Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Scholars and their marginalia in the Qing dynasty (1644-1911) Wei, Yinzong 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2019_september_wei_yinzong.pdf [ 23.64MB ]
Metadata
JSON: 24-1.0379407.json
JSON-LD: 24-1.0379407-ld.json
RDF/XML (Pretty): 24-1.0379407-rdf.xml
RDF/JSON: 24-1.0379407-rdf.json
Turtle: 24-1.0379407-turtle.txt
N-Triples: 24-1.0379407-rdf-ntriples.txt
Original Record: 24-1.0379407-source.json
Full Text
24-1.0379407-fulltext.txt
Citation
24-1.0379407.ris

Full Text

 SCHOLARS AND THEIR MARGINALIA IN THE QING DYNASTY (1644-1911)  by Yinzong Wei M.A., Peking University, 2014  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Asian Studies)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  June 2019 © Yinzong Wei, 2019   ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  Scholars and Their Marginalia in the Qing Dynasty (1644-1911)  submitted by Yinzong Wei  in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Asian Studies  Examining Committee: Bruce Rusk, Asian Studies Supervisor  Leo Shin, Asian Studies, History Supervisory Committee Member  Siobhan McElduff, Classical, Near Eastern, and Religious Studies Supervisory Committee Member Erik Kwakkel, iSchool (Library, Archival and Information Studies)  University Examiner Ross King, Asian Studies University Examiner   Additional Supervisory Committee Members:  Supervisory Committee Member  Supervisory Committee Member     iii Abstract  Marginalia are a variety of writings and symbols drawn by readers on the pages of books. In Chinese history, marginalia were rare in both records and physical books before the late Ming (1368-1644). The early-Qing (1644-1911) calligrapher, bibliophile, and textual scholar He Zhuo (1661-1722) devoted himself to reading and collating books and composed marginalia on hundreds of titles. After his death, the composition and transcription of marginalia started to become a popular scholarly practice. The transcription of marginalia helped to build up a rather efficient model of transmitting information, knowledge, and thought among scholars. It formed a particular scholarly culture—a systematic way for scholars to think and behave. This study explores how this scholarly culture took form, gained momentum, and shaped scholarly styles and scholars’ lives, thoughts, mental states in the Qing dynasty. The main part of this study is made up of four chapters. Chapter 2 introduces the characteristics of marginalia in comparison with other Chinese interpretive texts. Chapter 3 is a case study of He Zhuo’s marginalia. Focusing on He’s marginalia on the Hou Hanshu (History of the Later Han Dynasty [25-220]), this chapter explores He’s reading habits and scholarly practices, and their influence on later scholars and readers. Chapter 4 is concerned with questions of who participated in the practice of transcribing marginalia, how different participants were involved in the process of transcription, and their motives and attitudes. Chapter 5 analyses different transcriptionists’ colophons, so as to explore their private lives and mental states. Qing scholars spent an enormous amount of time and energy composing and transcribing marginalia. They were concerned with both content and form of marginalia. In this process, scholars not only tried to accumulate knowledge, but also pursued its aesthetic values. They inherited reading habits and scholarly approaches from Ming scholars, and developed their own way of reading, doing research, and living.    iv Lay Summary  Marginalia are a variety of writings and symbols drawn by readers on the margins of books. Centring on marginalia, this study explores the reading practices and the scholarly culture of Qing Dynasty China (1644-1911). Beginning in the early Qing period, more and more scholars devoted themselves to reading and collating ancient texts. They gradually developed a habit of writing marginalia while reading, transcribing other readers’ marginalia, and editing marginalia in various ways, which formed a particular scholarly culture. This study explores how this scholarly culture took form, gained momentum, and shaped scholarly styles, scholars’ lives, thoughts, and mental states in the Qing dynasty.    v Preface  This dissertation is original, unpublished, independent work by the author, Yinzong Wei.     vi Contents      Abstract ………………………………………………………………………………………. iii     Lay Summary ………………………………………………………………………………… iv     Preface ………………………………………………………………………………………… v     Contents ………………………………………………………………………………………. vi     List of Tables…………………………………………………………………………………. vii     List of Figures ……………………………………………………………………………….. viii   Conventions ………………………………………………………………………………….... x     Chart of Historical Periods Discussed ………………………………………………………… xi 1. Introduction …………………………………………………………………………………… 1 2. Chinese Interpretive Texts: Annotation, Commentary, and Marginalia ……………………… 15 3. The Reading Seed He Zhuo and His Marginalia …………………………………………….. 89 4. Transcription: Transcriptionists and Relationships Built with Marginalia …………………  124 5. Transcription: Time, Space, and Scholarship ……….……………………………………… 176 6. Conclusion: Marginalia, Scholars, and the Intellectual History of the Qing ………………... 218     Bibliography ………………………………………………………………………………... 230     Appendix I  Books Containing He Zhuo’s Marginalia and Their Transcriptions ………….. 244    vii List of Tables  Table 3-1    Table of the sexagenary cycle……………………………………………………... 111 Table 4-1    Marginalia employed by Ruan Yuan’s edition of the Three Classics…………….. 143 Table 5-1    Gu Guangqi’s colophons in the Huayang guo zhi held at the National Library of  China …………………………………………………………………………….. 186       viii List of Figures  Figure 2-1     One half leaf of a Qing edition of the Xixiangji …………………………………. 46 Figure 2-2     Genealogy of the transmission of the Yijing of New Text School in the Western Han ………………………………………………………………………………. 61 Figure 2-3a    A manuscript copy of the Chunqiu guliangzhuan ………………………………. 64 Figure 2-3b    First leaf of juan 2 of a Southern Song edition of the Zhouyi zhushu …………… 65 Figure 2-4      First leaf of juan 1 of a Southern Song edition of the Shiji ……………………… 67 Figure 2-5      Two half leaves of a Ming edition of the Maoshi zhenya ……………………….. 70 Figure 2-6      One leaf of the Xin qie Sanzang chushen quanzhuan ……………………………. 72 Figure 2-7      First half leaf of a multicolor edition of the Yinfujing …………………………… 75 Figure 2-8      Two half leaves from a Ming edition of the Shijing …………………………….. 76 Figure 2-9a    One half leaf of the  Hanshu containing marginalia in four colors ……………… 79 Figure 2-9b    One half leaf of the  Li Yishan shiji containing marginalia in four colors ……….. 80 Figure 2-10    One half leaf of the Jixuan ji containing marginalia on the layout of the book … 83 Figure 2-11    One half leaf of the Qian shu ……………………………………………………. 85 Figure 4-1      Last half-leaf of the Shitong with Jiang Gao’s transcription of He Zhuo’s  marginalia ………………………………………………………………………. 130 Figure 4-2      Leaf 22b of juan 20 of the Shitong with He Zhuo’s marginalia transcribed by  an unknown transcriptionist …………………………………………………….. 131 Figure 4-3      Transcriptions of He Zhuo’s marginalia on the Zhouli zhushu …………………. 141 Figure 4-4      Leaf 1a of juan 1 of the San Tangren wenji with Zhang Yu’s marginalia …….. 150 Figure 4-5a    Leaf 1a of the Yulan shi in the Tangren xuan Tangshi with He Zhuo’s  marginalia ………………………………………………………………………. 168  ix Figure 4-5b    Leaf 65b of the Yulan shi in the Tangren xuan Tangshi with He Zhuo’s  marginalia and impression of seals ……………………………………………... 169 Figure 4-5c    Last half leaf of the Heyue yingling ji in the Tangren xuan Tangshi with He  Zhuo’s marginalia and impression of seals ……………………………………... 170 Figure 4-6      First half leaf of the Yulan shi in the Tangren xuan Tangshi containing He  Zhuo’s marginalia transcribed by an unknown scribe ………………………….. 171 Figure 4-7      Transcriptions of He Zhuo’s marginalia in a Ming copy of the Hou Hanshu …. 173 Figure 5-1a    First page and Table of Contents of the Wudai shiji with Gu Guangqi’s  colophon ………………………………………………………………………... 192 Figure 5-1b    Last leaf of the Wudai shiji with He Zhuo’s colophons transcribed by Yao Shiyu …… 193 Figure 5-2      Leaves 5b and 6a of juan 1 of the Li Yishan shiji with various scholars’ marginalia respectively transcribed in different colors ……………………………………... 200 Figure 5-3      Jiang Fengzao’s transcription of various scholars’ marginalia on the Dushu minqiuji … 204 Figure 6-1      First half leaf of the “Su Xueshi wenji jiao yu” ……………………………………. 225    x Conventions  Chinese names and terms are Romanized according to the pinyin system, except for names better known in another form such as that of Confucius and those preferred by authors writing in European languages. Translations are my own unless otherwise indicated; official titles follow Hucker, Dictionary of Official Titles. Zhushu , pingdian ! and pijiao  are respectively translated as annotation, commentary and marginalia. Chinese rare books are referred to by juan (lit. “scroll,” similar to “chapter”); page numbers of texts cited from rare books are given unless there is no clear page number in the book. Call numbers of rare books that were in Chinese characters will be changed to the initials of the characters. For example, xianshan  is replaced by XS; xianpu  by XP.      xi Chart of Historical Periods Discussed  BCE Western Zhou  1045?-771 Eastern Zhou  771-256 Spring and Autumn Period  771-479  Warring States Period  479-221 Qin Dynasty  221-207 Han Dynasty  206 BCE-220 CE Western Han  206 BCE—9 CE CE Eastern Han  23-220 Three Kingdoms  220-280 Jin Dynasty  265-420 Southern Dynasties  420-589 Northern Dynasties  386-581 Sui Dynasty  589-618 Tang Dynasty 	 618-907 Five Dynasties and Ten Kingdoms  907-960 Song Dynasty  960-1279 Northern Song  960-1127 Southern Song  1127-1279 Yuan Dynasty  1279-1368 Ming Dynasty  1368-1644 Qing Dynasty   1644-1911 Republic of China  1911-1949    1 1 Introduction  In the past hundred years, studies on the history of Chinese scholarship and intellectual history1 in the Qing U  dynasty (1644-1911) have experienced a considerable transformation in ideas, approaches and methodology. Broadly speaking, what was emphasized by academic leaders and dominant views shifted from political factors to philosophical aspects, later to social-economic and cultural dimensions. Some scholars, especially those who focused on the philosophical aspects, such as thoughts, values and beliefs, claimed that they were concerned with the “internal” dimension of scholarship, while studies concentrating on political, social, economic, religious and cultural factors were deemed to be the “external” dimension.2 In this regard, the study of Qing scholarship fell into two categories: the external approach and the internal approach. The tension between the internalists and externalists was mainly created by the former. Of course, most scholars did not conduct homogenous research focusing on only one dimension. As with all other categories, this dichotomy is not a simple one. Different scholars usually employ different approaches and pay distinct attention to each dimension. And different approaches usually resulted in different evaluations of Qing scholars’ achievements. Zhang BinglinpX¡ (1868-1936), Liu Shipei4) (1884-1919), Liang Qichao N%‘ (1873-1929), and many nationalists in late nineteenth- and early twentieth-century China stressed                                                         1 Intellectual history in this study refers to the history of scholars (including scholars’ social network, sponsorship, and such), scholarship (including theory, knowledge, and scholarly information) and scholar’s thoughts (including such things as their beliefs, values, and ideas). 2 See, for example, Yu Yingshi I (Ying-shih Yü), “Some Preliminary Observations on the Rise of Ch’ing Confucian Intellectualism,” Ch’ing-hua hsueh-pao 11 (1975): 105-144.  2 political factors in the development of Chinese scholarship. 3 For example, in Liang Qichao’s comprehensive study of the scholarship of the Qing dynasty, Zhongguo jin sanbainian xueshushi 	(“b5/…$ (A history of Chinese scholarship in the past 300 years), he made a generally fair assessment of Qing scholars mainly according to their scholarly achievements.4 But when talking about the characteristics of Qing scholarship, he claimed that “The point of departure of Ch’ing (Qing) learning was a violent reaction against the Neo-Confucianism of the Song [1, 960-1279] and Ming [G, 1368-1644].”5 He stressed the discontinuity between Qing learning and earlier Song-Ming learning. This discontinuity, in Liang’s view, resulted from political pressure occasioned by Manchu rule. He argued, “When literary inquisitions took place all too frequently, scholars became increasingly concerned with self-preservation and dared not expound any doctrine that might arouse official suspicious. But the talents and intellect of brilliant and outstanding men could not remain unused forever; exegesis of ancient aphorisms and exhaustive searching into the semantics of technical terms could certainly be called [tasks which] ‘do not injure the world and do not conflict with men,’ and in these scholars found a refuge.” 6  Benjamin Schwartz has                                                         3 See Zhang Binglin, “Qing ru” U, in Qiu shu xiang zhu ˆJ‹S, annotated by Xu Fuguan :=† (Shanghai: Shanghai guji chubanshe, 2000), 132-175; Liu Shipei, “Qing ru deshi lun” U<-Œ, in Liu Shipei shixue lunzhu xuanji 4)$/Œ–Ÿ (Shanghai: Shanghai guji chubanshe, 2006),417-427; Liang Qichao N%‘ (Liang Ch’i-ch’ao), Intellectual Trends in the Ch’ing Period, trans. Immanuel C. Y. Hsü (Cambridge: Harvard University Press, 1959). 4 See Liang Qichcao, Zhongguo jin sanbainian xueshu shi 	(“b5/…$, Zhu Weizheng Luš, ed. (Shanghai: Fudan daxue chubanshe, 1985). 5 Liang, Intellectual Trends in the Ch’ing Period, 27. 6 Liang, Intellectual Trends in the Ch’ing Period, 47.  3 commented that Liang “stressed the peculiar repressiveness of the Manchu dynasty which forced the literati away from ‘practical statesmanship’ into the innocuous pursuit of pure scholarship.”7  There are two points needing our attention: 1) Liang focused on the Evidential Research (kaojuxue yC/) of the Qing dynasty (including disciplines such as philology, etymology, paleography, geography, epigraphy, astronomy, phonetics, and phonology) and largely neglected the Cheng-Zhu Learning of Principle (Cheng Zhu lixue lL^/) and Lu-Wang Learning of Mind and Heart (Lu Wang xinxue \>/); he concentrated on “pure scholarship” — scholarly methodology, knowledge — and paid less attention to philosophical dimensions. In this regard, he concluded that the Qing scholars contributed less in philosophical fields, and this was caused by an oppressive climate that prevailed under Manchu rule. 2) Liang stressed the determinative impact of political factors.8  This politics-oriented approach is rather influential. 9 However, it tends to oversimplify the situation, so that it soon drew a great deal of criticism. Qian Mu›m (1895-1990), Feng Youlan !ƒ (1895-1990), as well as many other scholars took a rather distinctive approach that they focused on the so-called “internal aspects” of Chinese scholarship. This approach was followed by Yu Yingshi’s studies on the “internal development of Neo-Confucianism” 10 and the “inner logic” of intellectual trends,11 i.e., “focusing on the inner                                                         7 Benjamin Schwartz, “Foreword” to Intellectual Trends in the Ch’ing Period, xviii. 8 See Liang, Intellectual Trends in the Ch’ing Period and Zhongguo jin sanbainian xueshu shi. 9  See, for example, Sun Qinshan .P& , Zhongguo guwenxianxue shi 	(#E[/$  (Beijing: Zhonghua shuju, 1994), 831-835. 10 Yu Yingshi, “Some Preliminary Observations on the Rise of Ch’ing Confucian Intellectualism,” 105. 11 Yu Yingshi, “Some Preliminary Observations on the Rise of Ch’ing Confucian Intellectualism,” 112, 128; “Qingdai sixiangshi de yige xin jieshi” U@AcF‡—, in Yu Yingshi, Zhongguo sixiang  4 evolution of intellectual history, ignoring political, economic and other external factors.”12 Yu came up with a new theoretical framework focused on the conflict between intellectualism and anti-intellectualism to analyze the development of Chinese thought in the late imperial period. One serious problem of this approach is that it has the tendency of detaching thought from its contexts, which might oversimplify the course that thought took and, of course, deviated from the historical facts. 13 Recent studies by Huang Jinxing (Chin-shing) ¢”~, Benjamin Elman, Hamaguchi Fojio, Lin Qingzhang, Edward Wang, and others have shown that Qing scholars, Qian-Jia Evidential Researchers and New Text scholars in particular, raised new questions and managed to supply various new answers. Evidential Research also had its philosophical dimensions and concrete ideas about how to cope with new practical affairs. These studies all showed the necessity of taking into account external factors in the study of the history of scholarship and thought. 14 Examples include,                                                         chuantong de xiandai quanshi 	(@Atc]Š—  (Nanjing: Jiangsu renmin chubanshe, 1992), 199. 12 Yu Yingshi, “Qingdai sixiangshi de yige xin jieshi,” 199. On Qian Mu, Feng Youlan and Yu Yingshi’s study, see also Qian Mu, Zhongguo jin sanbainian xueshu shi (Shanghai: Shangwu Yinshuguan, 1937); Feng Youlan, A Short History of Chinese Philosophy, Derk Bodde, ed. (New York: Macmillan, 1960); Yu Yingshi, Zhongguo sixiang chuantong de xiandai quanshi. 13 Yu Yingshi’s approach of studying intellectual history changed several times. But his idea of “inner logic” was one of the most influential explanations for the Ming-Qing intellectual transition. Therefore, this study focuses on this approach. 14 See Huang Jinxing, Philosophy, Philology, and Politics in Eighteenth- Century China: Li Fu and the Lu-Wang School under the Ch’ing (Cambridge: Cambridge University Press, 1995); Benjamin Elman, Classicism, Politics and Kinship: The Ch’ang-chou school of New Text Confucianism in late imperial China (Berkeley: University of California Press, 1990); Lin Qingzhang MB8 and Zhang Shouan 7,0, ed., Qianjia xuezhe de yili xue '/zcx^/, 2 vols. (Taipei: Zhongguo wenzhe yanjiu suo, 2003); Hamaguchi Fujio W"2+ž, Shindai kokyōgaku no shisōshiteki kenkyū UyC/@A$cgn  5 Elman’s study of the socio-economic and cultural dimensions of Evidential Research and the Changzhou New Text School in the middle and late Qing, 15 Thomas Wilson and Huang Jinxing’s studies of Confucian practice,16 Lawrence Kessler, Frederic Wakeman and Kent Guy’s studies of the relationship between scholars and the Manchu state, all of which demonstrated scholars’ attention to the external dimensions of scholarship and thought.17 The study of intellectual history                                                         (Tokyo: Kokusho kankōkai, 1994); Edward Wang, “Beyond East and West: Antiquarianism, Evidential Learning, and Global Trends in Historical Study,” Journal of World History 19.4 (2008): 489-519; Ge Zhaoguang ‚, Zhongguo sixiangshi, di’er juan: qi shiji zhi shijiu shiji zhongguo de zhishi, sixiang yu xinyang 	(@A$q s|s	(cfŽ@A}, second edition (Shanghai: Fudan daxue chubanshe, 2013), 365-392. 15 See Benjamin Elman, From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China, revised edition (Los Angeles: University of California, 2001); Elman, Classicism, Politics and Kinship: The Ch’ang-chou School of New Text Confucianism in Late Imperial China;  16 Thomas Wilson, “The Ritual Formation of Confucian Orthodoxy and the Descendants of the Sage,” Journal of Asian Studies 55.3 (August 1996): 559-84. Thomas Wilson, ed., On Sacred Grounds: Culture, Society, Politics, and the Formation of the Cult of Confucius in Imperial China (Cambridge: Institute for East Asian Studies, Harvard University, 2002); Huang Jinxing, Shengxian yu shengtu {}{; (Beijing: Beijing daxue chubanshe, 2005); Zhang Shou’an 7,0, Yi li dai li — Ling Tingkan yu Qing zhongye ruxue sixiang zhi zhuanbian j^——6*}U	€/@A’ (Shijiazhuang: Hebei jiaoyu chubanshe, 2001); Zhang Shou’an, Shiba shiji lixue kaozheng de sixiang huoli — Lijiao lunzheng yu lizhi chongxing sj/yc@AT——jDŒZ}jk˜d (Taipei: Zhongyang yanjiuyuan jindaishi yanjiusuo, 2001). 17 See Lawrence Kessler, “Chinese Scholars and the Early Manchu State,” Harvard Journal of Asiatic Studies 31 (1971): 179–200; Frederic Wakeman, “The Price of Autonomy: Intellectuals in Ming and Ch’ing Politics,” Daedalus 101, no. 2 (1972): 35–70; Kent Guy, The Emperor’s Four Treasures: Scholars and the State in the Late Ch’ien-lung Era (Cambridge: Harvard Council on East Asian Studies, 1987); Yang Nianqun O?w, Hechu shi Jiangnan?: Qing chao zhengtongguan de queli yu shilin jingshen shijie de bianyi „H“R”UKQt†cho}+Mri_c` (Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2010).  6 can hardly be thoroughly comprehended without paying attention to real people’s lives, thoughts, emotions, beliefs, and practices. Factors such as political environment, social structure, economic condition, and cultural ecology could all contribute to the understanding of scholars’ living conditions and offer new explanations of the evolution of thought and scholarship in the Qing dynasty. In this sense, previous research have significantly contributed to the study of Qing intellectual history. However, most of these studies have neglected a very important dimension of intellectual history: scholarly practices, i.e., how scholars collected books and documents, how they transmitted scholarly materials to each other, how they communicated with each other, how they read and made sense of various texts, whether they used similar methods and approaches to reading, and how their reading interacted with their scholarship, thought and life. It is essential to consider scholarly practices, especially reading practices, are essential to the study of intellectual history. Yet previous approaches paid too much attention to scholars’ published works. Based on some quotations and elements of thought detected in a scholar’s works, his or her reading diet was reconstructed in a cursory way, and the relationships between him and his predecessors were hastily defined. However, works can unfold only a part of one’s thought and, normally, only a small part of what one has read. There is actually a huge gap between what one has read and what one records, between what one thinks and what one publishes. To fill in the gaps and draw a more complete picture of a person’s thought and life, one effective method is to study his or her reading practices. To do so, it is necessary to take seriously sources created in the course of reading, between what one reads and what one composes. Marginalia, a variety of writings and symbols draw by readers on the pages of books, are sources of this kind. This study will take marginalia as its object, treat scholarly practices with marginalia as a cultural phenomenon, and explore how marginalia culture took form, gained momentum, and shaped scholarly styles and scholars’ lives  7 and thinking in the Qing dynasty. I attempt to answer questions of how Qing scholars treated the scholarly legacy left by Ming scholars, how marginalia culture contributed to the rise and development of evidential research, how evidential research expanded and penetrated into people’s lives, and so forth. Meanwhile, I also expect to clarify how different generations of scholars in the Qing dynasty perceived their positions and roles in the world in new circumstances.   Methodology This study will focus on the cultural aspects of marginalia for the following reasons: 1) the composition, transcription, editing and publishing of marginalia involved a variety of human practices; 2) these practices, as well as the content of marginalia, reveal people’s thoughts, beliefs, ideas, emotions and other intangible factors; 3) these practices and their products influenced the thoughts and practices of later generations, permeating, in some way, into modern China. The methodology of this study is mainly derived from recent literature on the “new” history of books, an interdisciplinary hybrid of intellectual history, cultural history, new bibliographical studies, and social history. “The purpose” of the “new” history of books, Robert Darnton says, “is to understand how ideas were transmitted through print and how exposure to the printed word affected the thought and behavior of mankind.” 18  Darnton emphasizes the adjective “new” because he distinguishes this new discipline which “can be called the social and cultural history of                                                         18 Robert Darnton, “What is the History of Books,” in The Kiss of Lamourette: Reflections in Cultural History (New York: Norton, 1990), 107. Here Darnton focuses on printed artifacts. In this paper, I use “book” very broadly to include any written or printed text, such as manuscript written on parchment, bamboo, silk, stone, paper, and all kinds of imprints. “Text” in this study is a defining term, denoting an agglomeration of the abstract verbal signs and symbols. Books are the physical embodiments of texts.  8 communication by print” 19 from traditional “bibliography,” which was regarded primarily as “the servant of scholarly editing and textual criticism which, in turn, was perceived to be the servant of literary studies.” 20  Traditional bibliography concentrated on the book itself, especially on the text, the verbal information carried by the book, and tried to seek the so-called “authentic original text” that might be the closest to the “author’s intention.” 21 Put simply, it was a study of what kinds of books or texts we have; or precisely, what books human beings have. Under the influence of social and cultural history, as well as the new history of books, bibliography entered into a new stage, as demonstrated by D. F. McKenzie’s new definition of bibliography in his influential book, Bibliography and the Sociology of Texts: “Bibliography is the discipline that studies texts as recorded forms, and the process of their transmission, including their production and reception,” that “studie[s] the composition, formal design, and transmission of texts by writers, printers, and publishers; their distribution through different communities by wholesalers, retailers, and teachers; their collection and classification by librarians; their meaning for ... and their creative regeneration                                                         19 Darnton, “What is the History of Books,” 107. 20 S.J. Michael F. Suarez, “Book History from Descriptive Bibliographies,” in The Cambridge Companion to the History of the Book, ed. Leslie Howsam (Cambridge; New York: Cambridge University Press, 2015), 121.  21 See Jerome J. McGann, A Critique of Modern Textual Criticism (University of Chicago Press, 1983). To know more about “author’s intention,” or “authorship,” see Roland Barthes “The Death of the Author”, in The Result of Language, trans. Richard Howard (New York: Hill and Wang, 1986); Michel Foucault, “What is an Author?” translated by Josue V. Harari, in Aesthetics, Method, and Epistemology, ed. James D. Faubion (New York: The New Press, 1998); Roger Chartier, “Figure of the Author,” in The Order of Books: Readers, Authors, and Libraries in Europe between the Fourteenth and Eighteenth Century, trans. Lydia G Cochrane (Stanford: Stanford University Press, 1994); Jack Stillinger, “A Practical Theory of Versions,” in Coleridge and Textual Instability: The Multiple Versions of the Major Poems (New York: Oxford University Press, Inc., 1994), 169-185.  9 by, readers.” It is a discipline that can be defined as a “sociology of texts,” in which “texts” encompass verbal, visual, oral, and numeric data, in the form of maps, prints and music, of archives of recorded sound, of videos, and any computer-stored information.22 For social and cultural historians, the history of books as a discipline has been established on two simple premises: 1) “books make history,” i.e., “books are the primary tools that people use to transmit ideas, record memories, create narratives, exercise power, and distribute wealth;” and 2) “books are made by history: that is, they are shaped by economic, political, social, and cultural forces.” 23 In this domain, the history of reading occupies  a position of considerable importance. If we center on reading (not books, texts or readers but reading itself), our study can be divided into three stages according to the process of reading: 1) before reading, 2) the act of reading, and 3) after reading. There is no clear division between any two of them, but this division can make the complicated situation clearer and simplify the analysis and explanation. The study of the first stage concentrates on how books get into a reader’s hands, i.e., when, where, and in what ways a book reaches a reader. This inquiry is concerned with the production, circulation and collection of books. Focusing on the agents involved in the whole process, Darnton comes up with a model, the “communication circuit,” which includes authors, publishers, printers, shippers, booksellers, binders and readers. 24 In “First Step Toward a History of Reading,” Darnton suggests that we study the external as well as the internal history of reading. For external aspects, there are two main types of study: microanalysis (including quantitative social history, making                                                         22 D. F. McKenzie, Bibliography and the Sociology of Texts (Cambridge; New York: Cambridge University Press, 1999), 12-15. 23 Simon Eliot and Jonathan Rose, “Introduction” to A Companion to the History of the Book (Malden, MA: Blackwell Pub., 2007), 1. 24 Darnton, “What is the History of Books,” 107-135.  10 comparisons between countries, cross-statistical comparisons) and microanalysis (including study of library catalogues, notarial records, subscription lists, records of lending libraries).25 “Having studied it as a social phenomenon, historians of reading can answer many of the ‘who,’ ‘what,’ ‘where,’ and ‘when’ questions.” 26  However, these studies can only answer how texts were transmitted, but cannot answer how ideas were transmitted and received, let alone how they affected the thoughts and behavior of readers, or, as Darnton put, “how exposure to the printed word affected the thought and behaviour of mankind.” 27 Therefore, we also need to study the second stage, the act of reading. Darnton suggests that we take into account “how people actually read” and also pay attention to reading clubs, family reading practices, and other reading practices. 28 Roger Chartier has made an in-depth study of such practices. In “Texts, Printing, Reading,” he poses a big question about reading: “How can a text that is the same for everyone who reads it become an ‘instrument of discord and battle between its readers, creating divergences between them, with each reader having an opinion depending on his own taste?’” 29  Given this diversity of readers and multiplicity of their abilities and expectations, Chartier proposes to approaches to a history of reading: “reconstructing the diversity of older readings from their sparse and multiple traces, and recognizing the strategies by which authors and publishers tried to impose an orthodoxy or a prescribed reading on the text.”30 He suggests that two perspectives that are often pursued separately be conjoined in such a history: 1)                                                          25 Darnton, “First Step Toward a History of Reading,” in The Kiss of Lamourette, 157-167. 26 Darnton, “First Step Toward a History of Reading,” 157. 27 Darnton, “What is the History of Books,” 107. 28 Darnton, The Kiss of Lamourette, 168. 29 Roger Chartier, “Texts, Printing, Reading,” in The New Cultural History, ed. Lynn Hunt (Berkeley: University of California Press, 1989), 154. 30 Chartier, “Texts, Printing, Reading,” 155-8.  11 Study of the text itself (“the study of the way in which texts and the printed works that convey them organize the prescribed reading,” as in the history of genres and in the bibliographical studies); and 2) Study of interpretive communities (“the collection of actual readings tracked down in individual confessions or reconstructed on the level of communities of readers — those ‘interpretive communities’ whose members share the same reading styles and the same strategies of interpretation”). Different from the “communication circuit” proposed by Darnton that centres on books, “interpretive communities” is a model for the study of reading individuals and groups, obviously centered on reading practices: We need to develop indicators of the major divisions that can organize a history of reading practices (of the use of texts, even the uses of the same text) — for example between reading out loud, for oneself or for others, and reading in silence; between reading inwardly and privately and reading publicly; between religious reading and lay reading; and between “intensive” reading and “extensive” reading. Beyond these macroscopic cleavages, the historian must seek to determine the dominant paradigms of reading in a community of readers in a given time and place. ... The mode of reading, which is dictated by the book itself or by its interpreters, provides the archetype of all reading, whatever kind it may be. 31 Similarly, Jonathon Ross also advocates: Broadly, then, the history of reading is the history of interpretation ... of books, magazines, newspapers, advertising bills, films, radio programs, musical performances, school lessons, and adult education classes. The rationale behind this method is simply this: we can only                                                         31 Chartier, “Texts, Printing, Reading,” 158, 166.  