scratch from Master Fu’s memory (xxvi–xxxii)—and, second, that the traditions of the Shang shu scholarship claiming to originate from Master Fu may have in part been retrospectively constructed (xxxii–xxxviii). The introduction illustrates the complex nature of these traditions through the analysis of the structure of the Shangshu dazhuan, which consists of multiple textual layers with different degrees of authority, including such peculiar elements as “commentary within commentary,” “explanation within commentary,” and “enunciations by Confucius” (xxxviii–xlv). The mention in the Shangshu dazhuan of Shang shu chapters absent from the canonical Shang shu and fragments of an alternative sequential outline (xu 序) complicate the history of the Shang shu itself, testifying to the continuing variance in the text of the canonical collection during the Western Han period (xlii–xliv).

Anyone who has consulted the Chinese editions of the Shangshu dazhuan knows that the text is challenging: reading the disjointed fragmentary passages requires not only an excellent knowledge of the canonical Shang shu, but also a solid understanding of Han-period intellectual history. Without this knowledge, it is difficult to make sense of the text’s contents. In the presentation of Lin and Vankeerberghen, this problem is solved through uniformly presented and accessibly written annotations. These annotations dramatically lower the entry barrier, paving a way for the broader use of the Shangshu dazhuan by scholars other than Shang shu specialists. The translation of each chapter is preceded by a brief, useful introduction summarizing the chapter’s contents in the canonical Shang shu and the main features of the fragments preserved in the Shangshu dazhuan. The English translation conveniently appears side-by-side with the Chinese original, according to the convention of the Classics of Chinese Thought series. Additionally, in the Chinese text, Lin and Vankeerberghen highlight the Chinese characters that the Shangshu dazhuan appears to have incorporated from the canonical Shang shu, making it easy to observe which parts of the canonical text the Shangshu dazhuan is explaining, and to what extent it engages in creative elaboration.

The volume succeeds particularly in showcasing the differences between the interpretations of the Shang shu in the Han period and in modern scholarship. The Shangshu dazhuan contains little information that could be classified under the labels of “history” (understood as the study of the past focused on chronologically arranged facts) or abstract “philosophy” (understood as speculative knowledge that can be appropriated and applied by a learner of any social status). Instead, the Shangshu dazhuan is emphatically preoccupied with the practicalities of rulership. Through the lens of the Shangshu dazhuan, the canonical Shang shu re-emerges as a normative text, valued for the precedents set by the sage rulers of the past which later rulers were obliged to follow. This perspective significantly rectifies simplistic approaches to the Shang shu that ignore the early exegetical tradition, resulting in fundamentalist bias and historically untenable readings.

One question that Lin and Vankeerberghen should have investigated more deeply is the nature of the text of Shangshu dazhuan in the edition by Wang Kaiyun 王闓運 (1833–1916), which was chosen by Lin and Vankeerberghen as the basis of their translation and study. Wang Kaiyun’s text was in turn based on an earlier edition by Lu Jianzeng 盧見曾 (1690–1768). Questions surround both the contents and the origins of Lu’s edition. In Lu’s edition (and consequently, in the volume by Lin and Vankeerberghen), the passages listed under the titles of individual Shang shu chapters are structurally incomplete and often show no relationship to one another—a feature characteristic of texts Qing scholars reconstituted (ji) from quotations, such as Shiben 世本 (Origins of Lineages) or the so-called “ancient” version of the Zhushu jinian 竹書紀年 (Bamboo Annals), to mention two well-known examples. However, unlike these reconstituted texts, Lu’s edition of the Shangshu dazhuan indicates no sources for most of its passages. What could be the origins of such a text? One hypothetical possibility is that Lu based his edition on a text that was damaged, but in such a way that its overall structure was not affected: the division into juan and chapters somehow remained intact. Such peculiar selective damage, however, is unlikely. It is difficult to imagine how it would have occurred in the context of manuscript or print culture: typically, damage leads to disappearance of continuous textual units, such as chapters and the larger juan containing multiple chapters, but not to disappearance of individual passages within chapters. Besides, this kind of damage is very different from the kind described by Sun Zhilu (lxxviii), who indicated that the fragmentary version of the Shangshu dazhuan he had at his disposal did not suffice to undergird any reconstruction. Lu Jianzeng never explained the provenance of the manuscript on which his edition was based, only mentioning that he acquired it “from a book collector in Wuzhong [present-day Suzhou]” (lxxix).

Lu’s inability or unwillingness to provide specific information about the sources of his edition raised suspicion already among his contemporaries. Lin and Vankeerberghen mention that some scholars, such as Yan Yuanzhao 巖元照 (1773–1817) believed that the manuscript may have been acquired from Hui Dong 惠棟 (1697–1758), a famous scholar from the locality of today’s Suzhou. But even if this suggestion is correct, what were the origins of Hui Dong’s text?