12 understand the mentality of a given audience by reconstructing (as far as possible) its cultural diet, and then asking how that audience interpreted those cultural experiences. 32 Through study of the production, circulation and collection of books, as well as the reading practices of “interpretative communities,” it is not difficult to comprehend audiences’ cultural diet, but it is difficult to know how they interpreted those cultural experiences, or to understand the emotions, thoughts, beliefs and values underlying their reading practices. Hence, it is not possible to fully understand how books and reading affected the thought and behaviour of mankind. Reading is more than getting the general main points of the text; reading will alter the mental condition of a reader; the words, illustrations, the format of the book, and even the wormholes in the paper can be triggers of new thoughts; reading can change the intellectual life of readers as well as their political and daily lives. It is not enough just to find out what the reader read and how he or she read it. We should go on to examine the effectiveness and consequences of reading, to ask not what reading could do, but what reading has done to mankind. Therefore, we have to study the third stage of reading, a stage after the act of reading that is concerned with readers’ responses — reading reception and its effects on people’s intellectual and daily lives. “It should be possible,” Darnton says, “to develop a history as well as a theory of reader response. Possible, but not easy; for the documents rarely show readers at work, fashioning meaning from texts, and the documents are texts themselves, which also require interpretation.”33 The main problem of the study of this stage is sources. The most important sources we can use to recover the historical reader include: (1) Autobiographical documents; (2) Commonplace books;                                                         32 Jonathan Ross, “Arriving at a History of Reading,” Historically Speaking 5 (2004): 39. 33 Darnton, “First Step Toward a History of Reading,” 157.  13 (3) “Albums” or miscellanies of texts; and (4) Marginalia.34 Among these, marginalia are of the utmost importance, because they are composed during the act of reading and thus reveal the process of reading and the mind states of the reader while engaged in reading. Heather Jackson, William Sherman, and Anthony Grafton have done a great deal of pioneering work on the study of marginalia.35 Their contributions and the features of marginalia they have covered will be discussed in detail in chapter two.  Structure of the Dissertation The main part of this study is made up of four chapters, focusing on the composition, transcription of marginalia, observing the transformation of scholarly styles, scholars’ thoughts, emotions, beliefs, and private lives within these practices, and their influence on later generations.  Chapter two introduces marginalia, along with two related forms of textual response, zhushu Sa (annotations) and pingdian ‰£ (commentaries), of which the former usually focused on the meaning of the main text, while the latter addressed the literary features. Because all of the three are texts that provide interpretation, explanation, appreciation, and evaluation of a pre-existing “main text,” they can be defined as “interpretive texts.” I will first discuss the contents and characteristics of these interpretive texts, and then examine their physical features, such as their                                                         34 Stephen Colclough, “Readers: Books and Biography,” in A Companion to the History of the Book, eds. Simon Eliot and Jonathan Rose (Malden, MA: Blackwell Pub., 2007), 53-57; See also Darnton, The Kiss of Lamourette, 154-187. 35 See Anthony Grafton, “Is the History of Reading a Marginal Enterprise?” Papers of the Biblographic Society of America 91(1997): 139-157; Heather Jackson, Marginalia: Readers Writing in Books (New Haven: Yale University Press, 2001); William Howard Sherman, Used Books: Marking Readers in Renaissance England (Philadelphia: University of Pennsylvania Press, 2008).  14 visual appearance, where and how they were contained, and how they circulated in history. Through this comparison, I expect to illustrate the essential features and various important characteristics of marginalia. I begin, in chapter three, with a case study of the marginalia of He Zhuo Y (courtesy name Qizhan3e, alternate name Yimen xœ, 1661-1722), an early Qing scholar and calligrapher. He, who worked as an imperial compiler (zuanxiu v) in the imperial court, composed marginalia on hundreds of titles, which were highly influential throughout the Qing dynasty. I focus on He Zhuo’s marginalia on the Hou Hanshu 9VJ  (History of the Later Han Dynasty), closely analyzing their characteristics and their influence on later scholars and readers.  In subsequent chapters I trace the transcription of marginalia and examine how the marginalia culture took into form and gained momentum in the Qing dynasty. Also centering on He Zhuo’s marginalia, chapter four is concerned with questions of who participated in this practice, how different participants were involved in the process of transcription, and what their motives and attitudes within this practice were; chapter five analyzes colophons composed by various transcriptionists (guolu zhe •™z), so as to explore their scholarly lives, beliefs and emotions.    15 2 Chinese Interpretive Texts: Annotation, Commentary and Marginalia  In the late imperial period, the Chinese empire was actually a reading empire. The administration of the state was built upon a complete textual system: the state orthodoxy was derived from the Confucian Classics; the dissemination of information was facilitated via a network of memorials and other official documents; communication among scholar-officials relied on letters transmitted by their family servants even when they were living in the same street; for the illiterate, imperial announcements were publicly read by the educated in the city and by the local gentry in rural areas. Before one could, with any luck, pass the civil service examinations and become an official, he had to be well-read in the Four Books (sishu »Ə), be familiar with the Five Classics (wujing ;Ⱦ), and at least conversant with several histories (shi ¢), philosophical works (zi ç) and works of belles-lettres (ji ˸).1 Moreover, much of the spare time of scholar-officials and some merchants was occupied with reading of various kinds, ranging from classical texts to popular works (such as fiction, drama, short stories, and other miscellaneous works).2 All of these texts—from official                                                         1 In pre-modern China, books classified into the jibu ˸ˌ were mainly verse and pross and criticisms of them, and only verses and prose were deemed belles-letters. About civil service examination, bureaucracy, and elite culture, see Peter Bol, “The Sung Examination System and the Shih,” Asia Major, 3rd ser., 3, 2 (1990):149-71; Benjamin Elman and Alexander Woodside, eds., Education and Society in Late Imperial China (Berkeley: University of California Press, 1994); Benjamin Elman, A Cultural History of Civil Examination in Late Imperial China (Berkeley: University of California Press, 2000); Civil Examinations and Meritocracy in Late Imperial China (Cambridge, Massachusetts; London, England: Harvard University Press, 2013). 2 About the reading public and reading practices of late imperial China, see Martin W. Huang, “Author(ity) and Reader in Traditional Chinese Xiaoshuo Commentary,” Chinese Literature: Essays, Articles, Reviews  16 documents to private letters, from memorials addressed to the throne to imperial documents directed at subjects, from classical works to popular works—could contain allusions to ancient texts. The reading, understanding and reception of ancient texts, especially the Five Classics, was a crucial task both for elite subjects and the emperor, not only in their public activities but also in their private lives and mental states. Thus, the first key question for us concerns how texts were interpreted. The study of the interpretation of ancient Chinese texts has long centered either on a particular scholar or on a particular text. Few took into account the features of the text per se that were differently interpreted or how these texts circulated in history, which both determined the reception and historical significance of ancient texts and their various interpretations over time. For example, zhushu Ǒȍ ,  pingdian ʐ̣ and pijiao ŝƢ  are different in content and nature but are indiscriminately translated into “commentary” or “annotation” in English,3 and this has confused                                                         (CLEAR) 16 (1994): 41-67; Anne E. McLaren, “Ming Audience and Vernacular Hermeneutics: The Uses of The Romance of the Three Kingdoms,” T‘oung Pao 81:103 (1995): 51-80; Anne E. McLaren, “Constructing New Reading Publics in Late Ming China,” in Printing and Book Culture in Late Imperial China, Cynthia J. Brokaw and Kai-wing Chow, eds. (Berkeley, Los Angeles and London: University of California Press, 2005), 152-183; Robert E. Hegel, “Niche Marketing for Late Imperial Fiction,” in Printing and Book Culture in Late Imperial China, 235-66; Cynthia J. Brokaw, “Sibao’s Customers and Popular Textual Cultural in the Qing,” in Commerce in Culture: the Sibao Book Trade in the Qing and Republican Periods (Cambridge: Harvard University Press, 2007), 513-533; Yuming He, Home and the World: Editing the “Glorious Ming” in Woodblock-Printed Books of the Sixteenth and Seventeenth Centuries (Cambridge: Harvard University Asia Center, 2013). 3  For instance, Daniel K. Gardner and Kai-wing Chow use “Confucian commentary” to denote interpretative text on the Confucian Classics. Martin W. Huang and David Rolston also use “commentary” to stand for comments and evaluations on Chinese fiction. See Daniel K. Gardner, “Confucian Commentary and Chinese Intellectual History,” Journal of Asian Studies 57:2 (May 1998): 397-422; Kai-wing Chow, “Paratext: Commentaries, Ideology, and Politics,” in Publishing, Culture and Power in Early Modern  17 our understanding of the features and functions of different kinds of interpretations of ancient Chinese texts. Therefore, I use the term “interpretive text” to denote all texts that provide interpretation, explanation, appreciation and evaluation of a pre-existing text, or the “main text.” In comparison with the main text, interpretive texts are secondary—they are generated by the main text, and for this reason, they always cling to the main text. Where there is no main text, there is no interpretive text. Interpretive text can be regarded as one kind of paratext, a concept that was proposed by Gérard Genette. In Genette’s theory, paratext consists of elements such as titles, subtitles, intertitles, prefaces, notes, epigraphs, illustrations, book covers, reviews, private letters, and many other kind of secondary signals that are related to the main text. The paratext performs the function of guiding or even controlling the reading and interpretation of the main text. It can decide how a text should be read.4 Genette stressed the transactional nature of paratexts between texts and readers, as he once put it: “It [paratext] not only marks the zone of transition between text and non-text, but also a transaction.”5 However, in comparison with other paratextual elements (such as illustrations, book covers, reviews, author’s letters), annotation, commentary and marginalia are usually physically closer to the main text, and they are mainly concerned with providing interpretation, explanation, appreciation and evaluation of a main text. Their nature is fundamentally “interpretive.” Therefore, this chapter will focus on these three interpretive texts,                                                         China (Stanford: Stanford University Press, 2004), 149-188; Martin W. Huang, “Author(ity) and Reader in Traditional Chinese Xiaoshuo Commentary,” 41-67; David. Rolston, Traditional Chinese Fiction and Fiction Commentary: Reading and Writing between the Lines (Stanford: Stanford University Press, 1997). 4 On paratext, see Gérard Genette, Paratexts: Thresholds of Interpretation, trans. Jane E. Lewin (Cambridge: Cambridge University Press, 1997). 5 Gérard Genette, “The Proustian Paratext,” in SubStance: a Review of Theory and Literary Criticism 17.2 (1988): 63.  18 discussing their contents, features, forms, and how they circulated in imperial China. By so doing, we can gain a deeper  understanding of the nature of marginalia.  Contents and Features It is because of their different modes of interpretation that I draw a distinction between zhushu, pingdian, and pijiao, and translate them respectively into “annotations,” “commentaries,” and “marginalia.” Annotations primarily aimed to draw out the philosophical and political meanings of texts based on glosses of meaning and sound. Commentaries in China concentrated on the literary features of the main text and aimed to help readers appreciate the beauty of the main text and improve the reader’s skill at composing classical poetry and prose. Marginalia are readers’ hand-written reading responses, which could include anything that readers come up with while reading. Zhushu/Annotations:  Proposing Meanings from the Classics The earliest interpretive texts in China can be traced back to the Spring and Autumn (770-476 BC) and Warring States (475-221 BC) periods. In his impressive monograph Writing and Authority in Early China, Mark Edward Lewis proposes a triangular model of “master—disciple—text” for the intellectual world during this era. This model elaborates how the various schools were organized and expanded, and how the thought of a given school was disseminated. In this book, Lewis emphasizes the crucial role that the text played for scholars in the process of the engaging with state affairs and expanding schools of thoughts in early China. He states that most of the schools in that time had a so-called “textual tradition,” that most of the schools were “text based,” and that even the “appearance of the master as an author ... was a function of, or a step toward, his  19 disappearance as the fundamental textual authority.”6 Texts, however, cannot speak for themselves. They have to be interpreted. Interpretation, even when orally transmitted from the master to the disciple, was made up of exegeses, glossing, paraphrasing, and discussion of the textual variants. These interpretations, primarily aimed at drawing out the meaning of texts, scarcely commented on the aesthetic features of the main text. So, we might think of “annotations” as their equivalent in English.  Historically, Chinese annotations actually had various names that could reveal their features. The earliest annotations on Confucian Classics were called zhuan/chuan \ or ji ʉ; later also referred to using such terms as jie ʄ, gu Ųxun ʈ, zhangju ȫž, shuo ʖ, or shuoyi ʖɇ.7 According to the Shuowen jiezi ʖŶʄé (Explaining graphs and analyzing characters), the original meaning of zhuan/chuan is “to pass on.” The influential Shuowen specialist Duan Yucai ƿǴɻ (1735-1815) pointed out that the extended meaning of zhuan/chuan is “to extend the meaning” and that zhuan/chuan in both zhuanzhu \Ǒ (annotation) and liuchuan Ǔ\ (to spread)                                                         6 Mark Edward Lewis, Writing and Authority in Early China (Albany: State University of New York Press, 1999), 57-69. 7 It is recorded in the “Yiwen zhi” ɲŶĻ of the Hanshu ǚƏ that: for the Yi ƃ (Book of Changes), there was the Zhoushi zhuan ¬dž\, and the zhangju ȫž of Masters Shi ż, Meng ì and Liangqiu’s Ʀ*; for the Shu Ə (Book of documents), there was one work of zhuan , the zhangju of Masters Ouyang Ƹ˱, Greater Xiahou ÐT and Lesser Xiahou’s, the jiegu ʄŲ of Masters Greater Xiahou and Lesser Xiahou, and Master Ouyang’s shuoyi ʖɇ; for the Shi ʒ (Book of songs), there were Master Mao’s gu xun zhuanŲʈ\, and Master Han’s ̆ gu Ų, shuo ʖ, neizhuan f\ and waizhuan Ò\; for the Li ȣ (Book of rites), there were one work of ji ʉ and one work of Mingtang yinyang shuo ƁÈ˭˱ʖ; for the Chunqiu Ƅȥ (Spring and autumn annals), there were the zhuan of Masters Gongyang hɆ, Guliang ȨƦ, Zuo Č, Zou ˎ and Jia’s Ø, and several zhangju and zaji. See Hanshu (Beijing: Zhonghua shuju, 1962), 1701-1784.  20 uses this extended meaning. In other words, zhuan/chuan means to transmit the meaning of the Classics. The original meaning of ji is “to record” or “to write down,”9 similar to zhuan/chuan. Ji annotations were actually disciples’ records of their masters’ interpretations of the Classics. The well-known Chinese historian Zhang Xuecheng ȫîʕ (courtesy name Shizhai ø̥, 1783-1801) noted of its meaning: The three zhuan of the Spring and Autumn Annuals recorded what meanings the annotator had heard of and drew out meanings from the Classics, so they can also be called “ji.” The two ji annotations to the Rites transmitted their explanations and circulated attached to the Classics, so they can also be called “zhuan.” 4Ƅȥ5%ö/\¤ʉŚɒSȾʬɇ˺ʙ/ʉ¡33Ⱦ4ȣ58Ř/ʉ¤\lʖ˫ȾɎɷ˺ʙ/\¡33The origin meaning of jie is “to cut an ox horn into halves;” “to analyze” is a derived meaning of this word.11 This implies that jie annotations analyzed the Classics and explained their meanings.                                                         8 See Shuowen jiezi zhu ʖŶʄéǑ, composed by Xu Shen ʍŏ and annotated by Duan Yucai ƿǴɻ (Shanghai: Shanghai guji, 1981), juan 8A, 377. 9  See annotations to the Shangshu ÿƏ  and the Liji ȣʉ , in Shangshu zhushu ÿƏǑȍ  (photo-reproduction in the Shisanjing zhushu Œ%ȾǑȍ, Taipei: Yiwen, 2007), juan 5, 68; Liji zhushu ȣʉǑȍ (photo-reproduction in the Shisanjing zhushu Œ%ȾǑȍ, Taipei: Yiwen, 2007), juan 13, 262. 10 Zhang Xuecheng, Wenshi tongyi jiaozhu Ŷ¢ʺɇƢǑ, annotated and collated by Ye Ying ɪǼ (Beijing: Zhonghua shuju, 1985), juan 3, 248. 11 See the annotation of the Liji and Shiji ¢ʉ, in Liji zhushu, juan 50, 845; Shiji (Beijing: Zhonghua shuju, 1963), juan 9, 399.  21 Shuo and shuoyi mean to declare the meaning of the Classics. The pronunciation of xun ʈ was similar to that of shun ̉ (“to obey” or “to follow”), so according to Duan Yucai, xun means to explain the meaning of the Classics to other people and lead them to follow the right principle.12 Gu Ų is the ancient form of gu ʏ, which means “to explain the meaning of ancient words.”13 Its derived meaning is “to explain the main text.”14 Zhangju ȫž, according to Liu Zhao’s }ƅ annotation of the Hou Hanshu, means “to divide the main text into sections and analyze the meanings of the sentences.”15 However, no book in the “Yiwen zhi” ɲŶĻ (Treatise on literature) of the Hanshu ǚƏ (History of the Han Dynasty) bears a title describing it as zhu Ǒ, nor is there any annotation to the Confucian Classics called zhu in the entirety of the Hanshu. Zhu was first used to mean “to annotate” in the biography of Zheng Xuan ˏdz (courtesy name Kangcheng ğŕ, 127-200) in the Hou Hanshu. Still, in the Hou Hanshu,  Zhang Kai’s Ĩƫ  biography states that Zhang Kai had composed “zhu to the Shangshu” (OÿƏǑ), but there is no evidence that this was the title of Zhang’s book.16  Few annotated books recorded in the “Jingji zhi” ȾȵĻ  (Treatise on the Confucian Classics and other books) of the Suishu ˲ Ə (History of the Sui) had zhu in their titles.17                                                         12 Shuowen jiezi zhu, juan 3A, 91. 13 Shuowen jiezi zhu, juan 3A, 92. 14 See the annotation of the “Yiwen zhi” of Hanshu, juan 13, 1708; see also Maoshi zhushu 。ʒǑȍ (Taipei: Yiwen, 2007), 11. 15 Fan Ye ɥƋ, Hou Hanshu, annotated by Liu Zhao }ƅ (Beijing: Zhonghua shuju, 1973), juan 28A, 955. 16 Fan Ye, Hou Hanshu, juan 35, 1212; juan 36, 1243. 17 Annotated books that were entitled with “zhu” include Xiaojing mo zhu ëȾ̢Ǒ by Xu Zheng İŴ, Hanshu zhu ǚƏǑ by Lu Cheng ˰Ǜ, and Hanshu jizhu ǚƏ˸Ǒ by Jin Zhuo ƈお. See the “Jingji zhi” of the Suishu (Beijing: Zhonghua shuju, 1982), juan 32, 933; juan 33, 953.  22 Most of them were only identified as having zhu in the author’s notes of the “Jingji zhi.”18 According to the Shuowen jiezi, zhu means “to irrigate or to fill with water.” The Liyi shu ^ȣȍ (Sub-annotation of the Rites and Ceremonies) writes: “the meaning of zhu is ‘tapping meaning from the Classic,’ like water dripping off an object.” ǑɍǑɇŻȾ'ɣLj/Ǒǫ  Similarly, Duan Yucai said, “Zhu means to lead [water] to a proper position, and so explicating the Classic in order to illuminate its meaning is called ‘zhu’” Ǒ/:ɍĥ/ƒŚ˅3Ų˒ȾCƁlɇƍǑ .  In addition, according to the “Jingji zhi” of the Suishu, yinyin ̇˴ (or yin ̇) annotations and jijie ˸ʄ annotations gradually increased in number, and yishu ɇȍ (or shu ȍ) annotations flourished during the late Han to Sui period under the influence of Buddhism.20 Yinyin annotations focused on the phonetic aspects of ancient texts; jijie means “collected annotations.” Shu originally meant “to unblock;” its derived meaning is “to remove misunderstandings,” i.e, to explicate a text.21  After the Sui dynasty ˲  (581-617), zhu, shu, and zhushu became the most common appellations of annotations, although new names such as zhengyi ƹɇ (orthodox meaning), suoyin                                                         18 For example, the self-annotation to the Zhouyi ¬ƃ of a 7 juan edition in the “Jingji zhi” of the Suishu writes “annotations by Yao Gui äʁ,” to that of a 13 juan edition writes “annotations by Cui Jian Ĉʂ and Master Fu [,” to that of a 10 juan edition writes “annotations by Master Lu ģ.” The self-annotation to the Shiji of a 80 juan edition writes “annotations by Pei Yin ɼ̔.” The Shuijing LjȾ has Guo Pu’s ˍに annotations for a 3 juan edition and Li Daoyuan’s ːˁ` annotation for a 40 juan edition. See the “Jingji zhi” of the Suishu, juan 32, 910; juan 33, 953, 982, 984. 19 See Xushen, Shuowen jiezi, annotated by Duan Yucai, juan 11A, 555; Yili zhushu ^ȣǑȍ (photo-reproduction in the Shisanjing zhushu Œ%ȾǑȍ, Taipei: Yiwen, 2007), juan 1, 3. 20 See the “Jingji zhi” of the Suishu, juan 32-35, 903-1104. 21 See Xu Shen, Shuowen jiezi, juan 14B, 744.   23 ȼ˴ (searching for the hidden meaning), and xinyi Źɇ (new interpretation) were created by different generations of annotators. All these names demonstrate that the core feature of Chinese annotation was to draw out the meaning of ancient texts, and that annotations were oriented toward explicating meaning, or “yi jing qi yi” SȾʬɇ(to propose meanings from the Classics) in the words of Zhang Xuecheng.Daniel K. Gardner claims that, within the Confucian tradition, Classics have no fixed meaning and “there was no such thing as a timeless, normative reading of a Classic.” Through creating a complex relationship between the Classics, audiences, and the commentarial tradition, annotators shaped the understandings of the Classics and gave them new meanings and significance in particular contexts.22  Since annotation is meaning oriented, the explanation of words and sentences became its main content. In this sense, as Lu Zongda ˰ó˂ and Wang Ning ǵù put it, ancient annotations provide a great number of sources with which to probe the meanings of the words of ancient texts. Nevertheless, they also remind us that although the explanations in ancient annotations might be more accurate because they were produced close to the time when the main text was composed, they should not be treated as the only basis for our study, inasmuch as they were scarcely able to avoid biases that could generate a considerable number of incorrect explanations. Ancient annotators aimed to transmit the “correct meanings” imparted by their teachers and they claimed to be rigorously representing their teachers’ opinions.23 In other words, transmitting philosophical                                                         22 Daniel K. Gardner, “Confucian Commentary and Chinese Intellectual History,” Journal of Asian Studies 57:2 (May 1998): 397-422. See also John B. Henderson, Scripture, Canon and Commentary: A Comparison of Confucian and Western Exegesis (Princeton: Princeton University Press, 1991). 23 See Lu Zonda ˰ ó˂ and Wang Ning ǵù, Xungu yu xungu xue ʈʏɛʈʏî (Taiyuan: Shanxi jiaoyu chubanshe, 1994), 17.  24 and political meanings was the annotators’ ultimate end; explaining words and sentences was just the means to that end.24 The well-known late-Qing Chinese scholar Pi Xirui ȕ˚ǽ (courtesy name Lumen ̛ˡ, 1850-1908) argued that classical study traditions were handed down from teachers to disciples in the Western Han period, and within clans in the Eastern Han period.25 Whether this distinction between the transmission methods of the Western and Eastern Han is correct or not, Pi Xirui correctly realized that different schools’ interpretations of the same Classic were, most of the time, not the same. The combination of different interpretations and various political cliques generated different schools; conflict between these schools in turn magnified the difference among different interpretations. Annotations could supply an arena for different schools, because the fundamental nature of the annotation was to transmit the sages’ Dao ˁ (Way). The Confucian Classics were transmitted orally either by different schools or by different clans before and during the Eastern Han. The great classicist Zheng Xuan was considered to have learned from various schools, adopting the strengths of both the orthodox New Text School (jinwen pai ]Ŷǒ) and the Old Text School (guwen pai Ŷǒ) in the interpretation of the Confucian Classics so as to bridge the two centuries of rivalry between them.26 His effort was actually to construct a new interpretive system, i.e., to unify and interpret all the Confucian Classics according                                                         24 See also Wang Li ǵ~, “Zhongguo yuyanxue shi” +Àʔʅî¢, Zhongguo yuwen +ÀʔŶ 3(1993): 233. 25 Pi Xirui ȕ˚ǽ, Jingxue lishi Ⱦîƽ¢ (Hong Kong: Zhonghua shuju, 1961), 136. See also Hu Pu’anɔƱñ, Zhongguo xunguxue shi +Àʈʏî¢ (Shanghai: Sanlian shudian, 2014), 3. 26 Whether or not there were centuries of rivalry between the New Text and the Old Text in the Han is an area of debate. But from historical records and Han scholars’ annotations of Confucian Classics, we know that the New-Text classicists and the Old-Text classicists employed different interpretive strategy.   25 to the theory of li ȣ (ritual) as expounded in the Classic of Rites.27 As Zhang Xuecheng put it, the nature of annotations of the Confucian Classics before and during the Eastern Han was that they “basically all drew out meanings from the Classics and each composed their own books, which is different from the annotations of later times”ɮȔSȾʬōlø¤əǨƏɛį)ȱǑə(¦3 .28 As a matter of fact, various annotations after the Eastern Han also “proposed meanings from the Classics” by explaining the words and sentences. Qiao Xiuyan ¸ȤĊ pointed out that yishu ɇȍ annotations from the Wei ̘ dynasty (261-225) to the Tang ° dynasty (618-907) were not seeking for the truth, but rather comprehending the laws of nature and human affairs. In order to express their own understanding, annotators sometimes drew farfetched analogies and made strained interpretations.29 Scholars of the Song dynasty ò (960-1279) were not satisfied with Han and Tang annotations. They composed new annotations bearing new interpretations, printed them and disseminated them all over the empire. Eventually, these new interpretations supplanted Han and Tang annotations and became the orthodox textbooks of the civil service examinations. The conflict between the new and old annotations was actually conflict between different orthodoxies, so annotations cannot be seen as mere explanations of words and sentences.30 Scholars of the Qing                                                         27 An annotation to the “Zaji” ˻ʉ of the Liji ȣʉ states: “The study of the Rites is actually Zheng Xuan’s study” (ȣƆˏî). See Liji zhushu, juan 40, 713. See also Qiao Xiuyan ¸ȤĊ, Yishuxue cunwang shi lunɇȍîê<¢ʘ (Taipei: Wanjuanlou, 2013), 177; Hua Zhe ɨ¶, Li shi Zheng xue ȣƆˏî (Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2018). 28 Zhang Xuecheng, Wenshi tongyi jiaozhu, juan 3, 248. 29 See Qiao Xiuyan, Yishuxue cunwang shi lun, 171-176. 30  See Kai-wing Chow, “Paratext: Commentaries, Ideology, and Politics,” 149-188; Thomas Wilson, Genealogy of the Way: The Construction and Uses of the Confucian Tradition in Late Imperial China  26 dynasty also armed themselves with techniques of annotation to dispute with Song and Ming scholars. Their annotations and other scholarly practices cannot be considered pure evidential research on the words and sentences either, because Qing evidential scholars had their own philosophy.31 Take the Shijing ʒȾ (Book of songs, Book of poetry, or Poetry classic) as an example to illustrate how one text was interpreted in different annotations. The Shijing was a collection of just over 300 poems from the Zhou dynasty (1045 BCE - 256 BCE). From very early times, it had been one of the centerpieces of the Confucian tradition, and was referred to as Classic. There were various schools of readings of the Shijing during and before the Han dynasty, which vary not only in interpretation but also in the text of the poems. Unfortunately, most of them are lost, leaving just a small number of fragments quoted in other texts. However, the text and interpretation of the Mao school survived. Its text is called the Maoshi (Mao’s poems), and the interpretation of the text                                                         (Stanford: Stanford University Press, 1995), 47-59; Gu Yongxin ̎《Ź, Jingxue wenxian de yansheng he tongsuhua: yi jingu shidai de chuanke wei zhongxin ȾîŶDzȓɸȂ­ʺVˆ——CʶƇAȓ\yǨ+ĸ (Beijing: Beijing daxue chubanshe, 2014). 31 About the various interpretative tradition of different annotations of Confucian Classics, see also John Makeham, Transmitters and Creators: Chinese Commentators and Commentaries on the Analects (Cambridge: Harvard University Press, 2003); John B. Henderson, Scripture, Canon and Commentary: A Comparison of Confucian and Western Exegesis. About the philosophy underlying evidential research, see Lin Qingzhang Ɵőĭ and Zhang Shou’an ĨÏñ, ed., Qianjia xuezhe de yili xue 5ºîɍȓɇǺî, 2 vols. (Taipei: Zhongguo wenzhe yanjiu suo, 2003); Hamaguchi Fujiō ǝœ÷Î˶, Shindai kōkyogaku no shisōshiteki kenkyū ǗAɌŮî9ŁŌ¢ȓȞȩ (Tokyo: Kokusho kankōkai, 1994); Ge Zhaoguang ɬbd, Zhongguo sixiangshi, di’er juan: qi shiji zhi shijiu shiji Zhongguo de zhishi, sixiang yu xinyang +ÀŁŌ¢Ȯ8“ $)ȸɚŒ2)ȸ+ÀȓȝʜŁŌɛXD, second edition (Shanghai: Fudan daxue chubanshe, 2013), 365-392.  27 is made up of Mao’s annotations (Mao zhuan 。\) and a “Preface” (Shi xu ʒĚ) to each poem.32 In the order of the Maoshi, the 24th poem is “Ye you si jun” ˔ƒƾ̝ (In the wilds is a dead doe, Mao #23). This poem reads: ˔ƒƾ̝ ȑɦ‡/3 ƒÜœƄ ¥Îʓ/3  ƟƒƱƯ ˔ƒƾ̛3 ȑɦȻƜ ƒÜßǴ3  ɟɎɘɘj ǢŎŖĐj ǢP“3©3 In the wilds is a dead doe; With white rushes we cover her. There was a lady longing for the spring; A fair knight seduced her.  In the wood there is a clump of oaks, And in the wilds a dead deer With white rushes well bound; There was a lady fair as jade.  “Heigh, not so hasty, not so rough; Heigh, do not touch my handkerchief. Take care, or the dog will bark.”33 Literally, this poem tells a romantic story of a knight (according to Waley’s translation) who encountered a beautiful woman and his affair with her. Arthur Waley translated shui Đ into “handkerchief” with a footnote saying: “Which was worn at the belt.” In Wen Yiduo’s ɒ#Ó study, “shui” was used to cover woman’s private part, i.e., it is a kind of underwear. 34 In these modern readings, this poem is very erotic.                                                         32 About the Shijing and the Mao school, see Steven Jay Van Zoeren, Poetry and Personality: Reading, Exegesis, and Hermeneutics in Traditional China (Stanford: Stanford University Press, 1991), 7-115. See also The Book of Songs, trans. Arthur Waley (New York: Grove Press, 1996). Of the major translations and research aids, I have relied chiefly upon Waley, trans., The Book of Songs; and James Legge, trans., The She King or The Book of Poetry (Taipei: SMC Publishing Inc., 1991). 33 The Chinese text is quoted from the Maoshi zhengyi 。ʒƹɇ (photo-reproduction in the Shisanjing zhushu Œ%ȾǑȍ, Taipei: Yiwen, 2007), juan 1, 65-66; Waley, trans., The Book of Songs, 20-21; and James Legge, trans., The She King or The Book of Poetry, 34. 34  See Wen Yiduo ɒ#Ó, Shijing tongyi ʒȾʺɇ (Wuhan: Hubei renmin chubanshe, 1994), 339-340.  28 In the Maoshi, each poem was introduced by a “lesser preface” (xiaoxu þĚ) employing a didactic interpretation that stated the poem’s title, topic, ritual use, and sometimes supplied paradigmatic historical events for the poem. 35 The preface of this poem reads: “Ye you si jun” expresses disgust at the lack of ritual. Throughout the realm there had been great disorder, and oppressive men insulted women, so that lascivious manners spread. Through the transforming influence of King Wen, even in such an age of disorder, there was still a dislike of the lack of ritual. 4˔ƒƾ̜5ŊǢȣ33Ö'Õ6ĩƊș˯!ŕǕ̐3ɺŶǵ/ˆ˺ȋ6)ǰŊǢȣ3336 This implies that this is a poem of satire, showing disgust at the chaotic and disorderly world, a world where rites has been ruined. But the meaning of the lesser preface is actually very ambiguous. Zheng Xuan composed his annotation based on the Maoshi, and then followed the interpretation offered in the lesser preface with his own annotation. His annotation reads: “Lack of ritual” means [getting married] without a matchmaker and proper betrothal gifts, i.e., [the girl] was forced to marry. This refers to King Zhow’s (the last king of the Shang dynasty, c. 1600-1046 BC) time. ǢȣɍǨ(ȇæÞ˵Ĕ(ɚɖCŕƂʙȹ/)337                                                         35 See Steven Jay Van Zoeren, Poetry and Personality, 80-115. 36 Maoshi zhengyi, 65. 37 Maoshi zhengyi, 65.  29 Zheng follows the lesser preface and meanwhile guides the reading to the ritual of marriage, which neither the lesser preface nor the poem ever mention. While annotating “There was a lady longing for the spring / A fair knight seduced her,” Zheng writes: There was a chaste virgin thinking of meeting the knight according to ritual in the mid-spring. The fair knight asked a matchmaker to accomplish it accordingly.  ƒʣÜŁEƄCȣɛȈƑ¥ÎPæ[ˁŕ/338 For this poem, Zheng claims that the lady was a “chaste virgin,” and that the knight asked a matchmaker to help him complete the ritual of marriage and then marry her. Zheng’s annotation of “Heigh, not so hasty, not so rough” reads: The chaste virgin wanted the fair knight to come according to ritual. She asked the knight not to be so hasty, nor to be so rough. Meanwhile, she is disgusted at the lack of ritual at that time when oppressive men insulted women [and warns the knight not to do so]. ʣÜƵ¥ÎCȣQɘɘǤɟ33—ȏƇǢȣĩƊ/Ȉșɖ339 In this reading, the chaste virgin told the knight to follow the rites and warned him not to violate ritual. In Zheng’s reading, the purport of this poem is to praise the chaste virgin and set a moral model for people to learn from, and the core meaning and significance of this poem is about a particular notion of rites. As mentioned above, Zheng Xuan tried to unify all the Confucian Classics according to the theory of ritual and the Classic of Rites. His efforts are demonstrated in this annotation.                                                         38 Maoshi zhengyi, 65. 39 Maoshi zhengyi, 66.  30 The voluminous sub-annotations compiled in the Tang dynasty collected a large number of previous annotations and texts of other Classics, and tried to make a “Correct Meaning of the Five Classics” (wujing zhengyi ;Ⱦƹɇ). But in fact, the sub-annotations follow Zheng Xuan’s annotations; his annotations were treated as the “correct meaning” and all other quotations were used to support his arguments. If there were gaps between the Mao preface and Zheng’s annotations, the sub-annotation would attempt to close them. Sub-annotations purported to make all the Classics perfect, as least ostensibly.40 The lesser preface, Zheng’s annotations, and the sub-annotations of this poem mention “li” (rites, ritual) fifty-three times,41 while, in the Song scholar Zhu Xi’s (ƚǧ, 1130-1200) Shijizhuan ʒ˸\ (Collected annotations of the Shi), it does not appear once. Zhu Xi, known in the English literature as a Neo-Confucian scholar, philosopher, and politician, doesn’t interpret the poems according to ritual or the Classic of Ritual. What concerned him was to lead human nature to follow Heavenly Principle (tianli ÖǺ), so his interpretation focused on human beings’ inner nature and moral cultivation.42 About the main meaning of this poem, Zhu states: This stanza recounts the girl refusing the knight. She told the knight to come slowly, neither touching her handkerchief, nor alerting her dog, in order to say emphatically that he could not approach her. The idea of her chaste inviolability is clearly shown.                                                         40 See Van Zoeren, Poetry and Personality, 116-150; Sun Qinshan, Zhongguo guwenxianxue shi, 349-396.  41 Maoshi zhengyi, 65-66. 42 See also Van Zoeren, Poetry and Personality, 218-249.   31 ƺȫ-ʷÜçŢ//ʴʅãİİɎQǁƒŖ/Đǁ̖Ŗ/ǭCȁʅl(ɕș˜33lpǤ(¡Ǯ/ōɮ¡ʀȜ343 In Zhu’s interpretation, the persona became a heroic woman who sharply defended her honor. However, his annotation is a little self-contradictory. It first states that the girl told the knight to come slowly, and then states that she wanted to make the knight not able to approach her. This might be so because, literally, it is hard to get the sense from the poem that she fiercely refuses this knight. She is half refusing and half accepting him, and perhaps enjoyed this romantic tryst. This poem, according to Zhu’s philosophy, is thus morally corrupt. One of Zhu’s students, Wang Bo ǵƠ (1197-1274), despised this poem as a “lewd” (yinben zhi shi ǕÚ/ʒ) and advocated excising it from the Shijing.44  After the Song dynasty, various annotations continued to emerge. Some of them proposed new interpretations of the Shijing, others not. But most, if not all, were meaning-oriented. Every annotator tried to explain his own understanding of this corpus of ancient poems.45 For different scholarly lineages, not only were their interpretations different, but the text of the Classics they transmitted and annotated was also not the same. Ni Qixin Zlĸ has pointed out that the text of the Confucian Classics was actually “a multi-layered complicated overlapping construction” (ÓĄƴȓɽ˻˓ȌƬŕ).46 Different schools transmitted different texts, which                                                         43 Zhu Xi ƚǧ, Shijizhuan ʒ˸\, in Zhuzi quanshu ƚçgƏ (Shanghai: Shanghai guji; Hefei: Anhui jiaoyu chubanshe, 2002), vol. 1, 418-419. 44 See Wang Bo ǵƠ, Shiyi ʒȎ, ed. Gu Jiegang ̎̋| (Beijing: Pushe, 1935), 26-32.  45 About the interpretive history of the Shijing, see also Bruce Rusk, Critics and Commentators: The Book of Poems as Classic and Literature (Cambridge and London: Harvard University Asia Center, 2012). 46 Ni Qixin Zlĸ, Jiaokanxue dagang Ƣ„îÕȿ (Beijing: Beijing daxue chubanshe, 2004), 79-85.  32 were altered by generations of scholars, so that the text of Classics became more and more complicated. Each annotation was actually composed based on a particular version of the text. Few collators realised this, however. Having the idea that all annotations of one Classic were made based on the same text, most collators did not make any distinction between any two different texts annotated by different schools/scholars; i.e., they did not distinguish the study of the content of the Classics—such as the thought, political and social ideas— from the editing of the text. They altered the original text of the Classics according to their own study, which often confused a Classic’s textual lineage, and as a result, the editing of the text was trapped into a vicious cycle, whereby the more a text was edited the more it needed editing. The influential eighteenth-century Chinese scholar Duan Yucai was the first to advocated telling apart the “authenticity of the original text” (ěƙ/Ɔ̃) from the “veracity of the argument” (Ȫʖ/Ɔ̃), then making clear which text was annotated by which annotator and which annotator actually annotated which text. But in practice, Duan Yucai was still preoccupied with the idea of determining one authentic text according to his own understanding of “authentic principle.” His overconfidence in his evidential method prevented him from realising that the text he had collated was just one new text created by himself and that his annotation of it was another new interpretation of the Classics, just as previous annotators and collators had done. The prominent classicist Gu Guangqi ̎ĢÄ (1770-1839) noticed Duan Yucai’s self-contradiction and argued with him. He insisted that scholars should not alter the text and he carried out this rule when editing the Confucian Classics. Gu was very clear in theory and practice that the text was the text, annotation the annotation, and that different  33 annotations, even collations of the Classics, were based on different understandings of the Classics.47 The well-known modern textual scholar Yu Jiaxi Mº˞ (1884-1955) remarks: All scholarly works, ancient or modern, were created to meet practical needs for convenience. When transmitted for a long time and studied deeply, the meanings and principles [of them] become obvious. It is of course not correct that someone would consider the Book of Changes as songs on the hexagrams and the Spring and Autumn Annuals as court reports, and yet it is also incorrect to say they were not initially composed as such. ]îɹlwǢ(¼7ø/́ɿɎǨ/ǐCU[Ȅ\/.Ȟ/ȷɎįɇǺɫǡ3ĹƵC4ƃ5Ǩƶ4Ƅȥ5ǨƕɾƗ¡Ɏʙlâƙ(ǨƺɎOz=̃3348  Yu recognizes that the meanings and principles of the Classics were added by interpreters and annotators most of the time. Combined with different annotations, various texts of each Classic are of historical significance; they are an important part of Chinese textual culture.  Besides Confucian Classics, annotations of the texts of the Hundred Schools during the Spring and Autumn and Warring States Periods, although orally transmitted, were also meaning-oriented.49 So were annotations of histories during the Han dynasty, which was pointed out by Zhang Xuecheng:                                                         47 About the conflict between Duan Yucai and Gu Guangqi, see Ni Qixin, Jiaokanxue dagang, 307-316. 48 Yu Jiaxi Mº˚, Muluxue fawei ȗ˘îȐĵ (Beijing: Shangwu yinshuguan, 2011), 141. 49 See Lewis, Writing and Authority in Early China.  34 The grand historian [Sima] Qian composed one hundred and thirty juan and said [he would] “hide his book away in a famous mountain and transmit it to the right person.” Soon after, his grandson Yang Yun started to disseminate this book. After Ban Gu’s death, his History of the Western Han could not be thoroughly understood by scholars at that time. Thereupon, Ma Rong went to prostrate [himself] at Ban Gu’s house and learn from Ban’s little sister. Henceforth the study [of this book] started to became well known. [Si]ma and Ban’s books are familiar to people today, but had to be transmitted by particular persons and learned from particular teachers. This is because for specialized scholarship, ancient scholars must hand down personal insight in addition to the study of the scriptures, which was beyond the capacity of writing and should be transmitted orally from masters to disciples. They learned and handed down their scholarship to posterity. [Sima] Qian’s book was annotated by Pei Yin, [Ban] Gu’s by Ying Shao. The later annotators, in different families’ scholarly traditions, all express their family teachings. ¢ˆɫȒ%Œȳ-:ɱ/§ą\/l[30lįÒíƨŋâĎlƏ3ǹ¾4ǚƏ5ə¾į#ƇîɍƗɕʺƌ3̒ɶ-G˦'ĴlÜħ›ƩǤįlîâ̏3×̒ǹ/Ə][ʀ/ŅȜɎȋž\/ĹCl[›ʝĹƒŚəɍ[ûˡ/îĹƒǐÒ\ĸȯ{/Ś(˜zœũlIJɎșɛ\ɉlƩCÅ《.33ˆƏəɼ̔ǨǑ¾ƏəŒ‚OʄlįǨ/ǑɍǰɣĕözȔ˨löîɍ33Annotations of the Confucian Classics and texts of the Hundred Schools, and of histories in the Han dynasty were meaning-oriented, aiming to build a systematic interpretation of the thoughts,                                                         50 Zhang Xuecheng, Wenshi tongyi jiaozhu, juan 3, 237.  35 intentions and principles of the attributed author(s) of the text. This kind of annotation comprised the majority of pre-modern Chinese annotations. Annotations of histories and classical literary works (i.e., classical poetry and prose) after the Eastern Han dynasty were mostly concerned with explaining difficult or obscure words or expressions rather than building a systematic interpretation of the main text. However, some were still to some extent meaning-oriented. For instance, there was a tradition in Chinese literary theory that poetry was believed to express one’s intent ʒʅĻ . Therefore, annotators were supposed to base their explanations on their comprehension of the author’s intent. In short, annotations of poems do not have many literary appreciations of the main text, which is rather different from another type of interpretive text, pingdian/commentary.  Pingdian/Commentaries: In-depth Understanding of Literary Features Guo Shaoyu ˍȽɴ (1893-1984) claims that “if the Qing literati considered the Six Classics all histories, then the Ming literati read the Six Classics simply as literature.”51 This may not be correct for Ming annotations—books with “zhushu” in their titles—for, as is pointed out by Kai-wing Chow, the new annotations in the late Ming challenged the Cheng-Zhu Ȧƚ  interpretation established in the Southern Song dynasty (1127-1279) and thus helped to undermine the orthodoxy of the official ideology.52 These annotations, meaning-oriented as was pointed out above, cannot                                                         51 Guo Shaoyu ˍȽɴ, Zhongguo wenxue pipingshi +ŶŶîŝʐ¢ (Shanghai: Shanghai guji, 1979), 441. What should be noted is that “wen” Ŷ (literature) discussed in the imperial period is different from the modern concept “wenxue” Ŷî (literature). In this sense, Guo’s argument seems a little anachronistic. However, the essential feature of Chinese pingdian/commentaries were that they were mainly concerned with literary features of the text. Guo’s argument is right on this point. 52 Kai-wing Chow, “Paratext: Commentaries, Ideology, and Politics,” 149-188.  36 be understood as only concerned with the literary features of the Confucian Classics. However, for books with “pingdian” in their titles, Guo’s claim is correct.  Chinses pingdian, consisting of piping ŝʐ  (comments, criticism) and quandian ¿̣ (emphasis marks, literally it means “circles and dots,” similar to underlining and highlighting), can be translated as “commentary” in English. Different from annotations, Chinese commentaries were mainly concerned with the literary features, or “wenli” ŶǺ, of the main text.  Wenli as a literary concept was first proposed by Liu Xie }† (courtesy name Yanhe Ĭ­, fl. 5th century) in his highly influential work on literary theory, Wenxin diaolong (The literary mind and the carving of dragons). Wenli literally means “the order and organization of composition, the line of thought in writing,” but was often used to denote “literary features of the text” or “literary theory.” Speaking of the Confucian Classics, Liu Xie said: The truths (yi) [contained in the Classics] shape human nature and the affections (hsing-ch’ing) ... the language (tz’u) is the most finely wrought in the principles of literature (wen-li).  ,4;Ⱦ5ɇŽÆ0ŃŇʴ=‰ŻŶǺ353  Liu pointed out that the Confucian Classics were not only refined and profound in meanings, but also exquisite and elegant in language, which should be the model for later composers. Liu focused on the literary features of the Classics, so, after reading Liu’s Wenxin diaolong, Shen Yue ǎȺ (courtesy name Xiuwen HŶ, 441-513), one of the best-known scholars of the Southern Dynasties                                                         53 Stephen Owen, Readings in Chinese Literary Thought (Cambridge, Massachusetts and London: Harvard University Press, 1992), 195. See also Fan Wenlan ɥŶǞ, Wenxin diaolong zhu Ŷĸ˹̦Ǒ (Beijing: Renmin wenxue, 1962), 21.  37 Žƕ (420-589) praised Liu as “having an in-depth comprehension of literary theory” ǖijŶǺ .54 Zhang Xuecheng regarded the Wenxin diaolong, Shipin ʒ® (Classifications of poets), Wenfu Ŷʫ (Rhyme prose on literature) and some other works on Chinese literature as the origin of commentary.55 This is because commentaries focused on the literary features of the text and at the same time drew from these works a concept and general theory of style, language, structure, and other rhetorical features. Commentaries also had an “in-depth comprehension of literary theory.” Commentaries came into being during the Song dynasty (960-1279) under the influence of the development of civil service examinations. The earliest extant anthology with commentaries on classical prose, the Guwen guanjian Ŷ˧˛ (Key to composing classical prose) by Lü Zuqian «ȡʚ (1137-1181), was compiled to “show the right entrance to students” るîɍCˡı . The Chonggu wenjue ćŶʋ (Instructions in classical-prose composition) by Lou Fang Ʈƀ (Southern Song dynasty) was compiled to “highlight the key parts and benefit later students” šl˧˛Cʼnįî ; Wang Shouren ǵð? (style name Yangming ˱Ɓ, 1472-1529) stated that the Wenzhang guifan Ŷȫʱɥ (Rules and criteria of classical prose composition) by Xie Fangde ʛƞij (1226-1289) was “compiled only for the enterprise of the civil service examinations”ƆDZǨɝƩɍʌɏ .56 These books with commentaries have been primarily concerned with the                                                         54 Yao Silian äŁġ, Liangshu ƦƏ (Beijing: Zhonghua shuju, 1973), juan 50, 712. 55 Zhang Xuecheng, Jiaochou tongyi tongjie Ƣʟʺɇʺʄ, annotated by Wang Zhongmin ǵ˓LJ (Shanghai: Shanghai guji, 1987), juan 1, 12. 56 Siku quanshu zongmu tiyao »ĞgƏɀȗūɿ (Beijing: Zhonghua shuju, 1965), juan 187, 1698; Yao Bao äǸ, “Chong gu wen jue yuan xu” ćŶʋ”Ě (in Lou Fang comp., Chong gu wen jue, Siku  38 literary features of the main text, especially rules and patterns of composing classical prose. In these commentators’ eyes, methods of prose composition turned out to be derived from prose written by the wise men of the past. Therefore, one should be a sophisticated reader before becoming a good writer. The Guwen guanjian starts with “Kan wenzi fa” ȚŶéǐ (Methods of reading texts), which is then followed by “Lun zuowen fa” ʘOŶǐ (Methods of prose composition). In the main body of the book, the rules and patterns of prose composition are elicited by refined comments and evaluations of the literary features of the main text. In this way, “the right entrance” to civil examination candidates was shown.57 Annotations of ancient classics were meaning oriented, aimed at transmitting the perceived intention and thought of the past sages by explaining their words and sentences; commentaries were literary style-oriented, trying to lead readers to “comprehend literary features from the characteristics of the main text” ĴŶƙǬŃ+̊ņŶîǬŃ ,58 i.e., they left aside the author and only appreciated the text per se. Even for the Confucian Classics, what commentaries focused on were also literary aspects of the texts; their remarks on the meaning of the classics were largely clichés. For example, there is a commentary on the Shijing titled Maoshi zhenya 。ʒŦ˷ (The restoration of elegance in the Maoshi) printed in the Late Ming period. Each page of this book has three registers; in the top and bottom registers are various commentaries; in the middle are the                                                         quanshu edition), 1a; Wang Shouren ǵð?, “Wenzhang guifan yuan xu” Ŷȫʱɥ”Ě (in Xie Fangde ʛƞij, Wenzhang guifan, Siku quanshu edition), 1a. 57 See Lin Gang Ɵĉ, Ming Qing xiaoshuo pingdian ƁǗþʖʐ̣ (Beijing: Beijing daxue chubanshe, 2012), 44-57; Wu Chengxue ªŞî, “Xiancun pingidan diyi shu: lun Guweng guanjian de bianxuan, pingdian jiqi yingxiang” ǶêʡǠȮ#4..ʠ4Ŷk˟5ȓɂʹʡǠ˜lĮ¯, Wenxue yichanŶî,の 4 (2003): 72-84. 58 Lin Gang, Ming Qing xiaoshuo pingdian, 92.   39 main text of the Shijing and selected annotations by Mao. The commentaries on the top margin also explain the meaning of the poems. However, they are in very simple language, and more importantly, they explain the meaning through analyzing the relationship between different stanzas and the structure of the whole text. They contribute nothing to the creating of new meanings and any deeper understanding of the poems. The meanings they discussed were the literal meanings proposed in the Mao annotations. The commentaries on the bottom consist entirely of appreciations of the text. As for the “Ye you si jun,” one comment reads: The meaning of the two characters “huai chun” [longing for spring] is sublime. Do not read them too coarsely. œƄ08éȁĵɧȶȚ3 Another one reads: The four characters “shu er tui tui” [Heigh, not so hasty, not so rough] are much better than beautiful diction.  ɟɎɘɘ0»éàȁ9̞ʑ359 Compared with Zheng Xuan and Zhu Xi’s annotations quoted above, these commentaries are rather distinctive.                                                          59 These two commentaries are quoted from Maoshi zhenya 。ʒŦ˷ (woodblock edition carved in the late Ming period, held in the Research Institute for Oriental Cultures at the Gakushuin University), juan 1, 19b-20a.  40  Commentaries on poetry and prose showed new possibilities of interpretation in Chinese literary history. When being applied to works of fiction and drama, this new interpretive method changed the development of Chinese narrative literature. Fiction and drama occupied a very low position in elite assessments of Chinese writings. They were despised by almost all literati before the late Ming period; accordingly, composers and compilers were apt to abandon the right to sign their real names. This, in consequence, resulted in the absence of the author in fiction and drama.60 Poetry was believed to express one’s intent; prose, to convey the Way of the past sages; histories and texts of the Hundred Schools, to record the past wise men’s deeds and thoughts. The reading and interpretation of these texts was centered on a powerful “author” behind the text. However, there was no sage or wise man behind the text of fiction and drama most of the time. The absence of the author invalidated the author-centered interpretation, and exacerbated the low status of fiction and drama. A breakthrough in interpretive theory was needed to change the situation. This breakthrough was made by the fiction-drama commentaries of the Ming-Qing period, which did not concentrate on the sages’ intention and or the Way; nor did they see the text as a mere container of meaning. On the contrary, the commentators emphasized that meaning came from the characteristics of the text and that literary features were no longer a dispensable auxiliary. They believed that one was able to grasp the essence of a text by pinpointing its literary features. The great seventeenth-century literary theorist and commentator Jin Shengtan ˕ɐƷ (c. 1610-1611), stated: I have said that the Shui-hu chuan (Shuihu zhuan) is superior to the Shih-chi (Shiji), but nobody believes me. Really, I wasn’t talking nonsense. The truth is that in the Shih-chi words                                                         60 See Martin W. Huang, “Author(ity) and Reader in Traditional Chinese Xiaoshuo Commentary,” 51.  41 are used to carry events [i-wen yün-shih CŶʿ7] , while in the Shui-hu chuan events are produced from the words [yin-wen sheng-shih ¼ŶȂ7]. When you use words to carry events, you first have events that have taken place in such-and-such a way, and then you must figure out a piece of narrative for them. To use words to produce events, on the other hand, is quite different. All you have to do is follow where your pen leads. To cut down what is tall and make tall what is short is all up to you. 61 In Jin Shengtan’s opinion, in good works the text comes first. Meaning and significance do not inhere in the events and philosophy recorded by the text, but in how the text is written—the genre, rhetoric, structure, and other literary features. If, as Jin Shengtan put it, the text is more important than the events and philosophy recorded in the text, then the text is more important than the author. In this sense, fiction-drama commentary had liberated the text from the overweening control of the author, and the field where meaning and significance were generated is transferred from the author to the text per se.  A reader had to read the text carefully and analyze its characteristics thoroughly before being rewarded. And so “the heart of the reader suffers” ȚƏ[ĸɤ , wrote Jin Shengtan in his commentary on the Xixiangji ɾĠʉ (Romance of the Western Chamber). Interestingly, in the margin of the same page of Jin’s commentary, a piece of marginalia by a Qing scholar Wei Jirui ̘˳ǽ (courtesy name Shanbo µI, 1620-1677) states:                                                         61 Jin Shengtan ˕ɐƷ, “How to Read The Fifth Books of Genius,” trans. John C.Y. Wang, in How to read the Chinese Novel, ed. David L. Rolston (Princeton, New Jersey: Princeton University Press, 1990), 133.  42 I always say that reading is more difficult than composing, and yet Shengtan’s words are also saying that reading is more difficult. This was known only to Shengtan and me. Hearing this, Shengtan will certainly guffaw and yell at me: “You lousy devil!” ŖʙȚ˼ŻOǤɐƷƺʔ=ƆʅȚ˼ŻO3ƺňŖ˜ɐƷəȝ/3ɐƷɒ/Ĺ³ǤɄŖƍ ɋʨɋʨ062  Therefore, for works of fiction and song drama, even if the author was known, he was not the only originator of meaning; readers could play a more active role in producing meanings under the guidance of commentators, who can also be seen as sophisticated readers. This, as is pointed out by Martin Huang, indicated the “ascendancy” of the reader’s status. Also in this vein, Haun Saussy claims that this theory of commentary is similar to that of New Criticism, both paying great attention to the importance of close reading of the text.63 As mentioned above, some annotations, especially those of literary works in the Ming-Qing period, shared features with commentaries in that they also analyzed the literary features and discussed methods of composing poems and prose. Some fiction-drama commentaries also used zhu in their titles, because the majority of their contents were the explanation of words and expressions. Nevertheless, I am still inclined to call them “commentaries,” because the way they explained words and expressions was not the same as that of annotations.                                                         62 Wang Shifu ǵøȅ, Louwailou dingzheng tuozhu diliu caizi shu ƮÒƮʆƹáʎȮiŜçƏɾĠʉ, commentated by Jin Shengtan and annotated by Zou Shenmai ˎɐɗ (Early Qing edition cut by Zou Shengmai, held at the University of Auckland Library Special Collections), juan 2, 5ab. 63 See Huang, “Author(ity) and Reader in Traditional Chinese Xiaoshuo Commentary,” 50-51; Haun Saussy, “The Age of Attribution: Or, How the ‘Honglou meng’ Finally Acquired an Author,” Chinese Literature: Essays, Articles, Reviews (CLEAR), Vol. 25 (2003): 119-132.  43 The main concern of annotations was to transmit the authentic intent and thoughts of the past “sages.” So, finding the correct meaning of words and expressions in particular historical contexts was the core task of annotations. The method was usually twofold: (1) etymological investigation, i.e., exploring the original meaning and derived meanings of words and expressions; and (2) evidential research on historical events and contexts so as to make clear the accurate meaning of words and expressions in particular contexts. All of this was built on serious and reliable (at least in their own opinion) sources and logic. Therefore, this kind of interpretation was historical in essence. Even for the Learning of Principle (lixue Ǻî) and Learning of Mind-and-Heart (xinxue ĸî) during the Song and Ming dynasties, which came under the influence of Buddhism and were well known for their enthusiastic discussions of moral principles, spiritual experiences, and other philosophical ideas, annotations still raised points of discussion via annotating the Confucian Classics. In order to argue with the Han-Tang scholars, they had to do some etymological investigation and clarify historical events, as well. Commentators, by contrast, concentrated on literary features of the text. Having inherited the characteristics of the poetry and prose commentaries, writers of fiction-drama commentaries focused on language, rhetoric, structure, and such.64 Thus, commentaries did not care much about the etymology of the words. In the meantime, since fiction and drama were fictional, there often was no historical context that needed explaining. Therefore, what commentaries were concerned about in the case of ambiguous and important words were the cultural implications accumulated across history and the aesthetic features generated in this process. Fiction-drama commentaries in the Ming-Qing period were filled with various popular texts that annotators did not even bother to glance at, such as miscellaneous notes, Buddhist and Daoist texts, poems and prose, other fiction and drama, and folk stories, just name a                                                         64 See Lin Gang, Ming Qing xiaoshuo pingidan, 99-177.   44 few. Some commentaries cited many poems and lyrics of song dramas not only to illustrate where the main text might have come from but also to help readers appreciate the multi-layered meanings of the main text.  Let us take the Xixiangji as an example. The last two sentences of Act Three, “An Exchange of Verses,” are as follows:  Never again will I seek in dreams the blue palace gates/But only wait beneath the peach flower tree. n(¨̂Ǿ˩Ôe+—z–ˊȟƤɢƲe'Ȱ365 Zou Shengmai’s commentary on the phrase qingsuota ̂Ǿ˩ states: Door decorated with blue lattices    Meng Kang said: “The hollow part of the door was colored in blue.” [Yan] Shigu said: “The blue lattice means to carve into the shape of a chain of lattice and colour it in blue.” The inner part of the door is called ta. Fan Yanlong’s poem writes: “Being an official within the door decorated with blue lattice; Gazing from a distance at the phoenix pond.” ̂Ǿ˩  ìğƍ ˡĈȉřʵ˜+30đƍ ̂ǾɍyǨʼǾŶɎĈÊ/330ˡfƍ˩3ɥĬ̦ʒ Űô̂Ǿ˩˃Ɣ̙rnj3066 This commentary first explains the meaning of qingsuo ̂Ǿ by citing annotations of the Hanshu, and then explains the meaning of ta ˩. At this time, the meaning of this word is clear enough. But                                                         65 Xixiangji, juan 1, 32B. Translation from Wang Shifu, The Story of the Western Wing, trans. Stephen H. West and Wilt L. Idema (Berkeley, Los Angeles and London: University of California Press, 1995), 143. 66 Xixiangji, juan 1, 32b.  45 Zou Shengmai continus to quote Fan Yun’s (courtesy name Yanlong, 451-503) poem to show where this word came from, since Fan Yun was the first to use qingsuota. Similarly, Zou’s commentary on beiyejing ʢɪȾ (palm leaf sutra) writes: Palm leaf sutra    Buddhist sutras in western regions are mostly written on pattra leaves, so these sutras were called palm leaf sutras. Luobin Wang’s poem writes: “The palm leaf [sutras] transmit the words from the golden mouth [of the Buddha].” Liu Zongyuan’s poem writes: “I will hold palm leaf books in my leisure time; and read them after walking out of the Eastern Study.” ʢɪȾ  ɾÇNȾÓCʢɪƏ/ŲȾ§ʢɪ3̕ʩǵʒ ʢɪ\˕œ30ơó`ʒ ˣŤʢɪƏƻsƝ̥ʝ3067 This commentary also explains the word first, and then quots poems by two well-known Tang poets to illustrate where this word came from and how it was used historically. Interestingly, the first poem was composed by Meng Haoran ìǔǤ (689/691-740 ) rather than Luobin Wang. This mistake, as well as lots of other mistakes and the hastily-carved appearance of the pages, shows the low quality of Ming popular editions (see Figure 2-1). — Sometimes there are only citations of poems without any explanation of the words. These poems can contribute little for explaining the meaning of words. In fact, some poems were even more elusive than the main text per se. Nevertheless, pointing out the relationship of the words of the drama with that of classical poems could flaunt the extraordinary taste and erudition of the author and the commentator. For the reader, it would be a very different aesthetic experience to read these two parallel texts at the same time.                                                          67 Xixiangji, juan 1, 34a.  46  Figure 2-1  Leaf 1.34b of the Xixiangji, early Qing edition cut by Zou Shengmai. Source: General Library Special Collections Asian Languages PL2693 .H472. University of Auckland Libraries and Learning Services.   47 Fiction-drama commentaries were not produced to trace the original meaning of a text or transmit the past sages’ intent and thoughts; they instead encouraged aesthetic appreciation. The method they employed was not primarily historical investigation or evidential research, but rather the use of a set of literary concepts to build a system of appreciation. For instance, it is written at the beginning of the fourth act of the Xixiangji that there is a ceremony on the fifteenth day of the second month. Zou Shengmai claimed that the date should be the fifteenth day of the third month rather than the second. But in his commentary, there is no evidential or historical evidence and principle adduced. Rather, he refers to another drama and analyzes the writing strategy and its effect to illustrate what it means to be good writing composed by a talented author. In Zou’s analysis, what was concerned was “marvelousness” (shenmiao Ȣà), “spirit and reason” (shenli ȢǺ), and “metaphor” (biyu 、¹).68 This is to evaluate the main text and even determine its authenticity according to its literary features.  Fiction-drama commentaries aimed to explore the aesthetic connotations of the main text and guide the reader in appreciating its “marvelousness” so as to be moved by the main text. They employed textual characteristics as a means to stir the reader’s sentiment, and, sometimes, to educate common readers. At the beginning of juan 42 of the Sanguozhi yanyi %ÀĻǙɇ (Romance of the three kingdoms), Mao Zonggang’s 。