A definitive solution to this old puzzle was offered in a 2019 study by Hou Jinman 侯金滿.Footnote ¹ Although Lin and Vankeerberghen acknowledge Hou’s article as a “careful comparison between Lu Jianzeng’s and Hui Dong’s editions” (xcviii, n. 252), this brief mention does not do justice to Hou’s brilliant work and its implications. Hou’s study is based on a meticulous, passage-by-passage analysis of not just two, but three texts: Lu Jianzeng’s published edition; an unpublished manuscript of the Shangshu dazhuan by Hui Dong preserved at the National Library of China; and another unpublished manuscript of what appears to be an earlier work-in-progress copy of Hui Dong’s reconstitution project of the Shangshu dazhuan, preserved at the National Library in Taiwan.Footnote ² Hou examines in detail the structure, contents, and textual features of these three versions, establishing their filiation and temporal sequence.

According to Hou, Hui Dong’s manuscript preserved at the National Library of China is nothing but the Wuzhong manuscript mentioned by Lu Jianzeng. The structure of these editions is largely the same: both consist of four juan of main contents with no indication of sources, and one juan of “addenda” (buyi 補遺), where passages are accompanied by mentions of sources that they have been extracted from. The contents of the two editions differ somewhat: the main part of Lu Jianzeng’s edition contains 114 passages, while the manuscript contains only 94.Footnote ³ However, 93 of the 94 passages found in the manuscript have been incorporated into Lu Jianzeng’s edition, suggesting that Lu’s edition is an extension of Hui’s manuscript.Footnote ⁴ As for the “addenda,” the manuscript contains 61 passages, of which Lu’s shares 57 and adds others for a total of 90 passages. The four omitted addenda passages found their way into the main part of Lu’s edition. At some places, Lu Jianzeng’s edition mistakenly incorporates some of Hui Dong’s notes, misinterpreting them as Zheng Xuan’s commentary. These shared errors confirm unequivocally that Lu Jianzeng’s edition is derived from the National Library of China manuscript.Footnote ⁵ While the National Library of China manuscript mentions Hui Dong as the author of the “addenda,” Lu Jianzeng’s edition omits this mention. Considering that Lu Jianzeng’s edition appeared later, this removal of the traces of Hui Dong’s authorship must have been performed deliberately.Footnote ⁶

The next step in Hou’s analysis is to establish the relationship between the manuscript preserved at the National Library of China and the one preserved in Taiwan. Hui Dong’s authorship of the Taiwan manuscript is confirmed by the presence of a seal imprint “Personally certified by Hui Dingyu” (惠定宇手定本; Dingyu is Hui Dong’s cognomen) on every volume.Footnote ⁷ While the manuscript preserved at the National Library of China looks like a clean copy, its Taiwanese counterpart contains many marginal notes and corrections, suggesting that it was Hui Dong’s working copy. It also contains a set of special marks (rectangles, triangles, and L-shaped hooks), which Hui Dong appears to have used to organize his collation of passages from different sources. The arrangement of material in the volumes of the Taiwan manuscript is not very orderly, and the chronological principle is not maintained consistently: again, understandable for a working copy. Although the Taiwan manuscript contains 256 passages overall, substantially more than the184 passages in the National Library of China manuscript, it is less comprehensive in contents, because many of the 256 passages are repetitive or fragmentary. In the National Library of China version, some of them have been removed while others have been merged together.Footnote ⁸ Once again, some of the mistakes present in the Taiwan manuscript found their way into the National Library of China manuscript and thence into Lu Jianzeng’s edition, showing that the Taiwan manuscript is ancestral to both the other texts.Footnote ⁹

Most importantly, all passages in the Taiwan manuscript—not merely in the final “addenda” volume as in the National Library of China manuscript—indicate their sources. Hou concludes that creation of the “clean copy” that we witness in the National Library of China manuscript involved a structural reorganization. The text was divided into two parts. The larger, first part was disguised as a “continuously preserved” text with all references to sources and all indications of textual variants removed. Only the smaller part, labelled “addenda” preserved the features of an academic text-reconstitution project. To put it more bluntly, Hui Dong deliberately misled his readers by presenting the results of his philological reconstitution as a continuously preserved text.

It is not clear why Hui Dong and Lu Jianzeng deceived their readers in such a way. Although the suggestion that Lu Jianzeng’s edition was in fact a reconstitution by Hui Dong has been expressed before, there is a difference between having doubts and knowing for sure.Footnote ¹⁰ It is deeply disturbing to realize that a scholar of Hui Dong’s caliber was capable of a fraud that confused many generations of scholars. The translation by Lin and Vankeerberghen has also inherited the problems of Hui Dong’s fraudulent edition.

Does this diminish the value of Lin and Vankeerberghen’s publication? To an extent, it does. The absence of source references will be a problem for those who want to use this volume in historical research. They may have to use it in conjunction with other editions, such as the one by Chen Shouqi 陳壽祺 (1771–1834). But does the unfortunate choice of Wang Kaiyun’s edition as the foundation of their translation negate the other advantages of Lin and Vankeerberghen’s work? It certainly does not, and the volume is sure to become the definitive reference for the Shangshu dazhuan. Its publication may propel the field towards greater maturity, with a more respectful consideration of traditional exegesis in the study of canonical texts. A pioneering effort comes with its own challenges. One hopes that Lin and Vankeerberghen’s example will inspire more such innovative studies of insufficiently appreciated but crucially important texts.