óĉ commentary has: As for the pleasure of reading, if there is no great surprise, there is no great joy; no great uncertainty, no great delight; no great worry, no great comfort. When Zilong fought his way out and was tired with his horse, he encountered Wen Pin who was chasing him. This is a first worrying situation. When he saw [his lord] Xuande, [Xuande’s son] Ah Dou was silent. This                                                         68 See Xixiangji, juan 1, 34b-35b.  48 is an uncertainty. When Yide broke the bridge, Xuande was forced to the edge of the river and had no way out. This is a second worrying situation. When Yunchang meets [Xuande] on the overland route, their way was suddenly blocked the way by warships and they did not know that was Liu Qi. This is a surprise. When they boarded Liu Qi’s ships, they suddenly encountered another group of warships. They did not know that was Kongming, and this again is a surprising and worrying situation. [All of these] are like showing the fierce thunder coming and going and the furious billow rising and falling in the eye of the reader; it is hard to imagine that there can be such a marvellous illusion within one square foot [of paper]. ʝƏ/ƭ(Õ̖z(Õ·(ÕȎz(ÕĽ(Õłz(ÕŐ3ȋç̦ǀs˓Á[½̒1/į—ʾŶɑʸQƆ#ł!Ɏ˜ʀdzĶ/Ƈœ+˪ŷ(ʀɓńƆ#Ȏ!ɚɊĶźƳ/įdzĶɺƐŭʸɚNjˉƎǢ–ʰ—#ł!˜˾ˠſʰŪŒ/įĿʀNj&ŗɠůʰ(ȝƆ}ǻ—#̖!˜}ǻ¦ʲ/įĿ—ʀŗɠůʰ(ȝƆèƁ—#Ȏ#ł3Bʝɍț+ßǯ̀/#–#Qŀǜ/#ʬ#ɩ3(ō…ē/f-ƒßƺ/Ę3369 At the beginning of that juan, the commentary points out that the text of the Sanguo zhi yanyi can stimulate the sentiments of shock, doubt, and worry by a series of narrative strategies, which can even create the effect of “showing the fierce thunder coming and going and the furious billow rising and falling in the eye of the reader,” giving the reader a kind of extreme multi-sensory experience, and thus great pleasure from reading. All of these, the commentary reminds us, are completed “within one square foot [of paper].” One can only gain these sensory experiences and                                                         69 Luo Guanzhong Ʌʤ+, Sanguozhi yanyi %ÀĻǙɇ, Mao Zonggang 。óĉ, ed. (Hong Kong: Shangwu yinshuguan, 1974), juan 7, 25.  49 pleasures through close reading of the main text. More interestingly, one marginal comment in the Xixiangji mentioned above reads: “Interesting words…I cannot stop laughing” (ʭʔ11Ȭ(L).70 This response vividly illustrates the pleasure of the reader experiences while reading the Xixiangji and Jin Shengtan’s commentaries. Feng Menglong ̓Ô̦ (1574–1646), one of the greatest vernacular writers and editors of the late Ming Dynasty, also held the idea that stories had much infectious power. In his “Preface” to the 1620 edition of Stories Old and New, Feng writes:  Just ask the storytellers to demonstrate in public their art of description: they will gladden you, astonish you, move you to sad tears, rouse you to song and dance; they will prompt you to draw a sword, bow in reverence, cut off a head, or donate money. The faint-hearted will be made brave, the debauched chaste, the unkind compassionate, the obtuse ashamed. One may well intone the Classic of Filial Piety [Xiaojing] and the Analects of Confucius every day, yet he will not be moved so quickly nor so profoundly as by these storytellers.71 In Feng’s eyes, the stories of the storytellers can move an audience or reader more quickly and profoundly than the Confucian Classics. This confers recognition on the function and significance of fiction. Achieving this goal—to be moved by a story—depends on close reading of the text. Therefore, there is usually a chapter called “How to read” (dufa ʝǐ) at the beginning of a work of fiction or drama, composed by the commentator and guide the reader how to read according to the commentator’s suggestions. Moreover, the circles and dots used by the commentator, and                                                         70 Xixiangji, juan 5, 16a. 71 Feng Menglong ̓Ô̦, comp., Stories Old and New: A Ming Dynasty Collection, trans., Shuhui Yang and Yunqing Yang (Seattle and London: University of Washington Press, 2000), 6.   50 sometimes the simplest comments “miao”  à (marvelous) and “hao” Ý (good) remind the reader to pay attention to the text in a particular place. Commentators were actually reading experts, whose thoughts and literary theories unfolded in their commentaries. In this sense, we can see that the commentators also controlled the meaning of the main text; some of them even controlled the text per se — for instance, Jin Shengtan ofter altered the main text and claimed that he did so based on an “ancient text,” which was actually non-existent.72  Therefore, it can be concluded that annotations and commentaries have one thing in common, that is, they actively express their political, philosophical, and literary thoughts through the interpretation of pre-existing texts,73 which is different from marginalia.  Pijiao/Marginalia: Hand-Written Reading Responses In her book, Too Much to Know: Managing Scholarly Information before the Modern Age, Ann Blair discusses how a person’s reading notes helped him or her and other readers read. Because of this kind of usage, reading notes contributed much to the compilation of reference books in early modern Europe.74 The most direct function of reading notes was to help the reader read, re-read, and make sense of the main text. Some readers expected to publish their reading notes in the future even when they started to write them. This kind of note can be thought of as a draft of annotations, commentaries, or even monographs. But according to the sources I have encountered, readers rarely intended to publish his or her reading notes or make his or her voice heard by the public,                                                         72 See David Rolston, Traditional Chinese Fiction and Fiction Commentary: Reading and Writing between the Lines (Stanford: Stanford University Press, 1997), 4. 73 See also Huang, “Author(ity) and Reader in Traditional Chinese Xiaoshuo Commentary,” 53. 74 Ann Blair, Too Much to Know: Managing Scholarly Information before the Modern Age (New Haven: Yale University Press, 2010), 62-116.  51 and reading notes were merely a by-product of reading. Since most reading notes are records of the reader’s thoughts while reading, their content ranges from exegesis to comment, from historical investigation to textual criticism, from scholia to records of the reader’s daily activities. Thus, reading notes are essentially fragmentary in character. Because they seemed trivial, even meaningless sometimes, almost all were thrown away intentionally or unintentionally in both early China and Europe. In Europe, reading notes started to be treated as long-term tools and carefully preserved in the Renaissance,75 while in China, one kind of reading notes, pijiao, gradually became prevalent in the mid-seventeenth century. Some of them are well preserved in a tremendous number of extant pre-modern Chinese books. In Chinese, the word pijiao consists of two morphemes, pi ŝ (comments) and jiao Ƣ (textual criticism or collation), and refers to all writings and marks on the margins and interlinear parts of the book. Its English equivalent is marginalia.  Marginalia, the plural form of the Latin marginale, originally meaning “notes written in the margin,” is thought to have entered English from Latin in the early nineteenth century. 76 In Heather Jackson’s pioneering book, Marginalia: Readers Writing in Books, “marginalia” refers to “notes written anywhere in a book, and not merely in the margins.”77 The primary feature of marginalia is that they are hand-written notes attached to pre-existing written or printed texts. As Jackson puts it, “the essential and defining character of the marginal note throughout its history is                                                         75 Blair, Too Much to Know, 61-74. 76 Jackson, Marginalia, 7. 77 Jackson, Marginalia, 13.  52 that it is a responsive kind of writing permanently anchored to pre-existing written words.”78 Some scholars use “manuscript annotations,”79 “readers’ notes,”80 and the like to denote marginalia. In this study, “marginalia” will be employed to avoid any confusion with annotations and commentaries. Marginalia, annotations and commentaries are three different types of interpretive texts. The core feature of annotations is to bring out the meaning of the text. Commentaries in China concentrated on the literary features of the main text and aimed to help the reader enjoy the beauty of the main text. These two were both intended to be revealed to the public, or to be “published,” like a book composed by a writer who, most of the time, not only wants his or her voice to be heard by as many people as possible, but also wants to be understood by them. Marginalia have four basic features that differentiated them from annotations and commentaries: They are 1) being hand-written; 2) being responsive to the main text, but not confined by it; 3) privacy; and 4) uniqueness. (1) Being hand-written by the reader is an essential characteristic of marginalia, which ensures that marginalia are responsive to the pre-existing main text by the reader, and hence able to reveal the practice and mental process of reading. If marginalia are printed alongside the original text, editorial intervention should be taken into account, and the mental process of reading will undoubtedly be contaminated. These “printed marginalia” will hence become less different from traditional annotations, which are normally written or printed simultaneously with the original text                                                         78 Jackson, Marginalia, 81. See also Sherman, Used Books; Antony Grafton, “Is the History of Reading a Marginal Enterprise?” Monique Hulvey, “Not So Marginal: Manuscript Annotations in the Folger Incunabula,” Papers of the Bibliographical Society of America 92 (1998): 159-176.  79 Hulvey, “Not So Marginal,” 159-176.  80 Jackson, “Marginal Frivolities: Readers’ Notes as Evidence for the History of Reading,” in Owners, Annotators and the Signs of Reading, eds. Robin Myers et al. (New Castle, DE: Oak Knoll, 2005), 137-151.  53 as a kind of interpretation of or complement to the original text, and are designed to go along with the text they interpret.81 When marginalia are transcribed (guolu ˀ˘) by hand onto another copy of the original text, their content and form will likely be altered significantly by many scribes. The transcription of marginalia in China was a topic of great significance. I will discuss this practice in chapters four and five, and use several case studies to illustrate how this practice revealed the transcriptionist’s editorial intentions, opinions on scholarship, and attitudes towards reading. (2) The second feature of marginalia is that they are responsive to the original text, but not confined by it. Glosses, phonetic notations, commentaries, collating notes, paraphrases, extracts, cross-references, and analyses, which are related to the meaning of the main text are the main content of annotations and marginalia, but marginalia contain more information. Readers might copy onto the margin some words, phrases or sentences — which seem elegant or full of wisdom, or useful to the reader in some way — rather than just underline them; they might write down reminiscences or very personal opinions, which have nothing obvious to do with the meaning of the original text but may have been “triggered” by it; they might proudly put down their imitations of the calligraphy of the main text; they might curse the cat that urinated on the book; they might record their mood while reading, such as the marginalia on the Xixiangji discussed above; they might talk about the weather of that day, or draw various symbols or doodles as they are “using,” not just “reading” the book in their hands.  Annotations and commentaries both aimed to establish a kind of discourse with the main text and/or the author — explaining the meanings of original texts so that they can be understood in a                                                         81  See Raymond Clemens and Timothy Graham, Introduction to Manuscript Studies (Ithaca: Cornell University Press, 2007), chapter three “Correction, Glossing, and Annotation” and chapter eleven “The Bible and Related Texts,” 35-48, 181-191.  54 particular way; making some supplements or corrections to “vindicate” the argument of the original text; or sometimes arguing with the author. All in all, they are fixed texts that can be revealed to the public. However, sometimes, marginalia are readers’ personal opinions, which can be radical, caustic, offensive, or obscure, seemingly meaningless, and of course not ready for “publishing.” Moreover, marginalia can stand in no direct relationship to the main text. Some marginalia make plain that using a book, not just reading the text, can have other possibilities or consequences that may have no obvious relationship with the main points or literary meaning of the original text. In his research on Guillaume Budé’s (1468-1540) marginalia in Pliny, Vitruvius and Homer, a case study of scholarly reading practices in early-modern Europe, Anthony Grafton shows how a text could be used in various ways, and: Budé’s case suggests … reading had at least two central purposes, both practical, but neither familiar to us. The first was documentary, even archival: the scholar set out not only to converse with the ancients, using the classical tradition, but to document the progress he made in doing so. He made the books he read into a monument of his scholarship as personal and splendid as the books he wrote. ... reading was, among other things, a way for ambitious, powerful mean to assemble cultural capital for themselves and their friends and families. …… Budé also read ... to write. What he mastered, he could quote and interpret; what he annotated, he could reprocess. Reading in early modern Europe — at least learned reading — implied copying and sorting as well as scanning.82 This is a reading practice that is not particularly familiar to modern scholars and readers, but the situation was similar in late imperial China. For example, He Zhuo, a well-known scholar and                                                         82 Grafton, “Is the History of Reading a Marginal Enterprise?” 155-6.   55 calligrapher in his time, composed an enormous amount of marginalia on the Hou Hanshu, some of which show how he accumulated knowledge and learned how to write rhyming couplets. For instance, in juan 43 of the Hou Hanshu, the main text states: Previously, Emperor Guangwu and [Zhu] Hui’s father, [Zhu] Cen, both studied in Chang’an. They were old friends. After ascending the throne, [Emperor Guangwu] asked about [Zhu] Cen. At that time, Zhu Cen had already passed away. So [his son Zhu] Hui was invited [to the capital] and appointed as a Court Gentleman. wdƼɛƉǩĆYîˠñƒɞŲ˜’K》²ĆƇč- ƉţǨˋ3He Zhuo wrote the following piece of marginalia on this passage: Zhu Cen could be the topic of parallel couplets with Zhang Chong, [because] Cen had a son and Chong had a grandson. ƚĆ¡ýĨaĆƒçaƒí3	He Zhuo meant that Zhu Cen and Zhang Chong had the same experience. According to juan 45 of the Hou Hanshu, the well-known scholar-official Zhang Pu Ĩˑ  learned Classics from his grandfather Zhang Chong when he was young. Zhang Chong was Emperor Guangwu’s classmate and old friend. When Emperor Guangwu ascended the throne, he also asked about Zhang Chong, but Zhang Chong had also passed away at that time.85 Zhu Cen and Zhang Chong were both                                                         83 Hou Hanshu (woodblock edition cut at the Jigu pavilion [Jigu ge Ǎ˦] of the Mao 。 family in the late Ming, held at the Peking University Library; abb. PKU edition hereafter), juan 43. 84 Hou Hanshu, PKU edition, juan 43. 85 Hou Hanshu, PKU edition, juan 45.  56 Emperor Guangwu’s classmates; Zhu Cen’s son and Zhang Chong’s grandson were both celebrated scholar officials. So, He Zhuo said they can “be the basis for parallel couplets.” Sometimes, He Zhuo himself composed couplets and wrote them down on the margin of the book. For example, Zheng Hong’s ˏĦ biography (in juan 33 of the Hou Hanshu) recorded that Jiao Kuangǣʦ, Governor of Hedong Commandery, was implicated in a rebellion, and all his family were arrested. Jiao Kuang himself died on the road when he was being sent to the capital. The emperor was in a rage, and no one dared to argue for Jiao Kuang. Zheng Hong was one of Jiao Kuang’s students. He managed to redress the injustice done to Jiao Kuang and escorted Jiao Kuang’s family back to their home. A very similar story was recorded in juan 31 of the Hou Hanshu: Lian Fan ġɥ was also Xue Han’s ɰǚ student, and did the same thing for his teacher.86 When He Zhuo read Zheng Hong’s story, it brought to mind Lian Fan’s, and he thus composed a pair of couplets in the marginalia, as follows: Lian Fan has Xue Han’s corpse buried properly; Zheng Hong defended Jiao Kuang regardless of his own safety. ġɥŵɰǚˏĦʊǣʦ3)These examples show that He Zhuo did not just read to get the main point of the text; his reading experience is rather complex.  (3) Grafton’s study also shows that writing marginalia transformed Budé’s “printed books into unique [books],” which were “personalized possessions that documented his social position                                                         86 Hou Hanshu, PKU edition, juan 31 and 33. 87 Hou Hanshu, PKU edition, juan 33.  57 as well as his prowess as a scholar.”88  This refers to the third and fourth characteristics of marginalia: privacy and uniqueness. “If ‘private’ means exclusive to oneself,” Heather Jackson claims, “then reading is not a private but a social experience and practically any reader of a given time and place is as typical as any other.” 89 In another paper, Jackson shows how “books were passed around, using annotation as a way of sharing knowledge and opinions, sometimes as a semi-public reviewing process.”90 She defines marginalia as “semi-public,” and writes: No marginalia of the Romantic period were written under conditions of privacy. ... Marginalia were not then the secret utterances that they have for the most part become, but semi-public documents. ... These notes were designed for use, for show, for persuasion; they were oriented towards others, not self. They do not provide direct access to the mental processes of readers as they appropriate texts.91 Her argument seems reasonable. But “private” is a scalar concept of degree most of the time. Compared with diaries and secret letters, marginalia seem more public; compared with annotations and commentaries, they are more private. Designed to be revealed to the public (or a group of people), annotations and commentaries usually have very evident intentions. They either claimed to be objective, transmitting and interpreting the real meaning of the original text, to be disinterestedly telling the truth, or can be very specific and contextual, leading the interpretation                                                         88 Grafton, “Is the History of Reading a Marginal Enterprise?” 147-148. 89 Jackson, Marginalia: Readers Writing in Books, 256. 90 Robin Myers et al., “Introduction” to Owners, Annotators and the Signs of Reading, Robin Myers et al. eds. (New Castle, DE: Oak Knoll; London: British Library, 2005), ix. 91 Heather Jackson, “Marginal Frivolities,” in Owners, Annotators and the Signs of Reading, 145.  58 in a specific way. But most readers’ marginalia have no such scruples. — We should keep in mind that many marginalia that were too personal were cut out when they were edited for publication. It is true that some marginalia, as Jackson points out, can be a kind of combination of public and private functions. Some marginalia, such as those written under severe political pressure, can be very public, i.e., they were composed for the public or some particular audiences, such as the emperor, head of a state, and the like. But this kind of publicness is based on the privateness of marginalia, for if they were not supposed to be a kind of “private” document, they would not capture so much attention. Therefore, the privacy of marginalia is different from case to case, but in the spectrum of privacy, marginalia lie between diaries and annotations. (4) Marginalia can transform a book into something unique, and are also unique by themselves alone. In this aspect, they resemble holograph manuscripts. Few authors want to write a work twice; few readers will write the same marginalia on one title twice. Normally, there is just one copy of one particular reader’s marginalia on a given title. Therefore, books with marginalia usually have special meaning for the owner. For instance, a celebrated modern Chinese historian and book collector, Zhou Yiliang ¬#ɡ (1913-2001), once wrote to a friend, “My collections are basically not worth mentioning. But among them, there are many that my hands have gone over from morning to night and to which I have added marginalia. That is why I can never forget them for a moment” (#ɡɱƏƙ‹‹(ʮˁ±l+ÓƕÑŬŧś€ŝǑɍƆCʳľľ(ɕļɏ); he added: “There are no rare books in my collection. But after reading and collating them and writing marginalia in them, I feel them becoming my old friends and cannot bear to abandon them” (#ɡɱƏǢǷƙJȾƢʝŝúįUʃŔŔßŲ[(ĺŨƧ/Ȝ).92 Books                                                         92 Zhou Yiliang ¬#ɡ, Zhou Yiliang quanji ¬#ɡg˸ (Beijing: Gaodeng jiaoyu, 2015), vol. 10, 17, 20.   59 were thus personalized by their marginalia. They also acquired special meanings for other readers. A late Qing Chinese scholar-official, Pang Zhonglu ̧˝ǿ (courtesy name Yunshan Ɂą, 1822-1876), borrowed one copy of Sanguozhi %ÀĻ (History of the three states) from his friend Weng Tonghe Ɉ¦̨ (courtesy name Shuping ™Ė, 1830-1904) and had it in his study for at least two years, because this copy had Weng Tongshu’s Ɉ¦Ə (courtesy name Zugeng ȡĜ, 1810-1865) marginalia on it.93 Weng Tongshu was Weng Tonghe’s elder brother. He was also Pang Zhonglu’s friend. Besides, Pang Zhonglu admired his personality and scholarly achievements. After transcribing all the marginalia into another copy of his own, Pang rebound this book and wrote a short colophon at the back, saying, “[This book] has stayed on my desk for two years. Every time I opened it, I felt as though I was seeing an old friend” (Ãƥ̌ɍ8ėǂ#ăˢßʀŲ[). Besides these four features that are shared by marginalia in both China and Europe, Chinese marginalia in the late imperial period have one more distinctive feature: textual criticism assumed a very prominent position, both in relative quantity and importance. For some scholars, even a bad edition could become unique and valuable once it was collated by a famous scholar and bore his marginalia. For instance, there is a copy of Hou Hanshu held at the Shanghai Library (call number: XS816534-73). It is a woodblock edition carved in the late Ming by a commercial printing studio, which would never have been considered a rare book by Qing scholars. However, after transcribing various marginalia — especially that about textual criticism — from another copy into this copy, the owner, Zhang Yu ȫ; (courtesy name Shizhi Ĥ/, 1864-1934), wrote in the colophon at the back: “Because of  my hard work on this copy, have I not added one more rare book in the world?” (M/……9ƆƏ3(—ǨÖÍˤË#µƙ0)                                                          93 This copy is now held at the National Library of China (call number: SB06086).  60 Moreover, uniqueness also adds more value to books with marginalia. “Books with readers’ notes,” Jackson said, “continued to be cherished.” “Readers’ notes potentially had commercial as well as sentimental value…and might even turn out to be publishable and profitable.”94 Because of this, forgeries of marginalia by well-known readers are not rare in the book market, and thus authentication occupies a fundamental position in the study of famous readers’ marginalia.   Forms and Circulation Annotations, commentaries and marginalia are not only different in content and features, they were also supported by different materials and circulated in different ways in pre-modern China.   Annotations: From Oral Transmission to Written on Paper, from Separation to Combination As mentioned earlier, the interpretation of the Classics was mainly transmitted orally from master to disciple before and during the Western Han dynasty. During that time, bamboo strips and silk were the main writing materials. However, perhaps because bamboo was heavy and silk expensive, few annotations were written down. Comparing the “Yiwen zhi” of the Hanshu with the biographies of scholars and classicists in the Shiji and Hanshu, we can find that transmitters were far more numerous than books with annotations. For instance, according to the biographies of scholars in the Shiji and Hanshu, there were more than thirty-five well known classicists who transmitted the teaching of the Yijing ƃȾ (Book of Changes) of the New Text School. We can draw the genealogy of their transmission (Figure 2-2). However, there were just thirteen works of                                                         94 Jackson, “Marginal Frivolities: Readers’ Notes as Evidence for the History of Reading,” 144-145.  61 annotation recorded in the “Yiwen zhi,” and there is no evidence showing that most scholars wrote down their interpretations. As for the Old Text School, it was recorded in the “Rulin zhuan” _Ɵ\(Treatise on Confucians) of the Hanshu that the famous expert on the Yijing, Fei Zhi ʧȘ, employed the Tuan Ī annotation, Xiang ] annotation, and Wenyan Ŷʅ annotation, among others, to interpret the book, and that another expert, Gao Xiang ̗ș, interpreted the book according to divination studies. But no annotated books by either of them are recorded in the Figure 2-2 Genealogy of the transmission of the Yijing of New Text School in the Western Han. Drew according to the biographies of scholars in the Shiji and Hanshu.    62 “Yiwen zhi.” The most plausible reason is that they did not write down their interpretations.95 Hu Pu’an ɔƱñ put it this way:  [Scholars who transmitted the new text in the Western Han dynasty] all have their family traditions and/or adhered to their masters’ explanations. [The teaching] was transmitted from master to disciple within their schools from generation to generation without any distortion. Although the Erudite positions were established by the state for all the schools, scholars stuck to the teachings of their own, which were orally transmitted from master to disciple. This was the era of the Modern Text School. [At this time,] although there were exegeses, no one needed them. ¤ƒöǐ¤ƙđʖ˄ș\ũDž(˻63Ãűĝ/ʌȪÎ˺mɎê/Ãđħ/œɏș\zÌð#ö/ʖ3ƺ]ŶöƇA˺ƒʈʏɎǢ́ʈʏɍ3(Here “exegeses were not needed” actually means “exegeses did not need to be written down.” The explanation of the Classics was mainly orally transmitted from master to disciple or from father to sons and grandsons, which facilitated the study and transmission of a particular Confucian teaching becaming the prerogative of a family clan or the members of a school. Annotations started to be written down in large quantities during the late Eastern Han period.)This might be because that the technology of papermaking improved and cheaper and lighter paper, a more convenient writing material, was invented during the Eastern Han and finally replaced                                                         95 See Zhang Xuecheng, Jiaochou tongyi tongjie, 78-81. 96 Hu Pu’an, Zhongguo xungu xue shi, 3.  97 For example, there were ninety-four titles on the Yijing recorded in the “Jingji zhi” of the Suishu (among them, sixty-nine titles existed in the Tang dynasty). See Suishu, juan 32, 912. 63 bamboo strips and silk, becoming the main writing material during the Six Dynasties iƕ (220-589).98 Before the Six Dynasties, the text of the Classics and the annotations were usually not written together in one book.99 But during the Six Dynasties, most of the Confucian Classics and some pre-Han texts were written along with the annotations. For example, the Confucian Classics found at Dunhuang ųǥ were mainly annotated editions.  Shu, which actually interpreted both the Classics and the zhu (annotation) and thus can be called sub-annotation, began to spring up in the Six Dynasties and flourished in the Sui and Tang. But they began to be written/printed together with the text of the Classics only in the Southern Song dynasty, when the civil service examinations were already the major entrance to the state bureaucracy. At this time woodblock printing was widely utilized to produce Confucian texts, and commercial publishing started to flourish. Woodblock printing facilitated the mass production of books. The popularization of Confucian Classics that resulted from the development of the civil service examinations created a large market for these Classics furnished with annotations and sub-annotations, which were highly convenient for examination candidates. These printed editions followed the basic pattern of previous manuscripts. That is, characters are arranged vertically from                                                         98 See Tsien Tsuen-Hsuin (Qian Cunxun) ˙êʈ, Paper and Printing. Vol. 5, part I of Science and Civilisation in China, edited by Joseph Needham (Cambridge: Cambridge University Press, 1985), 23-84; Tsien Tsuen-Hsuin, Written on Bamboo & Silk: the Beginnings of Chinese Books & Inscriptions, second edition (Chicago: University of Chicago Press, 2004), 145-152. 99 Zhang Xuecheng, Wenshi tongyi jiaozhu, collated and annotated by Ye Ying, juan 3, 248. 100 See Zhang Lijuan Ĩ̞å, Songdai jingshu zhushu kanke yanjiu òAȾƏʎȍvyȞȩ (Beijing: Beijing daxue chubanshe, 2013), 5. 101 See Zhang Lijuan, Songdai jingshu zhushu kanke yanjiu, 228-402; Gu Yongxin, Jingxue wenxian de yansheng he tongsuhua, 1-16, 38-122.   64 the top to the bottom in columns; columns go from the right to the left; the text of the Classics is in big characters, and the annotation and sub-annotation are in double-lined small characters. The biggest difference between printed Classics and previous manuscripts is that the latter were mostly bound in long continuous sections (such as long scrolls), while the former were printed in folded leaves (ye ɪ. See Figure 2-3).102  Figure 2-3a  The Chunqiu Guliang zhuan ƄȥȨƦ\, manuscript produced in the third year of the Longshuo ̦Ɠ reign (663), held at the National Library of China, call number BD15345. Source: Di er pi guojia guji zhengui minglu tulu Ȯ8ŝÀöǷʥȵ§˘Â˘, Beijing: Guojia tushuguan chubanshe, 2010, picture 2539.                                                         102 Some manuscripts were bound in sutra binding with pleat-like leaves and whirlwind binding with continuous pages pasted together.  These two formats were believed to be created under the influence of Buddhist palm-leaf books, and perhaps spurred the creation of folded-leaf books. See Tsien Tsuen-Hsuin, Paper and Printing, 227-233.  65   Figure 2-3b  First leaf of juan 2 of the Zhouyi zhushu , woodblock edition carved in the early Southern Song period by the Tea and Salt Supervisorate of the Eastern Zhejiang Province, photo-reproduction in Zhonghua zaizao shanben +ɨnʻµƙ.   Although the sub-annotations consist of multiple layers of annotation, and different layers are usually different in interpretation, the latest layer, which would have been composed by the compiler of the sub-annotation, usually selects one interpretation and tries to reconcile the gaps between the different parts of the Classics and those between the Classics and the interpretations.103  That is to say, although sub-annotations are multi-layered, different layers are chronologically arranged rather than randomly put together. The final explanation was based on all the layers of annotation listed, and no layer can be easily detached from the others. The sub-annotation itself                                                         103 See John Makeham, Transmitters and Creators, 79-167, 253-347.  66 was one completely integrated interpretive system and was able to circulate separately from the Classics.  Printing the Confucian Classics with their annotations and sub-annotations in one book was first practiced by local governments to provide more convenient versions of the Classics for examination candidates. Before long, commercial presses found that printing the main text and its annotations in one book could be a marketing strategy that attracted more consumers. They learned from the governmental press and produced more versions of Classics with annotations and sub-annotations. Moreover, they also applied this approach to the printing of other kinds of books. For example, the three annotations of the Shiji, the jijie ˸ʄ (collected annotation, by Pei Yin ɼ̔, around 430 AD), suoyin ȼ˴ (probing the meaning, by Sima Zhen £̒ʣ, 697-732), and zhengyi ƹɇ (correct meaning, by Zhang Shoujie ĨðȲ, around 1075 AD), were printed as one book in the Southern Song dynasty by Huang Shanfu ̟µ×, a publisher. This edition was not a good edition, for not only was it full of errors, but much of the zhengyi annotation was also cut out (see Figure 2-4).  This kind of commercial printing illustrates the influence of politics and commerce on the production and circulation of annotations. And this kind of influence was much greater when it came to the printing of commentaries.104                                                          104 On the impact of commercial publishing on the content and visual appearance of Chinese books, see also Lucille Chia, Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th-17th Centuries) (Cambridge: Harvard University Asia Center, 2002), 24-62; and Susan Cherniack, “Book Culture and Textual Transmission in Sung China,” Harvard Journal of Asiatic Studies vol. 54, no. 1 (1994): 5–125.  67  Figure 2-4  First leaf of juan 1 of the Shiji, carved in the Southern Song period by Huang Shanfu ̟µ×, photo-reproduction in the Zhonghua zaizao shanben. The three different annotations were chronologically arranged after the sentence they explained, and separated by small circles.   Commentaries: Reshaping Chinese Books Commercial editions that printed various annotations together in one book seemed unsatisfactory to many, and have drawn continuous criticisms since they were invented. But the commercial printing of commentaries on poetry, classical prose, fiction and drama seemed more acceptable and exerted a significant influence on the textuality and physicality of Chinese books.  Commentaries focus on the literary features of a text, but also need to make sense of the main text first. So, the explanation of difficult words and expressions also takes up a considerable part of the commentary and resembles annotations. Similar to annotations, if commentaries have multiple layers, these different layers are typically chronologically arranged. However, unlike the  68 annotation, different commentators’ commentaries were usually listed in parallel rather than being formed into one completely integrated interpretive system. This is caused by the distinctive features of commentaries: Annotations focus on the meaning of the main text and can circulate separately from the main text, hence their location on the page is not that important, compared with that of commentaries. Even for an edition that had various annotations printed together in one book, different annotations were usually chronologically arranged after one sentence or one paragraph of the main text (see Figure 2-4).  Commentaries, on the contrary, cannot circulate separately from the main text. This is a matter of structure, literary features, language, and the rhetoric of the main text. Detached from the text they analyze, commentaries become unintelligible. For example, without the main text, the huge number of miao (marvelous), hao (good), and emphasis marks in commentaries would make no sense at all.  The location of commentaries on the page is not insignificant. They must be as close to the text they analyzes as possible. Furthermore, since commentaries are essentially helping the reader to appreciate the text rather than seeking philosophical meanings or historical truths, they  unfold more interpretive space, and the conflict between different commentaries is usually not that fierce. The same text can be appreciated from different perspectives. Most of the time, these parallel appreciations add to the aesthetic experience rather than cause contradiction. In this sense, publishers, as well as some commentators, liked to print several commentaries in one book. If book illustrations can also be seen as a kind of interpretation, we can say that book illustrations were also printed in parallel with various commentaries, and were active in the construction of the meaning field of the main text. As there were usually a variety of commentaries on one text, and they must all be printed at or near the same place, printers needed to come up with a method to contain all of them and meanwhile draw a distinction between any two of them. In the late Ming,  69 when popular printing was flourishing, printing enterprises were operated by many literati, writers and scholars who were not only good at reading and writing, but also skilled at printing and even bookselling. They not only did not reject commercial operations, but also actively embraced a variety of strategies that modified the textuality and physicality of Chinese books.105 One of the most commonly used strategies to differentiate various commentaries was to print them in multiple registers. For example, the edition of Xixiangji mentioned above has two registers. The lower register takes up about four-fifths of the page and contains the main text of the Xixiangji, the commentary composed by Jin Shengtan, and the sub-commentary composed by Wang Zhuoshan ǵŸą and others; the main text of the Xixiangji is in single-lined big characters; Jin’s commentary is in single-lined medium characters, and the sub-commentaries are in double-lined small characters. The upper register occupies one-fifth of the space and consists of Zou Shengmai’s commentary (see Figure 2-1). In this way, not only are the main text and various commentaries differentiated by font size, but Zou’s commentary is also distinguished by being placed in a separate registers. In the Ming-Qing period, there were also a great many commentaries on the Confucian Classics and histories, which analyzed the literary features of the text and aimed to help examination candidates with their study of prose composition. These commentaries were different from annotations of the Classics and histories, so Zhang Xuecheng despised them for only “discussing literary features” and argued that they should not be classified in the category of                                                         105 See Lucille Chia, Printing for Profit, 24-62; Ōki Yasushi ÕƖğ, “Minmatsu Kōnan ni okeru shuppan bunka no kenkyū” ƁƘNjŽ86 7sǪŶˆ9Ȟȩ, Hiroshima Daigaku bungakubu 50:1 (1991): 1-173.   70 Confucian Classics or histories in book catalogues.106 The Maoshi zhenya discussed above is one of them. This copy has three registers; on the top and bottom registers are various commentaries, and on the middle are the main text of the Shijing and selected annotations by Mao (see Figure 2-5).  Figure 2-5  Two half leaves of the Maoshi zhenya, woodblock edition carved in the late Ming, held at the Lianci shuyuan ɯnjƏˬ in Baoding Wõ. This copy and the one held at the Research Institute for Oriental Cultures at the Gakushuin University are the same. Source: Baoding Lianchi shuyuan shanben tulu WõɯnjƏˬµƙ¢ (Beijing: Guojia tushuguan chubanshe, 2014), 27.                                                         106 Zhang Xuecheng, Jiaochou tongyi tongjie, 13.  71 Ming-Qing book illustrations, as Yuming He puts it, not only interacted with the literary text as separate representational media, but also worked together with the text and created something new: “a new kind of chapter, a new kind of book, and eventually a new vernacular language.” 107 Resonating with the literary text, illustrations can supply additional sublime meanings and generate different experiences for the reader. Moreover, even when the quality of a printed illustration was very bad, repetition of familiar stock elements in it allowed quick recognition and ease of comprehension and appreciation, which aided the reader in their own “imagining.”108  Book illustrations actually supplied visual hermeneutics to the literary text. They were also a kind of commentary. Therefore, the pattern of “pictures above and texts below” (shangtu xiawen &Â'Ŷ), which was very common in Ming-Qing fiction, can also be seen as a multi-register layout (see Figure 2-6). Another distinctive strategy commonly employed by Ming-Qing publishers was multi-color printing. As early as the 1970s, the bibliographer Wang Zhongmin ǵ˓LJ pointed out that the multi-color printing that came into being in the late Ming in Huizhou ķċ resulted from the prevalence of literary commentaries.109 Wang also claimed that the Classics, annotations, and sub-annotations were already transcribed in different colors in the manuscript era.110 But there is little evidence attesting to the prevalence of multi-colour transcription in the manuscript era. As mentioned above, before the Eastern Han, annotations were written separately from the Classics.                                                         107 Yuming He, Home and the World, 142-150. 108  Robert Hegel, Reading Illustrated Fiction in Late Imperial China (Stanford, CA: University of California Press, 1998), 312-326. 109 Wang Zhongmin ǵ˓LJ, “Taoban yinshua fa qiyuan yu Huizhou shuo” ÛǪ‘xǐʬǘŻķċʖ, in Wang Zhongmin, Lenglu wensou oģŶɳ (Shanghai: Shanghai guji chubanshe, 1992), 69-92. 110 See Wang Zhongmin, “Taoban yinshua fa qiyuan yu Huizhou shuo,” 70-72.  72 There was no need to differentiate different texts in one book. During the Six Dynasties to the Figure 2-6  One leaf of the Qimiao quanxiang zhushi xixiang ji Ùàgșʎ˒ɾĠʉ, woodblock edition carved in the 11th year of the Hongzhi ĦǏ reign (1488), held at the Peking University Library. Source: Qimiao quanxiang zhushi xixiang ji, photo-reproduction of the Ming edition (Shanghai: Shangwu yinshu guan, 1957), 2.57a.    73 Tang, some Classics and important texts were copied in one book with their annotations. To differentiate the annotation from the main text, the annotation was transcribed one space lower than the main text, or in double-lined small characters, or interlineally in very small characters, or on the back of the paper, or sometimes in different colors. All these devices, resulting from the difficulty of standardization in the manuscript era, were used to tell the main text and the annotation apart; there was less need to differentiate between different annotations. During the era of print, single-lined, big-character main text with double-lined small-character annotation became the standard pattern of Chinese books; devices like multi-color printing were not common.111  Because annotations, and sub-annotations as well, focused on the meaning of the text, their location in the book was not as important as that of commentaries. Even when different annotations and sub-annotations were printed in one book, there was still no necessity to tell them apart. However, the situation of the commentaries was rather different. Publishers were eager to print as many different of commentaries as possible in one place. To distinguish them, some publishers employed multi-color printing despite the cost. Multi-color printing was thought to make a book far more elegant and beautiful. The well-known Ming publisher Min Qiji ˥̤F wrote in the “Fanli” qR (General Remarks) of the Chunqiu Zuozhuan ƄȥČ\ (Zuo annotation of the Spring and Autumn Annals):  The old editions that had comments and emphasis marks were all in black ink, which was disliked in artistic and literary circles. Now I have carved another edition, with the main text and annotation in black and the commentary in red. This edition was revised three or more                                                         111  See Huang Yongnian ̠《ė, Guji banbenxue ȵǪƙî (Nanjing: Jiangsu jiaoyu chubanshe, 2012), 183-184.  74 times and the cost of money and knife was not considered. Having one in your room, all will appreciate it in their hearts. For those who are just starting their studies and do not care about the commentaries, there is the black-ink part. ɞyqƒŝʐ¿̣ɍY””Ǫ̑ɲƟ•/3]Ÿy#ǪȾ\ȄÌŝʐCƚ3Ƣʟ(´%;Ɏ˙t/̄̃ŚʇȜ3Ƀ/Ē+ȋǢ(ĸʪ3lwîʗƩǢšŝʐzƒÌƙÃ3112 The text of the Chunqiu and the Zuo Annotation is in black; the text of commentaries is in red. This shows that the status of the Classic-annotation and the commentary was different in the publisher’s opinion. Beginners studying for the civil service examinations could ignore the red commentary. This means that commentaries did not place much emphasis on interpretation of the Classics. They were more like guides for appreciating literature and learning how to write prose. And, as a matter of fact, these dazzlingly beautiful commentaries were for the wealthy reader to “appreciate in their heart” (see Figure 2-7). Sometimes, multi-register printing, multi-color printing, and illustrations were all applied in one edition. An edition of the Shijing produced in the Ming and held at the National Archives of Japan is one of these. This edition has two equal registers. For the beginning chapter of the book, the upper register is Shijing jindan huikao ʒȾ˕,ŠɌ (Collected studies on the golden elixir of the Book of Songs), and the lower register is the Shijing nanzi ʒȾ˼é (Difficult words in the Book of Songs). They are actually two different works on the Shijing. For the main body, the lower                                                         112 Min Qiji ˥̤F, “Minshi jiake fenci Chunqiu Zuozhuan fanli” ˥džöyuƴƄȥČ\qR, in Chunqiu Zuozhuan ƄȥČ\ (Ming multi-color edition, carved by the Min family, held at the Harvard-Yenching Library), 3b.  75 register is the main text of Shijing with Zhu Xi’s annotations, while in the upper register are Gu Qiyuan’s ̎ʬ`  (1565-1628) commentaries (see Figure 2-8). This kind of textbook was obviously more attractive to young examination candidates compared with those that crowded too many small characters in one page and were printed only in plain black.  Figure 2-7  First half leaf of juan 33 of the Shiji chao ¢ʉ:, woodblock edition carved by Min Zhenye ˥ŦƩ in 1620. Courtesy of the UBC Library.   76    Figure 2-8  Two half leaves from a Ming edition of the Shijing, held at the National Archives of Japan, photo-reproduction in the Zhonghua zaizao shanben. This book was actually a combination of four books: Maoshi zhengbian zhinan tu 。ʒƹʞťŽÂ (Guide tables of the Shijing with Mao’s annotations), Shijing jindan huikao ʒȾ˕,ŠɌ (Collected studies on the golden elixir of the Book of Songs), Shijing nanzi ʒȾ˼é (Difficult words in the shijing), and the Shijing with Zhu Xi’s ƚǧ annotations and Gu Qiyuan’s ̎ʬ` commentaries.     77 Marginalia: Anywhere, Any Color When did people start to write marginalia? There is still no firm consensus on this question. There is some evidence that readers drew punctuation marks and special symbols to help in reading as early as early as the Spring and Autumn Period. 113Various kinds of punctuation marks are also found in the Dunhuang manuscripts. But it is still unknown whether these symbols were made by scribes while transcribing the main text or by later readers.114 They cannot be called “marginalia” according to my definition. In the Three Kingdoms period (220-280), as recorded in the Weilüe ̘Ȋ (Brief history of the Wei dynasty), Dong Yu ɭʾ drew red and black symbols in the Zuo zhuan. Dong’s practice can be seen as the origin of drawing emphasis marks.115 These bi-color punctuation marks can be seen as the earliest known instance of marginalia. In the Song dynasty, more readers made emphasis marks on what they read, and some of the emphasis marks were in multiple colors.116 However, it should be noted that it is still not clear whether readers in the Song wrote literary comments on book margins or not. As discussed above, pingdian/commentaries also came into being during the Song dynasty. The most reasonable                                                         113 See Li Ling ƛ˿, Jianbo gushu yu xueshu yuanliu ȴďƏɛîɹǘǓ (Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2004), 121-122. 114 See Wu Chengxue ªŞî, “Pingdian zhi xing: Wenxue pingdian de xingcheng yu Nan Song de shiwen pingdian” ʐ̣/ɜ Ŷîʐ̣ȓīŕɛŽòȓʒŶʐ̣, Wenxue pinglun 1 (1995): 24-33; Li Zhengyu ƛƹï, “Dunhuang yishu zhong de biaodian fuhao” ųǥ,Ə+ȓựȭɵ, Wenshi zhishi 8 (1988): 98-101. 115 See the annotation of Sanguozhi %ÀĻ (Beijing: Zhonghua shuju, 1964), juan 13, 420; Wu Chengxue, “Pingdian zhi xing: Wenxue pingdian de xingcheng yu Nan Song de shiwen pingdian,” 25-26. 116  See Wu Chengxue, “Pingdian zhi xing: Wenxue pingdian de xingcheng yu Nan Song de shiwen pingdian,” 26-27.  78 process might be that commentators first drew emphasis marks and literary comments in a book, and then this book was sent to the publisher for carving and printing. These drafts of published commentaries, in hand-written forms, can be regarded as marginalia. Yet none of them survive today.  Extant books show that late-Ming and Qing scholars continued drawing emphasis marks. Meanwhile, there are some marginal writings—including literary comments, collating notes, and colophons—that were composed by several late-Ming scholars, while the composition and transcription of marginalia flourished in the Qing dynasty, becoming a very prominent cultural phenomenon.117 Therefore, this study will focus on the features of late-Ming and Qing marginalia. Marginalia were notes hand written by the reader while reading. It is not difficult to imagine that their location in the book enjoyed tremendous freedom. Some famous books were held and read by quite a few sophisticated readers and have various readers’ marginalia. The first reader to write marginalia in a book usually used red ink, or sometimes black ink. To distinguish between earlier readers’ markings and their own, later readers would commonly use different colors (see Figure 2-9).                                                         117 See Chen Xianxing ˮcɷ, Zhongguo guji gao chao jiao ben tulu +Àȵȧ:Ƣƙ¢ (Shanghai: Shanghai guji chubanshe, 2014); Wei Li ̅~, Pijiao ben ŝƢƙ (Nanjing: Fenghuang chubanshe, 2003); Zhongguo guji zongmu +Àȵɀȗ (Beijing: Zhonghua shuju; Shanghai: Shanghai guji chubanshe, 2009-2012).  79  Figure 2-9a  One half leaf of the  Hanshu, woodblock edition carved in the Wanli reign. It contains marginalia in black, red, blue and yellow ink. Courtesy of the Shanghai Library (call number: XS816474-513)  80  Figure 2-9b  One half leaf of the  Li Yishan shiji, woodblock edition carved in the Shunzhi reign (1644-1662). It contains marginalia in black, red, green and purple ink. Courtesy of the Shanghai Library (call number: XS6889).  81 Most marginalia were written at the margins of books. Books printed in the Ming and Qing dynasties usually had very wide top margins, called the “heavenly head” (tiantou Ö̌), which normally were two or three times wider than the bottom margins.118 Top margins were well-suited to marginalia, yet bottom margins were also employed. The majority of Chinese printed books after the Yuan dynasty were printed on large leaves with the text in the centre. While binding, the leaves were folded by the center and then stacked to form a book. The side margins became the book’s spine, and the center of the text was in the book mouth. Therefore, technically, these books had no side margin. However, European books often have large side margins. Therefore, Heather Jackson notes that “the side margins are universally, in English-language books, the favoured place for the readers’ running commentary on the text.”119 In both pre-modern Chinese and European books, interlinear notes, punctuations marks, and other symbols were very prevalent. Sometimes marginalia can be written on interleaved sheets of paper pasted into the book, or placed between two leaves of the book. Marginalia about personal ownership or commentary on the whole book are usually written on the flyleaf, title page, or the empty page at the end of a book.  The location of marginalia may impact contents, which Jackson discusses briefly in her book. She writes of European marginalia: The notes that are in the closest physical proximity to the text are the interlinear glosses that traditionally move word by word, as readers’ aids, translating or defining or paraphrasing the original. ... Marks and commentary in the margin of the same page, however, express a distinct position pro or con, or offer supplementary material from an external source, such as literary                                                         118 Leslie Howsam, ed., The Cambridge Companion to the History of the Book, 98. 119 Jackson, Marginalia, 28.  82 parallels or additional evidence. The index at the back extracts from the whole text just those passages that the reader might want to refer to again, and the summary judgment at the front or back formulates an opinion that is decidedly the reader’s and not the author’s.120 The situation in China was similar, but since textual criticism took up a considerable part of the marginalia in late imperial China, there was more space for character variants from other editions as well as scholars’ opinions on which characters were authentic. In some books, reader’s comments, which were usually not short, were put in the top margin, while variants and scholars’ short opinions were written in the bottom margin.  For pre-modern Chinese books, the layout of different editions was a great indicator for readers in distinguishing between them. Thus, information regarding the layout of various editions was usually carefully recorded in the table of contents or at the first page of the first juan. For example, in the first page of the table of poets of the Jixuan ji ƪdz˸ (The best poems, in Tangren xuan Tangshi °[ˇ°ʒ [Tang poems selected by Tang people], woodblock edition carved by the Mao family in the Ming dynasty, held at the Shanghai Library, call number: XS839789-96), He Zhuo’s marginalia says: The manuscript copy has ten columns in each half page, and every column has sixteen characters. [For the table of poets,] there are four names in each column and the number of poems was not recorded. şƙŒɷɷŒiéǂɷ»[§(ʲę̑3                                                         120 Jackson, Marginalia, 49.  83 From the colophon at the end of the book, we learn that He Zhuo obtained a manuscript copy and collated this book aganist the manuscript copy. This Ming edition has eight columns in each half page and nineteen characters in each column. In the table of poets, there is just one name in each column, and it also records how many poems were selected.  He recorded the layout of the manuscript, circled the poem number, made some revisions to the heading, and added the name and title of the compiler and commentator. This indicates that He not only wanted to preserve the detailes of an old edition, he also advocated keeping the original layout in carving a new edition (see Figure 2-10).  Figure 2-10  One half leaf of the Jixuan ji, in the Tangren xuan Tangshi, woodblock edition carved by the Mao family in the late Ming. Courtesy of the Shanghai Library (call number: XS839789-96).  84 In some cases, the layout was also discussed in the main body of the book. For instance, there is a rare book held in the Asian Library at the University of British Columbia, which contains lots of marginal commentary merely about the layout of the text. Titled Qian shu ̡Ə (Book of Qian), composed by Tian Wen Ȇ˽ (1635-1704) and carved in the 29th year of the Kangxi ğけ reign (1690), this book is a gazetteer about part of today’s Yunnan Province. On every page of the book, the characters go vertically in columns from the top of the frame to the bottom, and the columns go from right to left. This is the typical layout of most, if not all, books cut before Kangxi’s reign. However, at the top of page 1b, the marginal notes read: “Move [words from] ‘ƕ’ (chao, dynasty) to another line, do not indent” (ƕ0Ÿɷ̈ƣú. See Figure 2-11). The marginalia on page 2b read: “Move [words from] ‘ɐÖç’ (Sheng tian zi, the Sage Emperor) to another line and [print], do not indent” (ɐÖç0Ÿɷ̈ƣ). This kind of marginalia appears on every page which has words like “the Sage Emperor,” “the Great Empire,” and so forth. The dates of the marginalia are difficult to verify. From the style, ink and other tiny hints, however, we can estimate that it was approximately composed in the mid-Qing. The composer of the marginalia — possibly the collector of this book — is by no means to be identified, either. But the repeated marginalia about the layout show that he was uneasy about the original layout of the book and desperate to change it into a new one — to begin a new line upon mentioning the emperor’s names to show respect to the emperor, a device was “raising of lines” (taitou Š̌).  In the late-Ming and Qing period, marginalia usually circulated in three ways: (1) along with the original text in the book in which it was first written down; (2) transcribed onto another copy of the same title; or (3) in printed form, either with the original text or separately.  85  (1) One reader’s marginalia usually circulated along with the book he or she had read. Therefore, the circulation of marginalia is essentially the circulation of books, which is, in turn, the circulation of information and knowledge itself. Thus, previous theories and models about the dissemination and circulation of books and information, such as Robert Darnton’s Figure 2-11  One half leaf of the Qian shu, woodblock edition carved in the Kangxi reign (1661-1722). The marginalia writes: “Move [words from] ‘ƕ’ (chao, dynasty) to another line, do not indent.” Courtesy of the UBC Library.  86 “communication circuit” and Roger Chartier’s “interpretation circuit,”121 can provide instructive methodological reference points for us to study issues of who composed what kind of marginalia, who acquired them and how, who was involved in the process of dissemination, why one should be interested in marginalia, and whether there were differences in circulation between books with marginalia and books without. In addition, various catalogues and readers’ biographical documents supply abundant information with which to build a profile of the marginalia of a particular composer and that of a particular title. Bibliographical studies in China have a long history and occupied an important position in pre-modern Chinese scholarship. “Bibliography is the most important requirement for study. It is here that you must start and only then will you be able to find your way,” said Wang Mingsheng ǵ̚Ȗ  (1722"1797), a prominent Qing historian. 122  The relatively complete bibliographical system in pre-modern China supplies abundant records on books. Many descriptive catalogues and some abridged catalogues have records of the physicality and textuality of a book, the history of its circulation, details of marginalia in it, if any, and the like. Based on these records, we can build up profiles of marginalia by a particular annotator or about a particular title. Moreover, books once belonging to pre-modern Chinese collectors — especially Qing scholars and collectors —are systematically held at libraries in China and all over the world. The variety of library catalogues can also help us to find real books in the library. A reader’s biographical documents can supply similar information such as when, where, and from whom a book was acquired or just read, and whether or not the book contained any hand-copied marginalia. Marginalia written at the                                                         121 Darnton, The Kiss of Lamourette, 107-135; Roger Chartier, “Texts, Printing, Reading,” in The New Cultural History, 154-175. 122 Cited in Endymion Wilkinson ed., Chinese History: A New Manual, third edition (Cambridge: Harvard University Asia Center, 2013), 936.  87 front or back of the book were called “colophon” (tiba ̍ʯ ), and usually included such information as a summary judgment of the book, statement of ownership, history of circulation of the book. These are all very useful sources for the study of the circulation of marginalia.  (2) The transcription of marginalia is a widespread phenomenon in China. It was especially prevalent in the Qing. In this period, marginalia by famous scholars often circulated in more than one copy, transcribed by their students and other scholars who were attracted by their academic value, by calligraphers who appreciated the aesthetic value of the handwriting and imitated it in another copy of the same title, or by booksellers who saw the marginalia as marketable—they also usually imitated the handwriting of the original annotator and forged the annotator’s seals to pass their copies off as originals.  Transcribers’ intentions always become involved while transcribing. Students wanted to conceal their teachers’ mistakes; scholars concentrated on the “useful” information and left other things out; some unscrupulous booksellers’ haste was also widely known — they randomly left off quite a few marginalia, made mistakes out of carelessness, and intentionally cut information they thought was “sensitive” or did not suit contemporary sensibilities. Therefore, transcribed marginalia are distinct from the original marginalia not only in form but also in content. They are excellent sources for studying later readers’ reading responses to previous readers’ reading responses.  Compared with printed marginalia, transcribed copies are typically closer to the original. If the original marginalia are lost, transcribed copies can still be used to reconstruct the original by comparing various transcribed copies and investigating the annotators and transcribers. We need to find out how many transcribed marginalia are still in circulation, who transcribed them from which copy, and how they circulated.  88 (3) Transcription made marginalia more accessible, while printing enlarged the circulation circle of marginalia and made their transmission more efficient. Marginalia were normally printed in two ways: a) Marginalia on several titles were collected and then compiled into one book; b) Marginalia were printed along with the original text, creating a new edition. In both ways, marginalia were often thoroughly edited, and sometimes changed beyond recognition as a result.  In late imperial China, the transcription and editing-publishing of marginalia became a nearly universal scholarly practice. Therefore, it is one of the best points of access to late imperial Chinese scholars’ intellectual world.   89 3 The Reading Seed He Zhuo and His Marginalia  One of the most common fallacies in evaluating the accomplishment of scholars is to consider their philosophical or religious thoughts, political contributions, or social activities, all of which seem ostensibly practical, or  put simply, useful. One result of this approach is that excessive attention is paid to scholars’ written works and their political and social activities, and thus questions about their quotidian practices of scholarship, how they perceived their scholarly activities and identity, and who in reality contributed to the development of scholarship per se are largely left aside. Modern intellectual history owes some attention to scholars who have been forgotten either intentionally or unintentionally, and to the practices of scholars more broadly. Taking the perspective of scholarly practice, it can be found that some little-known or unknown scholars actually made great contributions to the enterprise of scholarship, and that some well-known scholars might be different from, or at least more complicated than, what we had thought. The following chapters will introduce some little-known or even unknown scholars, focusing on scholarly practices related to marginalia in late imperial China, examining how different scholars composed marginalia on printed books, how they transcribed and printed marginalia, how they expressed happiness, confusion, and other emotions in marginalia, as well as how they contributed to the development of scholarship and the intellectual evolution in the Qing dynasty. The story starts with an early Qing scholar He Zhuo. This chapter will examine He Zhuo’s life, his practices with marginalia, and the characteristics and influence of his marginalia.     90 The Reading Seed: He Zhuo He Zhuo was a book collector, scholar and calligrapher. He was well known in his lifetime and had approximately four hundred students. Even though he did not pass the jinshi ɪ© examination, he was given this title by the Kangxi öƐ Emperor (r. 1661-1722)1 and then worked as an academic advisor in the Southern Study (Nanshufang zʼnĢ, lit. the emperor’s southern reading room) and as an editor in the imperial printing office in the Hall of Military Glory (Wuying dian ūȕŰ). He was appointed the tutor of the emperor’s eighth son and established a lifelong intimacy with him, which caused him to be stigmatized after his death. He Zhuo was an avid reader and accomplished scholar specializing in textual criticism. Even when he was imprisoned because of his political enemies’ false accusations, he bore this hardship with equanimity and still devoted himself to reading and textual criticism. After his death, the Kangxi Emperor praised him as a “reading seed” (dushu zhongzi ɓʼnǕ¾).2 “Reading seed” refers to a person who loves reading and is able to disseminate and hand down a set of values and thoughts of tradition and culture. In pre-modern China, it was believed that                                                         1 Interestingly, He Zhuo happened to be born in the first year of Kangxi’s reign and to die in the year of Kangxi’s death. 2 There is no modern biography of He Zhuo. On He’s life, see the “Appendix” of He Zhuo, Yimen xiansheng ji ǹʎJƩʙ,  woodblock edition cut in Gusu ºȣ in 1833, held at the Peking University Library, photo-reproduction in Xuxiu siku quanshu DZBœôPʼn ; Shen Tong Ÿÿ ,  “Hanlinyuan bianxiu zen shiduxueshi Yimen He xiansheng xingzhuang” ǼŗʒǰBə?ɓ©ǹʎ;JƩȨƙ, in He Zhuo, Yimen dushuji ǹʎɓʼnȺ  (Beijing: Zhonghua shuju, 1987), 1275-1277; Quan Zuwang PLJŎ , “Hanlinyuan bianxiu zeng xueshi Changzhou He gong mubeiming” ǼŗʒǰBə©ʍſ;S§džʄ, in Yimen dushuji, 1278-1280; Qing shi liezhuan ƅ‹_F, in Qingdai zhuanji congkan ƅ3FȺ‡], compiled by Zhou Junfu •ʯÑ (Taibei: Mingwen, 1986), vol. 104, juan 71.  91 culture (wen ij), literary culture in particular, was related to the fate of the state, and that reading seeds were those who carried on this responsibility of spreading and handing down the essence of tradition and culture. The Helin yulu ʵ ŗƟʞ (Stories written in the forest at the Buddhist temple, lit. Jade dewdrops in the crane forest) recorded that Zhou Bida •Ď¯ (1126-1204) once said:  The two Kings Xian of the Han dynasty both loved books, so their states lasted the longest. Can we let reading seeds vanish from literati families? Ɖ)Ɯƶ¶ʼnȀVFŸƶŊɮ©¯²ÎV‰=ɓʼnǕ¾Ȭē"3 One of the well-known reading seeds before He Zhuo was the orthodox Confucian scholar-official Fang Xiaoru ĶÁà (courtesy name Xizhi éƽ or Xigu éˆ, 1357-1402) in the early Ming period. Fang was famous for his loyalty to the Jianwen ùij Emperor (r. 1398-1402). When the Jianwen Emperor’s uncle, Zhu Di œŝ, usurped the throne to become the Yongle Emperor in 1402, he summoned Fang Xiaoru and demanded that he write an inaugural address that would compare his usurpation of the throne with the regency of the Duke of Zhou during the reign of his nephew King Cheng of Zhou in the early Zhou period. Fang refused and was executed with his entire family. One of the Yongle Emperor’s tacticians once warned Yongle that, “If Fang is killed, the reading seeds under heaven will be exterminated” (ůÁðɓʼnǕ¾ǫǂ).4 The implication was that if Fang were executed, the moral principles that tied to that particular reading seed would be harmed, leading to disorder. Of course, this is hyperbole, since the transmission of                                                         3 Luo Dajing Ƿ¯ǭ, Helin yulu ʵŗƟʞ (Beijing: Zhonghua shuju, 1983), 212. See also Zhou Mi •Ð, Qidong yeyu ʺŖʀɂ (Beijing: Zhonghua shuju, 1983), 380. 4 Mingshi Ŀ‹ (Beijing: Zhonghua shuju, 1974), juan 141, 4017-4021.  92 culture and tradition was never the business of any one person. All scholars who devoted themselves to reading and writing, collecting and printing books, collating and transcribing the texts of books, as well as those who carried on the spirits and thoughts inhering in books, can be considered reading seeds.  He Zhuo spent his whole life collecting, collating, reading, and enjoying books. It is recorded that he wrote an abundance of books, but almost all of them were burned or scattered for political reasons. The Yimen xianshang ji ǹʎJƩʙ (Collected works of Master Yimen, 12 juan), consisting of a few of He’s classical prose writings, poems, colophons and letters, was compiled and printed approximately one hundred years after his death. The Yimen dushuji ǹʎɓʼnȺ (Reading notes of Master Yimen, 58 juan), compiled by He’s son, nephew, and students after his death, was a collection of He Zhuo’s marginalia on eighteen titles. However, He wrote marginalia on many more books. From the Zhongguo guji zongmu and catalogues of various libraries in China and abroad, we find that He Zhuo’s marginalia exists on more than one hundred titles and were transcribed by a variety of scholars throughout the Qing dynasty on more than two hundred copies of books (see Appendix I: Books Containing He Zhuo’s Marginalia and Their Transcriptions). He Zhuo’s marginalia cover all the four branches (sibu œɸ) of pre-modern Chinese books. The majority of his marginalia are of two kinds: 1) various comments (including literary criticism, historical and philosophical comments, and the like) on the content of the book; and 2) collation notes and collation symbols inscribed on the text. Living in a transformative period of scholarship when the Song-Ming metaphysical approach began to be seriously challanged and the relatively positivistic school of “evidential research” (kaojuxue ǾĪÂ) gradually came into being, He Zhuo’s comments in his marginalia inherited the characteristics of the commentarial tradition that  93 flourished in the Ming dynasty, and his textual criticism and evidential research—although it sometimes seemed preliminary—impacted the scholarship of the mid- and late-Qing period.  Ironically, even though He Zhuo never passed the civil service examination or got even a juren ȑ0 degree, he was an expert in selecting classical prose and writing literary commentaries as aids to students taking civil service examinations, and he was well known by his contemporaries for this enterprise. At the same time, his learning was also despised by some scholars as the “study of the papers’ tail” (zhiwei zhixue ǨÜ!Â), that is, the study of the eight-legged essays for the civil service examinations, which was considered as occupying the most inferior place in scholarly hierarchy.5 A noted mid-Qing scholar Yu Zhengxie AŨƒ (courtesy name Lichu Ƥ`, 1775-1840) stated that “He Zhuo annotated books in the way of annotating eight-legged essays, and thus became well-known under heaven because of [his selecting and annotating] eight-legged essays” (;Ǝ5Ńijƈ°ƪĥŃijżĥʼn).6 This claim is partly true. The so-called “way of annotating eight-legged essays” refers to literary commentaries on classical prose. As discussed in chapter two, this kind of commentary focused on the literary features of the text and aimed to help examination candidates learn how to compose classical prose. According to the Yimen dushuji and the existing marginalia by He Zhuo, this kind of commentary only takes up a very small fraction of his marginalia, which usually covers a very wide variety of topics and subjects. Furthermore, the concerns expressed in He’s marginal comments vary with the features of the text. For the Confucian Classics, He Zhuo talked more about the Learning of Principle, self-cultivation and                                                         5 See Wang Shaoying ŷǩƸ, “Ruanshi chongke Shisanjing zhushu kao,” ʐŲɿeuǭŽưǾ, Wenshi ij‹ 3 (1963): 54. 6 Yu Zhengxie AŨƒ, Guisi cungao ƴèÀǗ, in Congshu jicheng chubian ‡ʼnʙğ`ǰ, vol. 430, juan 14, 431.  94 other matters discussed extensively by Song-Yuan scholars. For historical works, he commented on historical figures, historical events, and the literary features of the historical text. He was a noted calligrapher, and thus his comments on books about calligraphy and rubbings from ancient inscriptions contain not only abundant information and knowledge about the work but also appreciation of the calligraphy. In addition, He was an outstanding poetry critic, and his comments on Tang poems were highly praised and widely influential.  He Zhuo was especially well-known for his expertise on the history of the Western and Eastern Han dynasties; he specialised in the textual study of the Hanshu and Hou Hanshu. Therefore, his marginalia on these books can represent his academic style and achievements. Furthermore, I have chance to gain access to several copies of his marginalia on the Hou Hanshu. So, I will take He Zhuo’s marginalia on the Hou Hanshu as an example and examine in detail of their characteristics and influences.   Reading He Zhuo’s Historical Comments In the Hou Hanshu held at the Peking University Library, more than half of He Zhuo’s marginalia are concerned with textual criticism. Some items discuss the literary features of the text, and several discuss the composition of rhyme couplets, as mentioned in chapter two. In addition, there are many comments on historical figures and events, historiography, previous scholar’s annotations, and such. His comments contain a great number of refined insights and fruits of his research. My examination will start with his comments on historical figures. In traditional Chinese historical writings, people always occupied center stage. For instance, more than half of the contents of the Ershiwu shi (the 24 Histories plus the Qingshi gao ƅ‹Ǘ), which were deemed the “standard histories” (zhengshi Ũ‹), consisted of historical figures’  95 biographies (liezhuan _F). Other sections were also largely biographical. Biographical writings thus were the most important component of the standard histories in Imperial China. In this sense, how to understand and evaluate historical figures was a crucial task for researchers and readers. Of He Zhuo’s comments in the Hou Hanshu, chronologically the third standard history of the twenty-five, many were about historical figures. He Zhuo usually tried to understand historical figures within their historical context. He also explained the course of history based by analyzing the historical figure’s personality, status, intention, and other aspects. In He Zhuo’s brief biography composed by Shen Tong, Shen praised He, saying: “In [Master Yimen’s] marginalia, the comments on people must have traced the historical context they lived in and examined the internal and external factors [therein]” (Zʦɐōɇ0ǿĎɞVČVȫȮ). This is to make an appraisal of personages by ways of investigating events they experienced. For instance, Can Yong’s Biography in juan 60 of the Hou Hanshu reads:  In the sixth year of the Zhongping era, Emperor Ling died and Dong Zhuo became the Minister of Works [one of the Three Dukes who were the paramount dignitaries of the central government]. Hearing that Cai Yong had a good reputation, Dong summoned him to be an official, but Cai refused pleading illness. In great anger, Dong Zhuo said: “I have the power to execute one’s whole clan. If Cai Yong is so arrogant, he will soon be punished like this.” Dong also urgently ordered the local governor to nominate Cai Yong to visit him. Cai Yong had no choice but to go [visit Dong Zhuo] and was appointed Libationer.                                                         7 Shen Tong, Hanlinyuan bianxiu zeng shiduxueshi Yimen He xiansheng xingzhuang, in Yimen dushuji, 1277.  96 îTïʟëãȜyƓǚȃɳʲɦ!ǖƲÛy¯đȼŇĠkȉĹ0ȟɳɫDɠǿĸɟǂ‚\İäɷȑɳȾóɳćçbǵLjɽ8 Cai Yong ȟɳ (courtesy name Boji 9š, 132-192) was a great calligrapher, musician and scholar in the late Eastern Han dynasty. While serving Emperor Ling, he advocated that ceremonial practices be restored and criticized the influence of eunuchs in politics. When the warlord Dong Zhuo came to power and controlled the central government, he summoned Cai Yong to serve at the court. Cai Yong was compelled to comply, and was criticized because of his weakness by later scholars. He Zhuo’s comment in the margin reads: Chen Liu was very close to Luoyang, [this is why] Bojie was inevitably contaminated. For the same reason, Ciming (Xun Shuang ȗƕ) could not do what Kangcheng (Zheng Xuan ɻƝ) had done. ʓƭžʕɨ9šģ5MķŴŘ'ęĿ!ćƓöğ' This comment mentions three historical figures, Cai Yong, Xun Shuang and Zheng Xuan, all prominent scholars in the late Eastern Han. While Dong Zhuo controlled the central government, he summoned all of them to serve him. Cai Yong and Xun Shuang complied, but Zheng Xuan did not, so that Cai and Xun were thought to have lost their moral integrity while Zheng kept his. Differing from this common opinion, He Zhuo explained these three scholars’ actions using geographical factors. Actually, according to their biographies in the Hou Hanshu, Cai and Xun                                                         8 The main texts of the Hou Hanshu and He Zhuo’s marginalia cited in this chapter are from the Hou Hanshu held at the Peking University Library (the PKU edition).  97 both had a reputation for refusing to serve the government in a turbulent political era. But, as He Zhuo pointed out, Cai Yong lived in Chenliu (Kaifeng, Henan Province), and Xun Shuang in Yingyin (Xuchang, Henan Province), both of which were not far from Luoyang, Dong Zhuo’s capital. It was difficult for them to flee with their family members, and thus they could do nothing but to comply. However, Zheng Xuan lived in Gaomi (in Shandong Province), far from Luoyang and Chang’an (Xi’an, Shaanxi Province). His biography in the Hou Hanshu states: “[Later,] Dong Zhuo moved the capital to Chang’an. The high ministers nominated [Zheng] Xuan as the prime minister of the state of Zhao. [Zheng Xuan] did not go because the road [from Gaomi to Chang’an] was cut off.” (ȜyɯɹʍÅSȑƝƓɜƾɭĵȍ)9 He Zhuo was probably inspired by this record and tried to defend Cai Yong and Xun Shuang. Using geographical factors to explain historical figures’ practices rather than arbitrarily excoriating them, He’s comments, as Shen Tong put it, “traced the historical context they lived in and examined the internal and external factors [therein].” In Chinese historical philosophy, one of the most important functions of history was to take it as a mirror and draw lessons from it (yi shi wei jian 5‹ƍʌ). To learn from history, the most difficult part is not just knowing the people or events, but also unravelling the basic roles of the course of history from the numerous and complicated historical events that comprise it. As the Grand Historian Sima Qian ʭɯ (145-86 BCE) put it, as a historian, he expected “to examine into all that concerns Heaven and the human, to penetrate the changes of the past and present” (Ǚ                                                        9 Hou Hanshu, juan 35, 1209.  98 °0!ʖɩˆ1!ɔ).10 Only then can we learn from history and solve new problems with the assistance of the wisdom derived from it. A commonly used method for historians to make sense of the past was “historical induction,” i.e., to sum up particular historical events, compare them, search for similarities and/or differences, and arrive at some general laws concerning the course of history and the vicissitudes of human society.  There is much historical induction in He Zhuo’s marginalia. Compared to the comments in the Nian’ershi zhaji ú)‹ŒȺ (Notes on the 22 Histories), an eighteenth-century masterpiece on the standard histories composed by the great historian Zhao Yi ɜǽ (courtesy name Yunsong ʛâ, alternative name Oubei Ƨs, 1727-1814), He Zhuo’s comments were less refined, neither sufficiently insightful nor well arranged. Scattered throughout the whole of the Hou Hanshu, they were long neglected by scholars. However, most of them were made based on He’s close reading of the Hanshu and Hou Hanshu, and his expertise in the history of the Han and instructive for readers. Shen Tong observed of He that “[While discussing historical events, He Zhuo] always knows the beginning and the end, and understands the whole event” (ĎɩVʬÜƺVɔ).11 For instance, in the “Baiguanzhi san” (Treatise on Officials Part III) of the Xu Hanzhi,12 the main text has the following entry:                                                          10 Sima Qian, “Bao Ren An shu” ¦6Åʼn, in Ban Gu, Hanshu, 2735. Translation by Burton Watson, in Sources of Chinese Tradition: From Earliest Times to 1600, vol. 1, comps. William Theodore de Bary and Irene Bloom (New York: Columbia University Press, 1999), 372. 11 See Shen Tong, Hanlinyuan bianxiu zeng shiduxueshi Yimen He xiansheng xingzhuang, in Yimen dushuji, 1277. 12 The current version of Hou Hanshu is actually made up of two parts. The annuals (ji ǧ) and biographies (zhuan F), 90 juan in total, were composed by Fan Ye and are called “Hou Hanshu” in this study. The  99 Palace Aide to the Censor-in-chief, one person, ranked at 1000 bushels (shi). Ĉ‹0v。 The original annotation has: The aide to the Censor-in-chief. In old times, [the Palace Aide to the Censor-in-chief] was different from the Supervising Censor (who working in the local government) in that they worked in the palace and secretly reported to the emperor the illegal behavior of officials. When the Censor-in-chief became Minister of Works, [the Palace Aide to the Censor-in-chief] was left in the inner central court and became the leader of the Censorate. Later, [the Palace Aide to the Censor-in-chief] belonged to the Chamberlain for the Palace Revenues. Ĉ‹¯²!'Ȓa○Ĉ‹¡ŰÐȑʠżƒĈ‹¯²ɤƓǚaƭƓĈ‹ȎƞĄ‚ßÙó About the change of the political duties and power of the Palace Aide to the Censor-in-chief, He Zhuo commented on the top margin: Bao Xun’s biography in the Weizhi states: “[Bao Xun] was appointed ‘Gongzheng’ in the fourth year of the Huangchu period.” “Gongzheng” is actually Palace Aide to the Censor-in-chief. This also followed the meaning of “Xiaozai” (Deputy Prime Minister) in the Zhouguan (Rites of Zhou). But this position was not taken by high ministers, so the name does not reflect reality. In the Western Han, [the Palace Aide to the Censor-in-chief] was still the vice prime minister, which followed the bequeathed law and meaning of the Zhouguan, while in the                                                         treatises (zhi ď), 30 juan in total, were composed by Sima Biao ʭ” (?-c. 306) and are called “Xu Hanzhi” in this study.  100 Eastern Han, it belonged to the Chamberlain for the Palace Revenues, so that the grand councillors could not ask about the personnel administration around the emperor. The grand councillors’ power was weakened and the old order was broken. 	ʳďʴoFʸ`œïƓÊŨÊŨ}Ĉ‹'ł.Ż	•ÈØÌ!ėƘ¯²ʥÓ#ȁȱ/ßhƾƚć	•ÈɱżȍŖ/ɤßÙófS昰¾åŒ!0(6ɣȀʱǬȰǂ Liu Zhao’s iŁ (lived in the Liang ś dynasty [502-557]) annotation of the Hou Hanshu citing a passage from the Zhouli, which states: “The Deputy Prime Minister is in charge of the laws about the officials in the palace, and puts into practice the governmental decrees in the palace.” Gan Bao’s annotation states: “[This] resembles the duty of the Palace Aide to the Censor-in-chief” (	•ǍØÌĨùɵ!Ë^5źƠË!Į4íÕŽŇȔĈ‹).13 Fan Ye’s original annotation also states that the Palace Aide to the Censor-in-chief “worked in the palace and secretly reported to the emperor the illegalities of officials.” That is, the Palace Aide to the Censor-in-chief actually had the duty and power to supervise all the officials of the central government. They were the aides of the Censor-in-Chief. In the Western Han, the Censor-in-chief became one of the grand councillors, i.e., the prime ministers, who were the leaders of the external government. The Palace Aide to the Censor-in-chief became the leader of the censors in the Censorate, still in the inner central government. Liu’s annotation also cited from the Hanshu, stating:                                                         13 Sima Biao, Xu Hanzhi, chapter 26, “Officials,” 3599.  101 Palace Aide to the Censor-in-chief, official of the Qin dynasty, ranked at 1000 bushels, located in the Orchid Pavilion within the palace, supervising Regional Inspectors in the external government, heading Attendant Censors in the inner central government, and overseeing the various officials. Ĉ‹〔È〕v。¡ŰȤŠĨ ǤǑʼn¬ɹɸd‹Oʥ?Ĉ‹ǦÒƵÔ14 Liu Zhao’s annotation of the Xu Hanzhi citing from the Hanyi reads: [The position of] Palace Aide to the Censor-in-chief was previously taken by officials ranked at 2000 bushels or selected from the highest rank of Attendant Censors. They worked in the inner central government, were seated separately in the morning court meeting, headed the Orchid Pavilion, supervised Regional Inspectors, and oversaw all the hundreds of officials. When they went out [to work in the local government], they ranked at 2000 bushels. į)v。Ɠ!ġɰ?Ĉ‹ʲǠ£ěŏŋƛ¢OĨȤŠǀɊäd‹ǦÒƵÔ[Ɠ)v。15 The privilege to sit separately shows that the Palace Aide to the Censor-in-chief had special power and authority in the emperor’s eyes and all the officials. It was recorded in Xuan Bing’s biography in the Hou Hanshu that  Emperor Guangwu Kū specially issued a decree and permitted the Palace Aide to the Censor-in-chief, Metropolitan Commandant (Sili xiaowei ʘř×) and Director of the Imperial Secretariat (Shangshu ling Úʼn4) to sit separately on special mats, which were                                                         14 Fan Ye, Hou Hanshu, juan 27, “Xuan Bing’s Biography,” 927.  15 Sima Biao, Xu Hanzhi, 26, “Officials”, 3600.  102 praised as the “the three who sat alone” (san duzuo ƛ¢).16 In that time, the Palace Aide to the Censor-in-chief actually carried out the duty of the Censor-in-chief. They enjoyed a very crucial position and high status. Therefore, He Zhuo was right to consider them vice prime ministers.  The core task of the Palace Aide to the Censor-in-chief was to inspect and supervise all the officials. Once reassigned to the jurisdiction of the Chamberlain for the Palace Revenues, they could directly serve the emperor, and hence the grand councillors lost this part of their power. He Zhuo said, “the grand councillors could not ask about the personnel administration around the emperor. The grand councillors’ power was weakened and the old order was broken.” He pointed out the transformation of the duties and powers of the Palace Aide to the Censor-in-chief and grand councillors. He Zhuo always paid close attention to the transformation of the duties and powers of all the officials, the evolution of political institutions and political thought. His comments inspired many later scholars. For instance, in juan 46 of the Hou Hanshu, Guo Gong’s biography says: [Guo] Gong transmitted his father’s teachings when he was young. There were always hundreds of people learning from him.  ɢÙFƔŞɍħĆȧìıƵ0 On the margin, He Zhuo writes: Codes and laws were also taught, so that the governance by officials was very good in the Han dynasty. When the Imperial Legal Test was abolished, this custom also declined. ă.ɍħįƉ!‘źǥȌĿż!ǐǶũʩȬǂ                                                         16 See Fan Ye, Hou Hanshu, juan 27, “Xuan Bing’s Biography,” 927.  103 This touches on the evolution of the legal system and intellectual history. In the Eastern Han, although Confucianism was claimed to be the official ideology, legal institutions were the foundation of the administration. The study of laws was independent from that of Confucianism and was very popular at that time. In this sense, the influence of Legalism and of Confucianism met as equals. However, the transmission and teaching of laws and Legalist thought has long been neglected in our scholarship because of the dominance of Confucianism in the later period. By stating that “codes and laws were also taught,” He Zhuo reminds us to pay attention to the transmission of laws and the Legalist thought. In the same juan, the main text of the Hou Hanshu reads: “[Guo Gong’s] father, [Guo] Hong, learned the Xiao Du lü (Law of the Little Du)” (ƔüǻØŕă). On the margin of this page, He Zhuo wrote three characters, “Xiao Du lü”, reminding the reader to pay attention to this book and the teachings of this school. Guo Gong’s father studied as a Legalist. He learned laws and taught laws. Guo Gong learned from his father and handed down his legal knowledge to later generations. This meant that the teaching of Legalism was transmitted from teachers to students and/or within family clans in the Han dynasty, just like the transmission of Confucianism. The main text of Chen Xian’s ʓ— biography in the same juan states:  Later, [Wang] Mang summoned [Chen] Xian for appointment a second time, but he refused on the pretext of serious illness. He then collected the laws, codes and other documents in his family’s possession, and hid them all in the walls. VĄȘĉĊ—ɫǖƳǣķł ĬIJVÎă4ʼnijƶ¨ȡ! He Zhuo’s marginal comment regarding this reads: Fu Sheng hid the Classics; Chen Xian hid the laws. 8Ʃȡǭʓ—ȡă  104 Fu Sheng 8p, known as Fu Sheng 8Ʃ (Master Fu), was a prominent Confucian scholar and classicist in the Western Han period. He was well known for hiding the Shangshu Úʼn (Book of Documents) within the walls to prevent its being burned by the first emperor of the Qin dynasty.17 Since Chen Xiao hid some legal texts within the walls, He Zhuo made a couplet, equating Fu Sheng and Chen Xian, which exhibits his grasp of the relationship between Legalism and Confucianism.  The Eastern Han empire was said to have been governed by Confucian doctrines, but the governing of officials at that time was very harsh, a state of affairs that violated those doctrines. This contradiction, Chen Suzhen ʓȣʈ pointed out, was due to the fact that the central government had to use cruel officials (kuli ɾ‘, also translated as “merciless judges”), skilled Legalists, to repress the power of wealthy families that was empire-wide, even though the empire claimed to follow the doctrines of Confucius 18  Chen’s explanation partly resolved this contradiction. Another important factor was the continuous transmission of Legalism and its education system—one could learn within his family and/or from other masters. Zhong Hao’s biography in the Hou Hanshu recorded that, much like Guo Gong and his family, Zhong Hao’s “was a well-known family in their commandery, specializing in criminal laws for generations” (Ɠɷț»™^ă).19 There must have been a great many families like this in the Eastern Han period. The Legalist school still possessed a significant social base; Confucianism had not yet dominated intellectual trends.                                                          17 See Sima Qian, Shiji, juan 121, 3124-3126. 18 Chen Suzhen ʓȣʈ, Chunqiu yu “Handao”: Liang Han zhengzhi yu zhengzhi wenhua yanjiu 	ŀǏȏ“Ɖɭ”QƉĮźȏĮźijrDžǙ (Beijing: Zhonghua shuju, 2011), 486-513. 19 Fan Ye, Hou Hanshu, juan 62, 2064.  105 At that time, Confucian teachings were gradually permeating the whole of society from high to low. He Zhuo made note of this too. In Dong Xuan’s ȜÉ biography in the Hou Hanshu (juan 77) , his marginal comment states: Dong Xuan, He Bing, and people of their like should not be listed in the “Biographies of Cruel Officials.” Li Zhang only killed criminals to excess when he was governor of Qiansheng prefecture. It wasn’t that they killed so many people out of a capricious whim. Fan [Ye] began this juan with these three persons perhaps because Emperor Guangwu made the governance of officials very harsh, so that the officials went further and became too cruel. This revealed that so-called good governance in this resurgent time [the Guangwu reign, 25-57 CE] was not perfect. However, my humble opinion is that there need not be “Biographies of Cruel Officials” [in the Hou Hanshu]. ȜÉ;!ƀƯ_!	ɾ‘Ŕǝ.Ė¡v$ŃɁĴƹɗɬƌʠ6›đ­ɁƇ'ȀȖʬƒũ0Ȟ5ùū‘(eƄ¶ƨfĎōN*ɾǿĿȐ!Ǹ¢łŐƺȁƏǛɋŖ/	ɾ‘F‰5ǜ Here, He Zhuo wrote that “my humble opinion is that there need not be ‘Biographies of Cruel Officials’ [in the Hou Hanshu],” arguing that Dong Xuan, He Bing and Li Zhang were not truly cruel officials who “killed too many people following their own capricious moods.” He must have made a comparison between the Western Han and the Eastern Han and found that in the Eastern Han, not only did the number of cruel officials decline dramatically, but “cruel officials” also became less cruel. “Cruel officials” were being “Confucianized” (rusheng hua HƩr), to employ a term proposed by a modern scholar, Yan Buke ʏŪL. Yan made this argument more clear by comparing the “Biographies of Cruel Officials” in the Shiji and Hanshu with those in the Hou  106 Hanshu. He found that there were fifteen cruel officials recorded in the Shiji and Hanshu, but only seven in the Hou Hanshu. Among these seven “cruel officials,” Dong Xuan was well-known for his upright and outspoken personality. When put into prison, he read and recited [Confucian books] from morning to night. He also abided by Confucian moral tenets. Another “cruel official” in the Hou Hanshu, Li Zhang, studied the Yan ʧ annotations to the Spring and Autumn Annals and was sufficiently well-versed to teach them. Huang Chang ʷľ had also once studied the Confucian Classics. Wang Ji ƠŽ loved to  read [Confucian] texts and their annotations. Only Fan Ye šņ, Zhou Ji •Ǯ and Yang Qiuʕƣ were inclined to the learning of Shen Buhai ƬÍ and Han Feiʢʠ, two Legalists in the Warring States period. Yet Shen Buhai and Han Fei joined hands with the Confucian scholars to fight against the eunuchs. In this sense, Yan Buke claimed, the cruel officials were Confucianized, or we can say that the Confucians were cruel-officialized (kuli hua ɾ‘r). That is, the cruel officials learned the Confucian teachings and were gradually influenced by Confucian doctrines; meanwhile, the Confucian scholars started to master more administrative skills (such as law) and threw themselves into practical affairs. Therefore, Yan Buke agrees with He Zhuo’s view that “there need not be ‘Biographies of Cruel Officials’ [in the Hou Hanshu].”20 This shows that He Zhuo’s view was rather novel and insightful for his time. This is a critique of the historiography of the Hou Hanshu. Marginalia of this kind were not rare. In these historiographical comments, He Zhuo usually criticized the collection, sorting, and trimming of various historical sources, and the writer’s writing skills, views, and attitudes concerning historical events and figures.                                                          20 Yan Buke ʏŪL, Shidafu zhengzhi yanshengshi gao ©¯²ĮźȩƩ‹Ǘ (Beijing: Beijing daxue chubanshe, 1996), 450-451.  107 In addition, He also paid much attention to previous scholars’ annotations of the Hou Hanshu, correcting them or making compliments. For instance, He corrected errors regarding the position of the Directorate of the Palace Library (Mishu jian Ǒʼn○), which was a very important position in the Han court, heading the Department of the Palace Library (Mishu sheng Ǒʼnƿ), which was in charge of collecting court documents and compiling official histories. But there was no Directorate of the Palace Library listed in the “Treatise of Officials” (Baiguan zhi ƵÈď) of the Xu Hanzhi. However, it was recorded in the “Annals of Emperor Huan” (Huandi ji Śëǧ) that the Directorate of the Palace Library was initially set up in the second year of the Yanxi øƑ (158-166) era. Liu Zhao’s annotation states: “Directorate of the Palace Library, one person, ranked at 600 bushels.” In the “Treatise of Officials Part II,” under Erudite of the Chamberlain for Ceremonials (Taichang boshi ±ì{©), Liu’s annotation states: “According to the ‘Annals of Emperor Huan,’ the Directorate of the Palace Library was set up in the second year of the Yanxi era.” This means that Liu’s annotator inclined to the belief that the Directorate of the Palace Library was subordinate to the Chamberlain for Ceremonials (Taichang ±ì). This, according to He Zhuo, is incorrect. In the “Treatise of Officials Part III” of the Xu Hanzhi (juan 26), under Chamberlain for the Palace Revenues (Shaofu Ùó), there was the entry “Head of the Orchid Pavilion (Lantai lingshi ȤŠ4=), ranked at 600 bushels.” He Zhuo’s marginal comment here reads: In the “Table of noble ranks and government offices” of the Hanshu (juan 19), the Palace Aide to the Censor-in-chief worked within the palace and the Orchid Pavilion was in charge of books and secret documents. It was recorded in the “Annals of Emperor Huan” [of the Hou Hanshu] that the Directorate of the Palace Library was initially set up in the second year of  108 the Yanxi era. The annotation cited the Hanguan yi states “Directorate of the Palace Library, one person, ranked at 600 bushels.” This official position should be listed under the Palace Aide to the Censor-in-chief and was subordinate to the Chamberlain for the Palace Revenues [in the Hou Hanshu]. It was omitted in this treatise. Liu’s annotation added it under the Erudite of the Chamberlain for Ceremonials. This is incorrect. 	·ʼnƵÈSȫĈ‹¡ŰȤȎĨ ǤǑʼn	ŚǧøƑ)ï`ǴǑʼn○ÈŽû	ƉÈGǑʼn○0〕TƵ。VÈƯ_ßķÙóȀũďɱ!iŲŽȯ*±ì{©!fʠ' By citing the Hou Hanshu, He Zhuo stated that the Directorate of the Palace Library should be listed under the Palace Aide to the Censor-in-chief and subordinate to the Chamberlain for the Palace Revenues, and the annotation supplementing it under the Erudite of the Chamberlain for Ceremonials was incorrect. He Zhuo not only pointed out the hasty omission in the main text of the Hou Hanshu, but also spelled out how the annotator had made a mistake. In the Wei dynasty (221-265), the Directorate of the Palace Library was initially subordinate to the Chamberlain for the Palace Revenues. This was supposed to follow the old custom of the Eastern Han. But the Tongdian ɩW (Comprehensive institutions) records that the Directorate of the Palace Library was subordinate to the Chamberlain for Ceremonials, following the record of Liu’s annotation in the Hou Hanshu. 21 There is still no final conclusion on the subordination of the Directorate of the Palace Library in the Eastern Han, but He Zhuo’s comments remind us that there may be a problem in the historical record of the Hou Hanshu and its annotation.                                                         21 Du You ŕ:, Tongdian ɩW (Beijing: Zhonghua shuju, 1992), 732-733.  109 In addition to the Hou Hanshu, He Zhuo also read and wrote a great deal of marginalia on the Shiji, Hanshu, Sanguozhi, Nanshi z‹ (History of the Southern Dynasties), Beishi s‹ (History of the Northern Dynasties), Wudaishi +3‹ (History of the Five Dynasties), Shitong ‹ɩ (Conspectus of historiography), Huayangguo zhi șʕŸď (Records of the lands south of Mt. Hua), Zhongwu jiwen“ǧȃ (Collected stories from the Wu Region), and Shuijing zhu ųǭŽ (Water classic with annotations). His marginalia on these histories, historiographical works, and historical-geographical works have a great deal of insightful comments. He Zhuo was also an expert in the appreciation and authentication for the rubbings of ancient stone inscriptions and calligraphic writings on silk and paper (beitie džê). His comments on these books, such as the Gengzi xiaoxia ji ò¾ʅ«Ⱥ (Records written in the summer of the gengzi year), and the Fashu yaolu żʼnȲʆ (Essential compendium on calligraphy) also have considerable academic and historical value. His marginalia on Tang and Song poetry were praised because his research was based not only on precise textual criticism, but also on precise comparisons between poetry and history, thus influential when it came to the reading and study of poetry, the history of Chinese literary criticism, and the history of reading. In short, although He Zhuo sometimes derived his practice from the approaches of annotating eight-legged essays, his marginalia were not the “study of the papers’ tail” kind. He contributed to the study of history, literary criticism, philology, and calligraphy, among other subjects.   A Pioneer of Textual Criticism In He Zhuo’s marginalia on the Hou Hanshu, there are also thousands of items concerned with textual criticism. At the end of juan 90, He’s colophon states:  110 When I read this book for the first time, I disliked its numerous mistakes [within the text]. After reading Liu’s Correction [of the Two Han Histories], I finally realized that good editions of this book were already rare in the Northern Song. This is because predecessors did not pay as much attention to it as to Ban [Gu’s] Hanshu. `ɓũʼn½VȻɏƓ­ƒȶiŲ	]ɄɊŜ 、¡sÆ}Dz™őǯ·0ɿ!·Ƣʼnį' He Zhuo pointed out that people previously did not pay enough attention to the Hou Hanshu, resulting in editions that had numerous mistakes in the text. Therefore, he made a great effort to correct these mistakes. In the roughly 3700 pieces of marginalia he appended to Hou Hanshu, more than 1600 of them were concerned with textual criticism.  He Zhuo was a book collector himself. He also had good relationships with the most famous book collectors in the early Qing period, such as the Mao ű family of Yushan Ȧà, the Ye Țfamily of Kunshan Ľà, Xu Qianxue, Qian Zeng, and so forth. He also worked as an editor in the imperial library for about twenty years. The relationships and activities supplied many opportunities for him to get access to many books, especially rare Song and Yuan editions. According to He’s colophons in the Hou Hanshu held at Peking University Library, he used several incomplete Song editions (can Songben ŮÆő) to collate the text of the Hou Hanshu. None of them are still extant today, so the variants collected by He Zhuo are the only evidence we have with which to recover these Song editions. He also employed the Yijing Hall (Yijing tang ǭ¤) edition and the Masha ʶŹ edition carved in the Southern Song, a Yuan edition, more than three Ming editions, and several “recent editions” (jinke ɨe) as he read and collated the Hou Hanshu. Sometimes when the text seemed problematic but there was no variant character from one of these  111 other editions to select, he would write down his opinions based on his own understanding and research on the text. To give one example of He’s collation: the biography of Emperor Huan in juan 7 of the Hou Hanshu says:  On the jiazi day of the first month in the spring of the first year of the Heping reign, [the emperor] granted amnesty to the [criminals of the] whole realm. The name of the reign was changed to Heping [peace]. On the jihai day [of this month], the imperial decree says…The second month… –îIïŀŨŌƫ¾¯ɚ°ĭI–îæ-ȽŇ……)Ō…… In the traditional Chinese calendrical system, a cycle of sixty terms was employed to reckon dates, which was known as the “sexagenary cycle” (ganzhi íī, lit. “Stems-and-Branches”). These sixty terms in the Eastern Han are listed in Table 3-1.  01.ƫ¾ jiazi 02.% yichou 03.Ï bingyin 04.| dingmao 05.ĝɧ wuchen 06.æè jisi 07.òw gengwu 08.ɥŐ xinwei 09.ªƬ renshen 10.ƴɼ guiyou 11.ƫĞ jiaxu 12.%- yihao 13.¾ bingzi 14. dingchou 15.ĝÏ wuyin 16.æ| jimao 17.òɧ gengchen 18 ɥè xinsi 19.ªw renwu 20.ƴŐ guiwei 21.ƫƬ jiashen 22.%ɼ yiyou 23.Ğ bingxu 24.- dinghai 25.ĝ¾ wuzi 26.æ jichou 27.òÏ gengyin 28 ɥ| xinmao 29.ªɧ renchen 30.ƴè guisi 31.ƫw jiawu 32.%Ő yiwei 33.Ƭ bingshen 34.ɼ dingyou 35.ĝĞ wuxu 36.æ- jihai 37.ò¾ gengzi 38 ɥ xinchou 39.ªÏ renyin 40.ƴ| guimao 41.ƫɧ jiachen 42.%è yisi 43.w bingwu 44.Ő dingwei 45.ĝƬ wushen 46.æɼ jiyou 47.òĞ gengxu 48 ɥ- xinhai 49.ª¾ renzi 50.ƴ guichou 51.ƫÏ jiayin 52.%| yimao 53.ɧ bingchen 54.è dingsi 55.ĝw wuwu 56.æŐ jiwei 57.òƬ gengshen 58.ɥɼ xinyou 59.ªĞ renxu 60.ƴ- guihai Table 3-1  Table of the sexagenary cycle   112 According to the sexagenary cycle, if the jiazi (No.01) day was in the first month, the jihai day (No. 36) cannot be in this month. He Zhuo found this problem in the text, and wrote in the margin of this page: If the amnesty was on the jiazi day, the jihai day, when political power was returned to the emperor, should have been in the second month. I suspect the date was incorrect. Ȕ5ƫ¾ɚfæ-ŭĮƯ¡)ŌƱĻōɄ The word “suspect” (yi Ʊ) reveals that He had not yet seen any variant character that would explain how either the jiazi day or the jihai day could be a correct record. He discovered this tiny mistake and made this judgment based on reasoning. After this comment, there was another sentence following a small circle: / In the Song edition, [“jihai” was written as] “jichou.” Æőæ This indicated that He later found a variant in a Song edition, in which “jihai” was written as “jichou” (No. 26). This confirmed his speculation that one of the dates was written incorrectly. Based on this solid evidence, He changed the character “hai” of the main text into “chou” in red ink.  The Hou Hanshu jiaokan ji ĄƉʼnřnȺ (Collation notes on the Hou Hanshu), composed by the great publisher and textual scholar Zhang Yuanji þIƋ (1867-1959), collects variants from almost all editions of the Hou Hanshu from the Song to the late Qing period. It was recorded in this book that the text here was “jichou” in the Shaoxing ǩȐ edition carved in the Southern Song period, and was “jihai” in almost all the Yuan and Ming editions. Zhang stated, “It was  113 written ‘jiazi’ above, so that there shouldn’t be a ‘jihai’ [in this month]. The Song edition is right” (ȸƫ¾Ĝōæ-Æőł).22 Zhang and He Zhuo held the same opinion, and their opinions should be taken into account when editing the Hou Hanshu in modern times. But the most widely-read and influential, punctuated and collated edition of the Hou Hanshu, the Zhonghua shuju șʼnÝ edition, has not adopted this suggestion. Instead, it employed a variant from another book, the Zizhi tongjian ɖźɩʋ (Comprehensive mirror for aid in government).23 This variant was discovered by a late Qing scholar Huang Shan ʷà while he was editing the Hou Hanshu jijie ĄƉʼnʙȷ (Collected annotation to the Hou Hanshu), composed by his teacher Wang Xianqian ƠJɌ (1842-1917). Huang put it: Yuan Hong’s Records [of the Later Han] has “jichou,” while the Tongjian has “yichou.” We should regard the text of the Tongjian as correct. ȭ	ǧ<æ	ɩʋ<%Ư5	ɩʋƓŨ24 Wang Xianqian used the text of a Ming edition, the Mao edition, to compose his collected annotations. He Zhuo wrote his marginalia on a copy of the same edition. The main text here has “jihai,” which is certainly wrong. Huang Shan found that it was “jichou” in Yuan Hong’s ȭÇ (courtesy name Yanbo “{, 328-376) Houhanji ĄƉǧ (Records of the Later Han) and that it was “yichou” in the Zizhi tongjian, and claimed the text in the Zizhi tongjian was right without any                                                         22 Zhang Yuanji þIƋ, Hou Hanshu jiankan ji ĄƉʼnřnȺ (Beijing: Zhonghua shuju, 1999), 41. 23 Hou Hanshu, juan 7, 295, 322. 24 Wang Xianqian ƠJɌ, Hou Hanshu jijie ĄƉʼnʙȷ (woodblock edition carved in the Xushou tang ȥ†¤ of the Wang family in 1915), juan 7, 1b.  114 explanation. The Houhanji and Zizhi tongjian are both other books (ta shu 2ʼn); they hold the same status when used to collate the Hou Hanshu. Therefore, in this situation, it is difficult to tell which one is right. However, the Shaoxing edition is another edition of the Hou Hanshu itself, and its text thus enjoyed more importance in collating the Hou Hanshu. Using the Shaoxing edition as its master copy, the text in the Zhonghua edition was “jichou” in the first place. Collators should keep this text and state that there are variants in other books, rather than change the text according to other books. He Zhuo’s marginalia here show that he was a very careful reader; his way of dealing with these variants demonstrated that he was also a meticulously scholar and collator. Unlike literary criticism and historical comments, textual criticism requires scholars to be more precise that their arguments must be based on solid and reliable textual evidence. Textual criticism was considered one of the fundamental sub-disciplines of evidential research, which characterized the scholarship of the Qing dynasty. He Zhuo was the first scholar who systematically undertook textual criticism of hundreds of titles. He led the discipline to flourish, so much so that he has been praised as the founder of textual criticism by modern scholars.25  The traditional narrative on Ming scholarship, especially classical studies and history, was that Ming scholars’ teachings were shallow and that almost all the scholars and students in that dynasty focused only on the civil service examinations, such that they spent all their energy discussing literary features of the text without first authenticating it, investigating the historical events therein, or having a deeper comprehension of the meanings of ancient Classics, histories and philosophies. Qing scholars, on the contrary, mostly concentrated on evidential research, a                                                         25 See Liang Qichao, Intellectual Trends in the Ch’ing Period, 69; Liang Qichao, Zhongguo jin sanbainian xueshushi, 290; Wang Shaoying, “Ruan shi chongke Shisanjingzhushu kao,” 54; Elman, From Philosophy to Philology, 211-214.  115 positivist approach to scholarship buttressed with recourse to reliable and authentic texts. What were the causes of this transition in scholarship between the Ming and Qing? How did it come to pass? Who contributed to this transition and in what ways? These questions are still not fully answered. 26 However, He Zhuo himself experienced this transition, which makes him a good model with which to study the evolution of thought and scholarship from the Ming to the Qing. When he was young, He Zhuo, like most students at that time, hoped to pass the civil service examinations and become an official. He devoted himself to learning how to compose eight-legged essays. Although he never passed the examinations, he became an expert in selecting and commenting on the eight-legged essays, and, ironically, helped a great many students pass the exam. Because of this, he became well known and was eventually given the jinshi degree by the Kangxi Emperor and appointed to a position in the palace.27 The Qingbai leichao ƅǔʨʁ (Qing anecdotes compiled by category) records: When Yan Qianqiu, name Ruoqu, got to know He Yimen, He was twenty-four sui. They talked about the exam writings every day.                                                         26 Lin QingzhangŗĚ… and many modern scholars have already started to question conventional opinions on Ming and Qing scholarship. Lin argued that a few scholars had already started to do evidential research since the late Ming, and that Qing scholars followed these late-Ming scholars’ steps and developed the evidential research to a new stage. See Lin Qingzhang, Mingdai jingxue yanjiu lunji Ŀ3ǭÂDžǙɇʙ (Taipei: Wenshizhe, 1994). 27 See the “Appendix” of He Zhuo, Yimen xiansheng ji; Shen Tong, “Hanlinyuan bianxiu zen shiduxueshi Yimen He xiansheng xingzhuang”, in He Zhuo, Yimen dushuji, 1275-1277; Quan Zuwang, “Hanlinyuan bianxiu zeng xueshi Changzhou He gong mubeiming”, in Yimen dushuji, 1278-1280; Qing shi liezhuan, in Qingdai zhuanji congkan, vol. 104, juan 71.  116 ʏƊ7ȔƦ`,;ǹʎ;ï)uœŬĻȏɑɇŃij28 This shows that when he was twenty-four sui, he was still devoted to the civil examinations, but by at least the thirty-first year of the Kangxi reign (1692), when He was thirty-one sui, he had started to collate books. This can be corroborated by all the colophons collected in the “Yimen tiba” ǹʎʦɝ (Master Yimen’s colophons) in the Yimen xiansheng ji.  After reading and collating one book, He Zhuo usually wrote a short colophon stating when, where and based on which edition(s) he had read and collated that book. The colophons in the “Yimen tiba” were all composed after the thirty-first year of the Kangxi reign; more than half were written during the fortieth to fiftieth years of the Kangxi reign.29 In addition, in the “Preface” to the Xingyuan ji Ȩɮʙ (Collected works of Lu Shen ʔƄ [1477-1544]), He Zhuo wrote:                                                         28 Xu Ke ąơ, Qing bai leichao ƅǔʨʁ (Beijing: Zhonghua shuju, 1984), 683. “Shiwen” Ńij, literally meaning “modern essay,” refers to the eight-legged essay (bagu wen Rȇij), which was a style of essay that examination candidates wrote in order to pass the civil service examinations during the Ming and Qing dynasties. Other alternative names include zhiyi cȢ, jingyi ǭǹ, and Sishu wen œʼnij. See Benjamin A. Elman, Civil Examinations and Meritocracy in Late Imperial China, 46-92. 29 He Zhuo, Yimen xiansheng ji, juan 9. Only the colophon of one book, the Ouyang Xingzhou wenji ŦʕȨ•ijʙ (Collected works of Ouyang Zhan Ŧʕɀ [courtesy name Xingzhou, 758-801]), records that it was written in the “yichou” % year of the Kangxi reign, that is, the 24th year of the Kangxi reign. But in the Tieqintongjianlou cangshu mulu ʊƥʃjŢȡʼnƼʆ and the Bisonglou cangshuzhi, ƷÆŢȡʼnď, the colophon is recorded being written in the “jichou” æ year of the Kangxi reign (48th year of the Kangxi reign). In the 24th year, He was occupied by composing eight-legged essays. It is more likely that he collated this book and wrote the colophon in the 48th year of the Kangxi reign. See Qu Yong ǁʉ, Tieqintongjianlou cangshu mulu (Woodblock edition carved by the Qu ǁ family in the Xianfeng reign [1850-1861]), juan 19, 33a; Lu Xinyuan ʔčƆ, Bisonglou cangshuzhi (Woodblock edition carved in 1882), juan 69, 14a-15a.  117 Early on, I followed the current fashion and focused on classical essays for the exams. … In the winter of the yihai year (34th year) of the Kangxi reign [1695], I studied the “Yueji” (Record of Music, one chapter of the Liji ǍȺ [Record of rites]) repeatedly. I closed the book and sighed when I read about the “deceitful sounds” and the “orthodox sounds,” and their “responses with human minds.” I regretted that I used to play the lewd music of the Zheng State and the Wei State in a Confucian house. I thought I was making rapid progress every day but I actually had forgotten the Way. Realising this, I abandoned all the various explanations. Instead, I obtained the sages’ Classics to read. I concentrated on the exegesis of the words and understood thoroughly the logical arrangement of the text. By so doing, I apprehended the general meaning of the Classics, and then the essence of the Way generally came out. ¹”ʗ@ƍǭǹ%-öƐuœïYÂǻĉ	ŠȺȍķ¼ȄŨȄĘĜ!ʖ÷~ȀťĔ’!ýɻȪķ¿Ɩ'Ļ´ʮȀɭfĐ ƺއɅň…Ȃ0ɘ0!ǭɓ!„ȳ"ȹįŋɩ"ŜƤćV¯ʱɭőƃ[30 This indicates that He Zhuo started to concentrate on classical studies in the thirty-fourth year of the Kangxi reign (1695). In short, his scholarly orientation shifted from study of eight-legged essays to classical studies during 1692 to 1695. Departing from the “current fashion” that focused on composing beautiful essays for the civil exams, this new approach, classical studies, aimed to comprehend the Way based on repeatedly and carefully reading the Confucian Classics. The reading and comprehension of the Classics necessitated precise explanation of the words and                                                         30 He Zhuo, Yimenxiansheng ji, juan 1, 6ab.  118 sentences, which need to be established on reliable and accurate texts. All of this caused the development of philology, bibliography, textual criticism, and many other evidential disciplines.  He Zhuo’s scholarly transformation relied on several conditions.  First, he became rather well known when he was young. He was granted the jinshi degree by the emperor and received a not particularly powerful but relatively stable position in the palace publishing office and imperial court. That is, He Zhuo was a professional scholar sponsored by the imperial family, whose main work was to read, do research, and collate and compile books. Second, He Zhuo had grown up in Suzhou, one of the economic and academic centers of Ming and Qing China. Some of the most famous book collectors and publishers gathered in that region. Therefore, it was easy for him to gain access to any kind of book, most importantly rare Song and Yuan editions. In addition, having worked in the palace for decades, He gained access to the imperial collection, which contained many rare editions. Third, according to modern scholars there was a “publishing boom” in the late Ming period. From this time on, print became the dominant mode of the circulation of texts, exceeding manuscripts.31 The dramatic increase in quantity of books and the proliferation of editions was one of the most important reasons for the development of textual criticism and the flourishing of marginalia. Most of He Zhuo’s marginalia were written on copies of Ming editions; the marginalia I discuss in this study are mostly written on editions printed after the late Ming. Moreover, more editions created more inconsistencies in a given text, which created new tasks for scholars: one needed to choose the “authentic” text before reading. Unfortunately, it was not always easy to determine an authentic text between different editions. Therefore, textual criticism weighed more and more in scholarship.  At the same time, disciplines related to textual criticism,                                                         31 See Ōki Yasushi, “Minmatsu Kōnan ni okeru shuppan bunka no kenkyū;” Kai-wing Chow, Publishing, Culture and Power in Early Modern China; Joseph McDermott, A Social History of the Chinese Book.  119 such as paleography, bibliography, phonology, and phonetics, became more and more important. Finally, with the development of commerce and transportation, the circulation of books and personnel became more and more convenient.32 Scholars could easily communicate with each other. During the Qing dynasty, scholars created an academic network swapping opinions and texts. This will be discussed in the next two chapters.   How Was a Scholar Stigmatized? He Zhuo never stopped collecting, reading, and collating books, or writing marginalia. As the Kangxi Emperor remarked, He was a real reading seed. Besides, based on his contributions to textual criticism and bibliographic studies, he can be regarded as one of the founders of evidential research. However, his moral character and scholarly achievement were long disparaged by some Qing scholars for political reasons; thus he has long been underappreciated—almost neglected—in modern scholarship. In the following section, I will show how He Zhuo was stigmatized. It is recorded in the Qingshi liezhuan ƅ‹_F (Biographies of the Qing History) that in the 42nd year of the Kangxi reign, He Zhuo was given the jinshi degree by the Kangxi Emperor, and then appointed Hanlin Bachelor (Hanlinyuan shujishi ǼŗʒõŽ©), working in the Southern Study (Nan shufang, zʼnĢ). Later he became the tutor of Yinsi Ȉnj (1681-1726), the eighth son of the emperor, and established a very intimate relationship with him,33 which is the root cause of his wrongful disparagement.                                                          32 See Cynthia J. Brokaw and Kai-wing Chow, eds., Printing and Book Culture in Late Imperial China. 33 See Qingshi liezhuan ƅ‹_F, juan 71, 27a, in Zhou Junfu ed., Qingdai zhuanji congkan, vol. 104, 838.  120 In the later period of Kangxi Emperor’s reign, his sons fought fiercely for the succession to their father’s throne. In the end, the fourth prince, Yinzhen ȈNj (1678-1735), succeeded and ascended to the throne in 1723, becoming the Yongzheng Emperor. In the power struggle, Yinsi was favored by most officials in the imperial court. He had a very large support network that included many top-ranking officials. Four years after the Yongzheng Emperor ascended the throne, when he had consolidated his rule, he started to slaughter his brothers who had fought against him for the throne, and this also brought disaster to the officials who had supported them. At this time, He Zhuo had been dead for years, but he was not spared. Jiang Liangji Ƞȓʰ wrote in the Donghua lu Ŗșʆ (Chronicle of the Qing Dynasty): In the third month [of the 4th year of the Yongzheng reign], the Great Academician and the Nine Chief Ministers presented a memorial to the emperor, saying: “Qian Mingshi the expositor-in-waiting composed poems and sent them to Nian Gengyao to glorify his accomplishments. For this extreme sycophancy, he should be dismissed and punished.” The emperor’s decree read: “In the past, Qian Mingshi, He Zhuo, and Chen Menglei all had literary fame. But they were dishonorable in conduct and their comportment was despicable.” ʚŨœïŌ¯Â©&Ǣ³ʪ?ɍC!ʇ<ȿĦəïǺ¥ǖlʣċEşɎɈĜʡȅźdzćļ’>·ʇ;Ǝʓ®ʜǢƶʤōij‰ĕȨŧǞǜɡxŵ34 Wang Xianqian ƠJɌ wrote in his Donghua lu:                                                         34 Jiang Liangji Ƞȓʰ, Donghua lu Ŗșʆ (Beijing: Zhonghua shuju, 1980), juan 27, 451.  121 On the jiazi day [of the sixth month of the 4th year of the Yongzheng reign], Chong’an the Imperial Prince Kang and other princes, Beile Princes, Beizi Princes, dukes, and the Manchu and Han officials civil and military, all presented memorials to the emperor impeaching Akina’s [lit., pig, denoting Yinsi] forty crimes. …Most of the time, [Yinsi] was controlled by his wife. One day when he was talking with He Zhuo, he suffered from her to laugh out loud outside the door with no compunction. He also brought up He Zhuo’s little daughter in his house and treated her as his own daughter. … After being stripped of the title of Beile Prince, he secretly used silver and horses to draw Rufu and others into his clique, and secretly conspired with the eunuch Li Yu and managed to dismiss Emuketuo who concurrently served in the Imperial Dining Room. He also aligned with He Zhuo who worked in the Hanlin Academy. They won undeserved renown and harboured disloyal thoughts. ʚŨœïTŌƫ¾öȵƠáÅŅɊƠɕmɕ¾SƈƉijū¯ȋǢSɑ³ʑVɴdzƙœuŤ……ȈnjîĻ†cķ¸Ļȏ;ƎUɆ6Ȇ7¸ʎ¬¯ǟ、ƿɲ‚Ö;Ǝ!ñµǎʫÄ5ƍçµ……ĺʡɕm!Ąń5ʂʭǢƗȲǪŶ》Ǣ0Nʹ‚Џ±○ŔƟĩʡȊĢȨɛ!€ǘLĤ‚ȏǼŗʒ;ƎžǪtʹƹ…ɒƊȝƮč35 This indicates that Yongzheng Emperor’s court spared no effort to damage Yinsi’s reputation, and He Zhuo was also drawn into the political brawl. One prominent modern scholar, Deng Zhicheng ɺ!Ƀ (1887-1960), pointed out that He Zhuo was lucky that he died earlier. Otherwise, as one of Yinsi’s trusted aids and a skilled counsellor, He Zhuo might not have his head after the fourth                                                         35 Wang Xianqian ƠJɌ, Donghua lu Ŗșʆ, “Yongzheng ba” ʚŨR, 58a-59a, in Xuxiu Siku quanshu, vol. 173, 174.  122 year of the Yongzheng reign.36 But, unfortunately, his reputation was ruined and a shadow was cast over his scholarly contributions. These decrees of the Yongzheng Emperor spread widely and were influential. Some scholars blindly—or intentionally—followed these descriptions and despised He Zhuo without carefully reading his works. For example, Chen Kangqi ʓö《 (1840-1890) stated: He Zhuo knew nothing about gratitude and was dishonourable in conduct, which can be seen from the decrees of the two reigns [i.e., the Kangxi reign and the Yongzheng reign]. People within the Four Seas please do not be shocked by his empty reputation. ;Ǝ!ɐĒǹȨŧǞɫĿȴ!QŏɉļǂƂX0©ðq5ƁȀʝ!37 Chen, like many other scholars from the Qing dynasty, employed an expression similar to those used in Yongzheng’s decrees. His knowledge of He Zhuo came from Yongzheng’s decrees rather than He Zhuo’s works and marginalia. In this, he claimed that He Zhuo had an “empty reputation” (fuming Ɓ). Yu Zhengxie, for his part, claimed that He Zhuo’s learning “was mostly what deceived the human heart” (Ʃî­ţč!Â).38 This claim was also influenced by Yongzheng’s decrees. None of these were impartial judgements. As a reading seed who preserved, restored and transmitted texts, and contributed much to textual studies, He Zhuo’s works and reputation should not be erased. I am not arguing that he was innocent in the princes’ battle for the succession to the                                                         36 See Deng Zhicheng ɺ!Ƀ, Qingshi jishi chubian ƅȿǧ(`ǰ, juan 3, in Zhou Junfu ed., Qingdai zhuanji congkan, vol. 20, 353. 37 Chen Kangqi ʓö《, Lang qian jiwen erbi ɶƊǧȃ)ǡ (Beijing: Zhonghua shuju, 1997), juan 8, 475. 38 Yu Zhengxie AŨƒ, Guisi cungao, in Congshu jicheng chubian, vol. 430, 431.  123 throne, rather, what I would like to emphasize is that we should focus on scholars’ scholarly practice and achievements while evaluating their thought and scholarly contributions.  What happened to He Zhuo also happened to other Qing scholars. In the following chapters, I will try to “rescue” some of them—many of whom may be unfamiliar to most modern readers—by investigating their scholarly practices, especially those related to reading and collating books, and composing and transcribing marginalia.   124 4 Transcription: Transcriptionists and Relationships Built with Marginalia   “Transcribe,” guolu ̠̀ in Chinese, means to hand-copy a text, and since marginalia were mainly handwritten comments, usually in printed books (as discussed in previous chapters), the production and circulation of them were a hybrid of print culture and manuscript culture. In this particular textual culture, the transcription of marginalia as a scholarly practice occupied a very fundamental position. Records and physical evidence of the transcription of marginalia found prior to the Qing are rare. In the Qing dynasty, however, the transcription of marginalia became a prevalent scholarly practice, and transcription characterized the circulation of marginalia to a considerable extent. This chapter and the next will take the transcription of He Zhuo’s marginalia as an example and examine by whom marginalia were transcribed and under what circumstances, how different texts were read and used by various transcriptionists, and what the meaning and significance this practice had for them.  This chapter will focus on various transcriptionists and the relationships formed based on marginalia. Generally speaking, the enterprise of transcribing marginalia was undertaken not only by professional scribes and scholars, but also by various other agents including book collectors, calligraphers, and book merchants earning a living by means of this scholarly practice. In the case of He Zhuo, his students, many later scholars, and some merchants were the most active transcriptionists of his marginalia. All of them, despite living in different temporal and spatial realms, were connected by He Zhuo’s marginalia and formed a kind of special relationship.    125 Students He Zhuo was well known in his own time and had approximately four hundred students,1 who were the first generation to transcribe his marginalia. Since He Zhuo was a successful scholar and calligrapher, the process of transcription for his students was not only an opportunity to accumulate research materials and academic knowledge, but also a way to practice their calligraphy. In a copy of a Ming woodblock edition of the Zhouli zhushu ·ɛǴȹ (The Rites of Zhou with annotations and sub-annotations) held at the Shanghai Library (call number: XST04967-80), there are a great many of Shen Tong’s ǭľ (courtesy name Guanyun z͟, alternative name Guotang ƴÍ, 1688-1752) marginalia and his transcriptions of his teacher He Zhuo’s marginalia. Most of He and Shen’s marginalia consist of collation notes and evidential research on the Zhouli zhushu. At the end of this book, there is a colophon composed by He Zhuo and transcribed by Shen Tong. It reads: In the bingxu year of the Kangxi reign [1706], I had the opportunity to see a [a copy of the Zhouli zhushu in a] Song edition repaired in the Yuan dynasty. I made one rough pass at collation. Noted by He Zhuo. įȘŨŅ:mīôȝgYƫɸƷ̠Ȕ˶ The Zhouli zhushu was a very common book in the late Ming and early Qing period, but when He Zhuo encountered an edition printed from blocks originally carved in the Song and repaired in the Yuan, he conscientiously compared the texts of a Ming edition and this Song-Yuan edition and                                                1 See He Zhuo, Yimen xiansheng ji, “Appendix;” Shen Tong, “Hanlinyuan bianxiu zeng shiduxueshi Yimen xiansheng xingzhuang,” in He Zhuo, Yimen dushuji, 1275.  126 recorded the variants (yiwen ȷƊ) in the Ming edition—one of the basic collation methods of Chinese traditional textual criticism, called duijiao ĉƷ (to collate according to comparison). Before long, one of his students Shen Tong devoted much time and energy to transcribing the variants onto another edition.  Shen Tong was a well-known classicist in the early Qing period and was considered by Quan Zuwang nɗƨ (1705-1755) to be one of He Zhuo’s most famed students.2 After He Zhuo’s death, it was Shen Tong who composed the xingzhuang ˎȠ (lit. brief biography) of He Zhuo. Shen Tong specialized in the study of the Three Ritual Classics (Sanli ɛ, i.e., Rites of Zhou, Rites and Ceremonies, and Records of Rites). He Zhuo’s marginalia, the variants from that Song-Yuan edition in particular, were very important information with which to study the text of the Three Ritual Classics. Therefore, transcribing his teacher’s marginalia, especially the variants and textual criticism, was a way for him to accumulate knowledge and research resources. As mentioned in the introduction, the most prevalent scholarly method in the Qing dynasty, evidential research, was an evidence-based approach mainly focusing on the study of ancient Chinese texts. Accordingly, reliable study was based upon reliable evidence, and reliable evidence first and foremost referred to reliable texts. Most Qing scholars had faith in the reliability of Song and Yuan editions. If Song and Yuan editions were unavailable, variants that could demonstrate the textual characteristics of Song and Yuan editions were a good substitute. This is one reason why He Zhuo’s marginalia, of which the majority were variants and other information regarding Song and Yuan editions, were treasured and transcribed in the Qing dynasty. Besides Shen Tong,                                                2 Quan Zuwang nɗƨ, “Shen Guotang mubanwen” ǭƴÍÐȝƊ, in Quan Zuwang ji huijiao jizhu nɗƨ͙ļƷ͙Ǵ (Shanghai: Shanghai guji chubanshe, 2000), 361.  127 other scholars such as Hui DongśƼ, Wu Xin ³Ƙ, Wang Xinfu Ȧ」Ü also transcribed He Zhuo’s marginalia for this title to other copies. This will be discussed in detail below. Another student of He Zhuo who made a great effort in the transcription of He’s marginalia is Jiang Gao ʾư (courtesy name Zizun è̦, alternative name Huangting ɳ:, 1683-1732). Jiang Gao received the jinshi degree in 1665, and was appointed Director of the Ministry of Revenue (hubu langzhong Ŭ̲̱" ), and subsequently Governor of Lianzhou İĜ , in Guangdong  province. Influenced by He Zhuo, Jiang Gao was an expert in book collection and identification of book editions. He Zhuo had taught him and identified book editions in his library Zhushu Tower (Zhushu lou ˽ơ。, alternate name Cishu lou ̃ơ。). So far, five titles in circulation today can be found that have He Zhuo’s marginalia transcribed by Jiang Gao:  1) Shitong «̛  (Conspectus of historiography), Ming woodblock edition cut by Zhang Zhixiang ĺ'˼, held at the Shanghai Library (XS832937-38);  2) Kunxue jiwen Ãðɺʘ  (Observations culled from arduous study), Ming woodblock edition held at the National Library of China (SB11096);  3) Xie Xuancheng shiji ˲üÊˢ͙ (Collected poems of Xie Tiao ˲Ɍ), manuscript copied by Jiang Gao in 1710 held at the National Library of China (SB08374); 4) Jiang Langxian Changjiang ji ̂Ǻ]͆ǧ͙(Collected works of Jia Dao ̂Ę, abb. Changjiang ji), Qing edition cut in Xi Qiyu’s Ĥ½ă Qinchuan shuwu ȩěơĐ during the Kangxi reign, held at the Fu Ssu Nien Library in Taibei;  5) Sanguo zhi ÆŎ (History of the Three Kingdoms), woodblock edition carved in the late Ming by the Mao family, held at the Shanghai Library (XS795562-69, Jiang Gao’s marginalia and  128 his transcription of He’s marginalia in this copy were transcribed by an unknown scholar named Diao Daigao ūͳ [alternative name Yueshanɻē]). There is a short colophon in the Xie Xuancheng shiji, stating: In the second [lunar] month of the gengyin year of the Kangxi reign [1710], I collated the Xie Xuancheng shiji in Master Yimen’s (He Zhuo) house, and copied this volume. Written by Jiang Gao, the Recluse of the Fragrant Cliff. įȘĪ”0ƦZʏ͇ģˋƷǔüÊˢ͙Ű̀xͰĖĊ͗ˁư In 1710, Jiang Gao was 27 sui, and He Zhuo 49. Jiang had learned how to authenticate book editions and collate classical texts from He. In He’s studio, Jiang copied not only a whole volume of Xie Tiao’s poetry anthology, but also his teacher’s marginalia along with the main text. The collating notes on the bottom margins have variants from a manuscript copy, another old manuscript copy, and a contemporary woodblock edition among others, which evidence He’s effort in collating the text of this poetry anthology. That copy is lost, so Jiang’s transcribed copy is the only source of information we have for it. Jiang Gao was also well known for imitating He Zhuo’s calligraphy. A Ming edition of the Shitong held at the Shanghai Library (call number: XS832937-38) is filled with He Zhuo’s marginalia, transcribed in red ink by Jiang Gao. We know that this edition was transcribed by Jiang because, firstly, there is a short colophon at the end of the book stating, “[This colophon] was composed by my teacher Yimen, and transcribed by [Jiang] Gao on the eleventh day of the seventh month of 1710” (ʏ͇ģ˟Ī”Ʀ’Ƒừ); and, secondly because the handwriting by Jiang in semi-cursive style small characters resembles that of He Zhuo (see Figure 4-1). Jiang was well known for his skill in imitating He’s handwriting. Therefore, marginalia transcribed in a hand  129 that resembles He Zhuo’s have been attributed to Jiang. In another instance, there is a copy of the Shitong in a Ming woodblock edition held at the Shanghai Library (call number: XST01248-53, see Figure 4-2), whose red marginalia looks almost the same as He Zhuo’s, so much so that a late Qing book collector, Deng Zhengan ̵̮̙ (courtesy name Zhengan ǔ͌, 1868-1939), initially thought it was He’s holograph. But after seeing an authentic holograph version of He Zhuo’s marginalia in another copy of the Shitong,3 Deng realized that the marginalia in this copy was instead a transcription. Because the handwriting was almost the same as He Zhuo’s, Pan Boshan ȉ—ē  (1904-1943), a well-known book collector, claimed that the marginalia had been transcribed by Jiang Gao.4 Just like his teacher, Jiang Gao became a book collector and textual scholar. He not only transcribed He Zhuo’s marginalia, but also collated many books carefully and wrote his own marginalia in exquisite handwriting. A mid-Qing scholar and poet, Peng Zhaosun Ŀhʿ (courtesy name XianghanȃǾ, or Ganting Ȭ:, 1769-1821) praised Jiang Gao: Mr. [Jiang] Huangting collated hundreds of Classics and histories. All of them were refined and done meticulously in red and yellow ink, and were praised in artistic and literary circles. Collectors treasure copies formerly held in the Zhushu Tower. ɳ:iȮŰƷˮʄ«Ɖ’ȿɠɀ#ͺɹ˴ˇƳůɡ˽ơ。ƫŅʕˆĵBȐć                                                3 This copy of Shitong that bears He’s handwriting is now held at the National Library of China, call number 11313. 4 See the colophon by Ye Jingkui ʻƟʽ (1874-1949) at the beginning of this book, and the colophons by Wu Zhizhong³Ŏŏ and Deng Bangshu ̵̮̙ at the end.  130  Figure 4-1  Last half-leaf of the Shitong, woodblock edition carved in the Ming dynasty. This colophon was composed by He Zhuo, and transcribed by Jiang Gao. Courtesy of the Shanghai Library (call number: XS832937-38).  131  Figure 4-2  Leaf 22b of juan 20 of the Shitong, woodblock edition carved in the Ming dynasty. The marginalia, which resemble He Zhuo’s handwriting, were transcribed by an unknown transcriptionist. Courtesy of the Shanghai Library (call number: XST01248-53).  132 Besides Shen Tong and Jiang Gao, other students of He Zhuo, such as Chen Shaozhang ͒Ɵ͟ (courtesy name Shangzhang ċɩ,1670-1747), Yao Shiyu æ̾ (courtesy name Yucai ȥˑ, alternative name Yitian (Ȱ, 1695-1749), and Wang Youdun ǩȱƈ (courtesy name Shitiao ģʳ, 1692-1758) also transcribed their teacher’s marginalia. It seems that He Zhuo sometimes also received feedback from them. In juan 38 of the Hou Hanshu held at the Peking University Library, the main text reads: “K̩uōƆ(” (Why did we fail so quickly?), for which He Zhuo’s marginal note states: [Chen] Shaozhang says: A character “ɰ” (ce) is missing after “̩” (ju). ċɩ˯“̩”ċ “ɰ” é This sentence was the last from General Zhang Zong’s ĺö speech to his commander when he was a general and their army was in a critical situation. Zhang asked to be the rear guard, saying:  I have heard that when one soldier exerts all his strength [to fight], he will be mightier one hundred men; once tens of thousands of soldiers face death with no regrets, this army will be invincible. I, Zhang Zong, now have an army of thousands of soldiers and have the great power. Why must we fail so quickly? şʘ–ȵˆȿ;ȸʸÜʣǚªB》ˎö=žtƉ“BųÚçK̩uōƆ( Here, Zhang Zong says that they still had thousands of soldiers and therefore victory was possible even their enemy had more soldiers. Thus it is more reasonable for the last sentence to be “Why do you hastily expect our failure?” rather than “Why must we fail so quickly?” Chen Jingyun was  133 right that “ɰ” (ce, to expect) should be added in this sentence. One of He’s students, Chen specialized in the Hanshu, Hou Hanshu and Tang poetry. He Zhuo mentions him several times in his marginalia and praises his erudition on the history of the two Han dynasties. In marginalia to a Qing edition of the Tangshi guchui »ˢͽ´  (Tang poems for accompaniments), carved in the 16th year of the Shunzhi reign (1659) and preserved in the Chongqing Library, the communication between the master and his students is more apparent. In this copy, Chen Jingyun first transcribed his teacher’s marginalia. He Zhuo then wrote down additional comments on both the main text and his marginalia as transcribed by Chen Jingyun, and Chen later wrote many comments on these previous texts. Wang Xinfu found that the marginalia in this book were written by He Zhuo and Chen Jingyun, observing that “He Zhuo’s handwriting is smooth and ornate; while Chen Shaozhang’s handwriting is  solemn and careful” (Kơǹ͸͒ơŖ˴).5 Concerning the content of the marginalia, he commented: “Most of He’s words consist of evaluation; most of Chen’s words are evidential research” (K˧ØˡDz͒˧Øʔſ).6 In juan one of this copy, there is a piece of He Zhuo’s marginalia on Liu Yuxi’s ‡ɜ͂ poem “Ku Lü Hengzhou” º¶ːĜ (Crying for Lü Hengzhou), transcribed  by Chen Shaozhang, which reads: “Seeing this piece, we can know that the talents of Liu [Yuxi] and Liu [Zongyuan] are far apart from each other” (˜ǕɵªBɑ‡□űˆɈŸ'̤). But later this sentence was crossed off and two other characters, “butong” ̛ (nonsense), were written alongside by He Zhuo himself. Chen Jingyun appended his own opinion as follows:                                                5 Wang Xinfu Ȧ」, Yishuxuan qie cun shanben shulu ˏ̌ƫờ (Shanghai: Shanghai guji chubanshe, 2002), 676. 6 Wang Xinfu, Yishuxuan qie cun shanben shulu, 676.  134 The comment about the talents [of Liu Yuxi and Liu Zongyuan] is one piece of my deceased teacher’s old marginalia, which I transcribed in the jisi year [1689]. It was later crossed out. The running comments on the two characters were written by my deceased teacher in the spring of the xinwei year [1691]. When talking about essays of the Tang dynasty, Master Ouyang also claimed that Liu [Yuxi] and Liu [Zongyuan] were equal, which suffices for us to know that the ancients initially did not think there was a difference between these two masters. It is appropriate that master [He] did not think his earlier comment was fair. űˆɈŸ'ˡ%Lğġ{̀iģʨŲŁ ŶŸ]Ǵ0éʕ…iģ̔ƩƙͯŲ,ǒ͔qˬ»AƊɩ8B‡□ö!ɡ̉ɑ§;10q‚ȓ̌̏úģB†ˡȐf,7  In this last piece, Chen referred to He Zhuo as his “deceased teacher”, indicating that it was written after He’s death (1722), more than 30 years after Chen first transcribed He’s marginalia (1689). From other examples recorded in the marginalia of this copy, we learn that during those thirty years, He Zhuo changed his opinions several times because of his accumulation of research materials. The most interesting point is that He Zhuo did not revise just his own marginalia editions, but also those where his marginalia were transcribed by his students. One possible reason is that He Zhuo wanted to get some feedback from his students or other scholars.  Sometimes, he also obtained new texts from his students. In the Su Xueshi ji ,ðÒ͙ (Collected works of Su Shunqin ,ʫǎ, 1008-1048, courtesy name Zimei èʎ, an outstanding                                                7 Wang Xinfu, Yishuxuan qie cun shanben shulu, 676.  135 poet of the early Song period) held at the Shanghai Library (XS824382-83), He Zhuo’s colophon, which was transcribed by Qian Taiji ́ǵ® (courtesy name Fuyi ̐ú, 1791-1863), reads: I am annoyed that the newly carved Suzimei ji has many mistakes. One of my students, Lü Yili of Shimen County, whose father [Lü] Wudang ranked second in the imperial examination of the wuxu year [1718], said his family had an old manuscript copy. It now belongs to one of his uncles. He has only kept the copy collated by his deceased father. Thus, I borrowed it and collated my own copy. …Written by He Zhuo in the third month of the gengyin year of the Kangxi reign [1710]. ƌ͈͚	,èʎ͙LȻuˠ˳ʢØðńɒ͇¶ŦǘŧŨǂɍȓͼiȮ'è˝uþƧʨ̼ƫ=Ġ€ĒˮțʖuiqůƷ'ƫȣêÂ[BƚǔįȘĪ”ƦƞƑKȔ˟ In the tenth month of this year, student Li brought his family’s old copy to me, which was a manuscript copy produced in Mr. Wu Wending’s Congshutang in my hometown and obtained by his grandfather Mr. [Lü] Wancun from the Qi family in Shanyin County. On the basis of this copy I revised the order in the table of contents and wrote this [colophon] at the end. There are tables of contents at the beginning of each juan in this manuscript copy. Younger generations have no chance to see the Song edition, yet this can serve as evidence of it. Zhuo. ƚǗʭƦ¶ȮŸuþʨƫʢ%µ̴³Ɗøq¦ơÍ̼ƫuɗƝ̯iȮŅ'ē͑ɕǤʕ,ÂŅƁǔɆ̀š「ʖ˟uŁ̼ƫǠš'ͯ ­ƧɆŁȮŅ:ô͚Ǖ8̉BȐſɐȔ  136 This colophon tells us that He had a chance to see a manuscript copy of Su Shunqing’s poetry anthology. He collated his own copy referencing it, writing down the variants and these colophons. This manuscript copy was brought to him by one of his student, Lü Yili, who was the son of the book collector Lü Baozhong ¶ʺŏ (courtesy Wudang ȓͼ, ?-1708) and the grandson of the well-known scholar, thinker and physician Lü Liuliang (courtesy name Yonghui ȯƞ, alternative name Wancun ƝƮ, 1629-1683). According to He’s colophon, Lü Liuliang got this manuscript copy from the Qi family in Shanyin (in Shaoxing, Zhejiang Province). The Qi family was renowned for book collection in the late Ming era. The catalogues of their collection, the Danshengtang cangshumu ȎȮ͈ơɆ (Catalogue of the Dansheng Hall) and Danshengtang dushuji ȎȮÍ˸ơ˟ (Reading notes written in the Dansheng Hall), are still very important documents for the study of book production, circulation and collection in the late Ming.8  Mid- and Late-Qing Scholars In addition to his students, He Zhuo’s marginalia were also treasured by scholars after him, some of whom spent significant time and energy transcribing them. Among them was the prominent classicist Hui Dong śƼ (courtesy name Dingyu øñ, alternative name Songya ƱĚ, 1697-1758). According to the Zhongguo guji zongmu and several other library catalogues, Hui Dong, cooperating with his father Hui Shiqi śÒÝ (courtesy name Tianmu ÛȞ, alternative name Bannong •̗, 1671-1741), transcribed He Zhuo’s marginalia on five titles, which are: 1) Zhouli zhushu, 2) Chunqiu gongyang zhushu ƙɞqʍǴȹ (The Spring and Autumn Annals of the                                                8 See Qi Chenghan ɕų΂, Danshengtang cangshumu ȎȮ͈ơɆ and Danshengtang dushuji ȎȮÍ˸ơ˟ (Shanghai: Shanghai guji chubanshe, 2015).  137 Gongyang tradition with annotations and sub-annotations), 3) Hanshu, 4) Hou Hanshu, 5) Sanguo zhi. Hui Dong was considered one of the earliest scholars who publicly and clearly advocated applying the Han scholars’ approach to the study of the Confucian Classics, as opposed to that of Song scholars. He once claimed that “The damage done by the Song scholars is even more serious than the Qin bibliocaust” (ôe'əȭƎɟȏ). In order to rectify the misinterpretations caused by Song scholars, Hui, along with other Han Learning scholars, tried to solve some fundamental textual problems at their origins by devoting themselves to the collation of the Confucian Classics.9 In this regard, He Zhuo’s marginalia, which not only included variants from rare Song and Yuan editions but also were preliminary researches on these textual problems, were of special significance. For instance, He Zhuo’s marginalia on the Zhouli zhushu, one of the Five Confucian Classics, were initially transcribed by his student Shen Tong and then by Hui Dong and other later scholars. Hui Dong transcribed them into a copy of a new edition that was cut in his own studio, the Hongdou Study (Hongdou zhai ɼ˺;). Based on that edition, he conducted his own research and recorded his own marginalia in it. Unfortunately, Hui Dong’s copy is not in circulation today; fortunately, his marginalia and transcription of He Zhuo’s marginalia were all transcribed into another copy by one of his friends, Shen Wotian ǭǫȰ. Shen’s copy is also lost, but a scholar-official, Wu Xin ³Ƙ (courtesy name Yunge ʱ͊, jinshi 1774) transcribed all these accumulated marginalia into yet another copy. This copy is a woodblock edition cut by the Maoǣ family in                                                9 See Qi Yongxiang Ȇǥɘ, Qian-Jia kaojuxue yanjiu -¿ʔſðɓɦ (Beijing: Zhongguo shehui kexue, 1998), 137-159.  138 the first year of the Chongzhen reign (1628) and now held at the Shanghai Library (call number: XST06322-31). At the beginning of this copy, Wu Xin’s colophon reads: This book was originally annotated and collated by the Hui family in Yuanhe. The red characters were Mr. Bannong’s (Hui Shiqi, 1671-1741) handwriting, and the green characters were written by Songya [Hui Dong] the recluse. They probably first transcribed Mr. He Yimen’s marginalia, consisting mainly of textual criticism based on comparison with a Song edition that had been repaired in the Yuan dynasty and preserved in the Palace. They continued to transcribe [variants from] the Yu family’s Wanjuantang edition … also referring to the Xu family’s edition, … an edition with commentaries on the meanings, … an edition with annotations quoted within the Classics … [and] sometimes an edition that they call the Jian edition [presumably a Jianyang imprint]. … It is recorded at the beginning of the book that it was collated against a Song edition and the Wanjuan edition; when the other editions were cut is unknown. There are some items beginning with “[Lu] Wenchao’s comment”, which were recorded by Master Lu the Hanlin academician. In the spring of the xinsi year of the Qianlong reign [1761], I borrowed it from Mr. Shen Wotian and started to transcribe [the marginalia into one of my copies], and finished in the winter of the renwu year [1762] … Written in Chusonglou, on the day following the Shangsi Festival in the gengyin year [1770]. Noted by Wu Xin. ƚơžƫVg¸śǤͻ‹ɼɯ•̗iȮů͋ʅɯ…ƱĚŊ±ů‰,ÚɻìKʏ͇iȮůƷmīôƲgYƫʊ̀LǤʸšÍƫ… Ƨłþƫ…˞ʏƫ…3Ǵƫ…!^Ƨɡijƫʕ…šͯǓ̎BôƫʸšƫƷuͭˮƫɑKƛů、," 139 ƧƉƺ2“ƊĹƸ”22ʕ=ðÒɅĝů˟-͕̔ġ'ƙLņǫȰǭˋZ̘̀Ӕ{íʖɪ…Ī”ġŁƑơ1ƽͧ。³Ƙ˟ There is the impression of Wu Xin’s seal at the end of this paragraph. According to Wu Xin’s statements in this colophon and the marginalia in this book, Hui used multiple editions to collate the text of the Zhouli zhushu. For Hui, He Zhuo’s record of the information obtained from the Song-Yuan edition was quite important. Although most Qing scholars were not satisfied with or were even fiercely opposed to Song Learning, they had a fetish for Song editions.10 Hui Dong had composed two books on the the Rites of Zhou: Lishuoɛ˫ (On Rites) and Jiujing guyi +ʄ§ʏ (Ancient meaning of the Nine Classics), but none of his marginalia in this Zhouli zhushu were included in these two works. These scattered marginalia are not unimportant, though, because some items can contribute to an understanding of Hui Dong’s original attitude toward Han and Tang dynasty scholars.11  Hui Dong’s marginalia also contained some important information about a Song edition of the Zhouli zhushu, a woodblock edition cut in Yu Renzhong’s L<C Wanjuan Hall (Wanjuan tang ʸšÍ), which is no longer in circulation today. This edition was rather famous in the Qing period, but it was already difficult to get access to it. So when Lu Wenchao got a copy, he rebound it and offered it to the emperor. Before the offering was made, Hui Dong had a chance to read it and                                                10 For instance, the noted Qing bibliophile Huang Pilie ͺȑ (1763-1825) called himself “Ningsong zhuren” Nô$; (Master of flattering Song editions). Most Qing scholars treasured Song editions and attached great importance to Song editions in book collation. 11 See also Wang Xinfu, Yishuxuan qie cun shanben shulu, 733.  140 managed to compare it with his Hongdou Study edition. Hui recorded this process in his colophon, which was transcribed at the end of Wu Xin’s copy: Master Lu Yayu [Wenchao] obtained an annotated Song copy of the Rites of Zhou and will offer it to the emperor. In the spare time  during the rebinding, I read it once and transcribed the variants into my copy. This copy has twelve juan in total, and each juan is bound in one volume. In the 28th day of the last month of the yihai year [1755]. Songya [Hui Dong]. ͘͞ɅqŅô、ƫʄǴ·ɛĈB̜ŇÂ˓Ȋ'ƠƷ̠͋ơs’0šǠšxƛ*7’0ƦĊ͐ֆƑƱĚ According to the various colophons at the beginning and end of Wu Xin’s copy, this book was later acquired by the scholar-official Han Yingbi ͤţ͏ (courtesy name Duiyu ĉˌ, alternative name Lüqing ʅ›,1800-1860), and more marginalia were added by Shen Chengtao ǭ˨ș and Zhang Yiqing ĺE› in the Xianfeng reign (1850-1861). After Han Yingbi’s death, his books were all scattered. This copy ended up in the collection of the famous book collector Ye Jingkui ʻƟʽ (courtesy name Kuichu Ż‚, alternative name Juan’an šɄ,1874-1949), a friend of Wang Xinfu. Ye Jingkui also possessed a copy that has Shen Tong’s marginalia. In the 1930s, Wang Xinfu spent about two years transcribing the marginalia of these two copies into another Ming edition. The transmission of He Zhuo’s marginalia on the Zhouli zhushu is outlined below in Figure 4-3. From Shen Tong and Hui Dong in the High Qing to Ye Jingkui and Wang Xinfu in the Republican era, He Zhuo’s marginalia were transcribed by generations of scholars. In this process, new sources and the opinions of the transcriptionists were added to previous marginalia.  141 Therefore, this was not only a process of transmission, but also a process of proliferation of scholarship, which can reveal how Qing scholars did their research.  Figure 4-3  The transcription of He Zhuo’s marginalia on the Zhouli zhushu.  Hui Dong’s main concern was the authentication of the text, so what he wrote and transcribed were mainly marginalia concerned with textual criticism. This is what many Qing scholars did. He Zhuo collated the Hedong xiansheng ji ǯƯiȮ͙ (Collected Works of Liu Zongyuan □ög). One of his friends, Wang Wenyuan Ȧʘ̤ (1663-1741), borrowed He Zhuo’s collated copy and collated his own copy based on He’s marginalia. 12 He Zhuo had collated the Changjiang ji. His                                                12 Wang Wenyuan’s colophon states: “I have been searching for the Hedong xiansheng ji for years, but got nothing. … In the winter, I got this copy from the bookstore in front of the City God Temple. All the  142 textual criticism was transcribed by Jiang Gao and later by one of the most prominent textual scholars of the Qing dynasty, Lu WenchaoɅƊĹ (courtesy name Shaogong ©ķ, 1717-1796).13 Other examples are too numerous to mention individually. Not only did He Zhuo’s marginalia enjoy this kind of privileged status, but that of other scholars, especially textual scholars such as Lu Wenchao and Gu Guangqi ͬIJÉ (courtesy name Qianli“̸, alternative name Jianpinȍˉ, 1770-1839), also circulated in this way. Transcribing prior scholars’ marginalia, reading them closely, and even adding new comments was one of the main forms of scholarly activity for Qing scholars. The text of ancient classics was shaped, and this scholarly practice influenced the reading and reception of traditional texts in the Qing dynasty and even today. For example, the most widely used and influential edition of the Thirteen Confucian Classics with collation notes produced under the charge of Ruan Yuan ͎g (courtesy name BoyuanGg, alternative name Yuntaiʱʤ, 1764-1849) was edited and carved based on some Yuan editions and many Qing scholars’ collations in their marginalia. The prefaces of these Thirteen Classics have listed the marginalia employed in these books, which are shown in the table below:14                                                information concerning layout and taboo characters was copied from a Song edition. There are quite a few omissions and errors. This fall, I saw in He Qizhan’s library his collated edition. So, I borrowed it, took it back and collated [this copy] for three days. … and stopped. So I procrastinated and only finally finished after more than twenty days.” (	ǯƯiȮ͙;'ƉĨȤ{ŅƎÊ͖ı†'ơʙuơˎǏ¡̪˭éŗȗôƫŴ̀Jʝȇʪ˩ͨØ=ɞƎKĮĥĔɎ:uƷƫÂ[Ǚ˞Ƒžơ%M̒Âʼn̟̥ʢ0’ͭƑäƷȵ). See Zhongguo guji gao jiao chaoben tulu, 678-679. 13 See Fu Zengxiang, Cangyuan qunshu jingyanlu, 1083-1084. 14 See the Shisanjing zhushu ’ʄǴȹ, Ruan Yuan, chief editor, photo-reproduced edition (Taipei: Yiwen yinshuguan, 2007).  143  Classic Source YiƗ (Book of Changes) Lu Wenchao Qian Sunbao ́îW (1624-) Shi ˢ (Book of Songs) Lu Wenchao Zhouli ·ɛ (Rites of Zhou) Hui Dong (incorporating his transcription of He Zhuo’s marginalia) Lu Wenchao Duan Yucai Yilidɛ (Rites and Ceremonies) Lu Wenchao Gu Guangqi Lijiɛ˟ (Record of Rites) Hui Dong Lu Wenchao Sun Zhizu îŎɗ (1737-1801) Duan Yucai ǝȥˑ (1735-1815) Chunqiu Zuozhuan ƙɞĞ`  (Zuo annotation on the Spring and Autumn Annals) Hui Dong (incorporating his transcription of He Zhuo’s marginalia) Lu Wenchao Duan Yucai Chen Shuhua ͒《ʷ (1730-1801)  Gu Zhikui ͬ'̝ (1754-1797)  Chunqiu Gongyangzhuan ƙɞqʍ` (Gongyang annotation on the Spring and Autumn Annals) Hui Dong (incorporating his transcription of He Zhuo and He Huang’s marginalia) Chunqiu Guliangzhuan ƙɞɤƹ` (Guliang annotation on the Spring and Autumn Annals) Hui Dong (incorporating his transcription of He Zhuo and He Huang’s marginalia) Duan Yucai Mengziìè(Mencius) He Zhuo and He Huang Lu Wenchao EryaȜ͘ (Explaining refined language) Hui Dong Lu Wenchao Xiaojing ëʄ (Classic of Filial Piety) Lu Wenchao Table 4-1  Marginalia employed in Ruan Yuan’s edition of the Three Classics   144 It can be seen from this table that He Zhuo, Hui Dong and Lu Wenchao’s marginalia contributed significantly to Ruan Yuan’s edition of the Thirteen Classics. Meanwhile, the marginalia of other scholars such as Sun Zhizu, Chen Shuhua, and Gu Zhikui also did fundamental work on this influential project, but their names are nearly unknown not only in the modern Western literature, but also in that of China (even Qian Sunbao’s date of death is unknown). This is a problem that deserves further consideration. Only after carefully examining these marginalia and their composers and transcriptionists can we have a deeper understanding of the creation and significance of Ruan Yuan’s edition of the Thirteen Classics. In the Qing dynasty, scholars were not only concerned with textual criticism; some were also interested in He Zhuo’s comments or both his comments and textual criticism. Therefore, they transcribed both, and also some non-literary marks and symbols. They were concerned not only with the meaning of the main text but also its form—the physicality of the book and the form and circulation of the marginalia per se. They paid much attention to the various scholarly practices related to the collection and reading of the book, the process of making sense of the text, and the production and transcription of marginalia. The practices themselves were of great significance to them. For instance, when Weng Tonghe ʐ¯΁ (courtesy name Shuping £ħ, alternative name Pingsheng ȫȮ, 1830-1904) was fifteen sui, he bought a copy of the Tang Liuxiansheng ji »□iȮ͙ (Collected works of Liu Zongyuan in the Tang Dynasty) with He Zhuo’s handwritten marginalia in it.15 Weng was so excited that he composed a poem about this book, wrote it down on a separate piece of paper and put it inside. About fifty years later, Weng transcribed He’s marginalia into a new copy. He wrote a colophon at the end of this copy:                                                15 This copy is now held at the National Library of China, call number SB06251.  145 I obtained the Liuxiansheng ji with marginalia from the Chen family’s Jirui Pavilion when I was fifteen sui. The red marginalia were rather illuminating. They were written by He Qizhan. I intended to make a copy but have not managed to. In my spare time this spring when I was lecturing in the Yuchun Palace, I collated it once roughly. My eyes were almost blind and my wrist unsteady. I am not as strong as when I was young, for which I sigh in sorrow repeatedly. Recorded by Weng Tonghe on the seventh day of the fourth month in the renwu year [1882]. LĨ’4ŅŲƫ	□iȮ͙ƎɢȪ。͒ǤƬơȚȕKĔɎŰ̋,ŞǍʠĆƫ––ƴ=ĨƙƎlɇǡƙý˱'ƠČǟ˕ɲĊɧɸƷ̠ɆɉʞȌ͢ňċĻ\ɐȐ'ǑӔÁƦ‚Ƒʐ¯΁˟ Weng Tonghe was a noted scholar-official and the teacher of two emperors, Tongzhi¯ǰ (r. 1861-1875) and Guangxu jʈ (r. 1875-1908). He was especially well known for his calligraphy and was very confident about it. At the end of the table of contents of the Tang Liuxiansheng ji on which Weng transcribed He Zhuo’s marginalia, Weng’s colophon reads:  The copy with Mr. Yimen’s marginalia was a Masha edition from the Yuan dynasty. This copy is a Ming edition produced in the Zhengtong reign. It is the same as the Yuan edition in its chapter order and phonetic notation, but different in the number of columns per page and characters per column. I transcribed Mr. He’s marginalia into this copy in the renwu year [1882]. After twenty years, in the renyin year [1902], I read it again and felt that the characters [I wrote] were big and beautiful, a bit better than Mr. He’s handwriting. Recorded by Weng Tonghe (Songchan) on the last day of the second month.  146 ʏ͇iȮŰŲƫVgƛ͹Ǯƫ,ǕƖǔʂƫʥg„ɵɮͦǴŗ¯ŚˎƉéƉØĄ3ȷʗLƎӔĨʠiȮˡƷƎxm̈0’ĨÓ”̹͋˚éÚʝɆ̍iȮŰ̋,0ƦƞƱɚʐ¯΁˟ This comment not only revealed Weng’s confidence about his calligraphy, but also showed that while transcribing marginalia, the calligraphy was a concern. However, there was no evidence (e.g. no seals) showing that the marginalia in the copy he got when he was fifteen sui were actually written by He Zhuo. As a matter of fact, it is more likely that they were transcribed by some unknown scribe. Besides calligraphy and textual criticism, Weng was also interested in the marginal comments composed by He Zhuo. He had transcribed He Zhuo’s marginalia on numerous titles, such as the Hanshu, Sanguo zhi, Shitong, Gengzi xiaoxia jiĪèǼÕ˟ (Record of whiling away the summer of the gengzi year), Han wen chaoͤƊŴ (Collected works of Han Yu ͥŝ), Tang Liuxiansheng ji, and Jiayou ji ¿ɖ͙ (Collected works of Su Xun ,Ƿ). For all of them he transcribed both the collation notes and He’s comments. There are two colophons by Weng at the end of the table of contents of the Shitong held at the Shanghai Library (call number: SX782275-79). The one in blue reads: I have a copy [of Shitong] with marginalia by [Qian] Muweng (Qian Qianyi) transcribed by Qian Xiangling. I collated this copy with it and find that the comments that start with “Feng says” in this copy were all by Muweng. He Zhuo perhaps concealed [the truth and avoided mentioning Qian Qianyi’s name]. I supplemented the emphasis marks in blue ink and added on the top and side margins several pieces of what was left out by He. Written by Tonghe in  147 the sixth month of the jiashen year [1884]. The emphasis marks seem to have been made by Xiangling rather than Muweng. þˆ́ȃ͠iȮʠȞʐŲƷƫȯBĉ‹%ɑǕƫmůɡͱ2ʕɀȞʐˡ,Kˀ˭'ʗÂB)ʮ˒uÅͻ!˒KůȇʕƉƺƎ͍¡ˎ]ȲȳrƦ¯΁˟ÅͻHȃ͠iȮůȐ͢Ȟʐ, The one in red states: The copy held in my house has Royal Attendant Wang Genzhai’s comments. So I transcribed them in this copy in purple ink. The emphasis marks are a little careless, and the comments unreasonable most of the time. So, I suspect that they were not composed by Royal Attendant [Wang Jun]. However, they point out the literary ingenuity of the text, so I transcribed them in order to make the text convenient for beginners to read. Noted by Shuping. þˆƫƧȦʬ;TŇˡ˧ÂBɿʮɯʠ'uÅͻͨȚȈʥˡ˧Ø̣ŭˋȺ͢TŇŰɯȕͨȶƊɩ'ãƃ!ʠ'UƎ‚ð˪˸£ħ˟ Qian Qianyi ́˰ɂ (courtesy name Shouzhi ¥', alternative name Muzhai Ȟ;, 1582-1664) was a prominent poet and scholar in the late Ming and early Qing periods. Wang Jun Ȧę (courtesy name Genzhai ʬ;, 1694-1751) was a scholar-official, specializing in history and geography. Weng transcribed their comments and emphasis marks. He also transcribed those that he suspected were not composed by Qian and Wang because “they point out the literary ingenuity of the text.” For Weng Tonghe, previous scholars’ comments were useful for “beginners” to grasp the literary features of the text. Meanwhile, we can also see that Weng paid lots of attention to the emphasis marks made by previous readers. Obviously, the “comments” here mentioned by him  148 were mostly literary comments; the emphasis marks were also devices for helping to point out and analyze the literary features of the text. As discussed in chapter two, literary comments and emphasis marks were introduced to analyze the literary features of classical prose in order to help students prepare for the civil service examinations in the Song dynasty, and in the late Ming were widely used in works of fiction and drama to guide readers in appreciating the aesthetic values of these texts. Scholars in the Qing dynasty still used them in their marginalia, and Weng also expended much effort transcribing them. Weng asserted that they were for “beginners,” but how many “beginners” could have access to his transcriptions? The more reasonable explanation is that the comments and emphasis marks were for Weng himself to appreciate the literary beauty of the text. In a time when evidential research was the mainstream scholarly approach, when pure beauty and pleasure were excluded from scholarship and scholarly practice, scholars managed to find other ways to meet this kind of need. Weng rationalized his practice of pursuing beauty from the text by asserting that it was for unspecified “beginners.” In the Qing dynasty, many scholars also paid significant attention to the form and physicality of both the main texts and marginalia. They tried their best to “transcribe” not only the literary writings in marginalia but also all sorts of symbols and even the way in which they were drawn. One example is Zhang Yu ɩ̾ (courtesy name ShizhiĶ', alternative name Mingyi ʵɶ, 1864-1934), a late-Qing book collector and textual scholar. He transcribed He Zhuo’s marginalia on several titles, such as the Sanguo zhi, Wudai shiji 4A«˟ (History of the Five Dynasties), Geshibian ǐˢʉ (Poetry Anthology, i.e., anthology of Li He’s poems), and San Tangren wenji»;Ɗ͙ (Collected works of three Tang essayists), just to name a few.  What is interesting about Zhang Yu’s transcriptions is that he transcribed not only the collation notes and comments but also the seals. In the San Tangren wenji held at the National  149 Library of China (call number: SB14604), He Zhuo’s marginalia as transcribed by Zhang Yu is preserved. On the first page of juan one of Sun Qiao’s îLJ anthology, Zhang’s marginalia records two of He Zhuo’s seal impressions (see Figure 4-4): Love the ancients without belittling the moderns (intaglio characters in square shape) ˃/=;/Š§;ȾƊƍ˜ “Talking about antiquity,” a dragon in the left and a tiger right (intaglio characters in rectangular shape) ˧§Ğ΀¬ˊȾƊ͆ƍ˜ On the page listing the table of contents of Li Ao’s ƭʒ anthology, Zhang recorded two more impressions of He Zhuo’s seals: He Zhuo’s personal seal    Qizhan (two square seals in relief characters) KȔɝ˜    ĔɎoƬƊƍ˜ This demonstrates that Zhang had seen the book read by He Zhuo and that he transcribed He’s marginalia and recorded the contents of the marginalia—such as the seals—into his own copy. Zhang’s handwriting is in semi-cursive style small characters that resemble He Zhuo’s. Although He Zhuo’s copy was lost, Zhang Yu’s effort of duplicating He’s marginalia and seals can help us see how He Zhuo’s original marginalia and seals looked.   150  Figure 4-4  Leaf 1a of juan 1 of the San Tangren wenji, woodblock edition carved in the Ming dynasty and mended in the Qing dynasty. The marginalia in this copy were transcribed by Zhang Yu. On this half leaf, there are three seal impressions that were transcribed by Zhang. Courtesy of the National Library of China (call number: SB14604).  151 Scholars’ Relationships Built with Marginalia In the process of transcription, marginalia were duplicated and disseminated to more books. Meanwhile, the scholars involved built up a kind of relationship, in which knowledge, scholarly approaches, and a particular set of values were also transmitted from one person to another and from one generation to another. The examples cited below show how Qing scholars were connected by marginalia. At the beginning of the Sanguo zhi held at the Shanghai Library (XS491435-54), there is a colophon composed by Wang Qisun Ȧʯî (courtesy name Nianfeng Œ˻, alternative name Lengqie shanren ƾIē;, 1755-1818) and transcribed by Fei Yuanshan ̀Ȅȁ (courtesy name Runquan ȋDZ, late 19th c.). It reads: This collated copy of the Sanguo zhi is held at the Shili House of my jinshi classmate Huang Yaopu. Yaopu obtained it from Chen Shuhua (alternative name Yequan), who previously lived in Wugang Prefecture. The marginalia that were transcribed by Yequan were from the Hui family’s Hongdou Study edition. The collation made by Feng [Ban] was well-known for its refinement, and in addition to He Yimen and Chen Shaozhang’s revisions and the Hongdou Study’s [owner’s] transcription of them, and later Yequan’s transcription—all these people were good readers of my hometown—making this copy error-free. Ǖ	ÆŎƷƫL¯Ĩͺ˂ÄʐÒɛďůˆ˂ÄŅ'ƃǖĕĜ'¯͒|DZ《ʷ|DZůĬʕśǤɼ˺;ƫ,ͱƷɡɹ¾ɥBKʏ͇͒ċɩiŁąǔʖɼ˺;̀'|DZĬ'Ǖu;ɀµ̴'¾˸ơʕúȓʪ̓ɐ  152 On the Wenxuan Ɗ̧ (Selections of refined literature) held at the Zhejiang Library (S3563), there were He Zhuo’s marginalia transcribed by Sun Qilu îǿ (alternative name Zhuxiang ɭ̴) and Wang Shijin ȦĶ̺. Wang’s colophon at the beginning reads: Recently, I happened to visit a bookstore and saw a Wenxuan commented on by He Zhuo and transcribed by Mr. Sun Zhuxiang. Zhuxiang was indeed a reading seed in my hometown. This copy was carefully collated and is worthy of appreciation. ̬ʕ^̠ơʙƧKˡƊ̧Vɭ̴îʠƫɭ̴ƃµ̭˸ơɠè,Ʒ˹ɹąÚÎŵȧ Details of Sun Qi’s life are unknown to us. Wang Shijin praised him as a “reading seed” in his hometown. Wang’s life is likewise unknown, but his marginalia and transcriptions of previous scholars’ marginalia were also very refined and precise. He was also a reading seed. Wang Qisun stated that He Zhuo, Chen Jingyun, Hui Dong, and Chen Shuhua were all good at reading. They of course are also reading seeds. For them, writing, reading, and transcribing marginalia were important practices in their lives. Their world was interwoven with different texts, among which those with marginalia were one of the most important kinds. They exchanged books with each other and transcribed marginalia from prior scholars and from the transcriptions of their contemporaries. Despite living in different times and spaces, they were connected by marginalia, and  relationships were formed based on practices related to the transcription of marginalia. In the case of He Zhuo, as discussed above, his marginalia were passed on to his students such as Shen Tong, Jiang Gao, and Chen Jingyun. They transcribed their teacher’s marginalia and learned from him how to collate books and improve their handwriting. In return, He Zhuo occasionally received feedback from his students; sometimes, he also received new texts from  153 them. He Zhuo also established various relationships with his teachers and friends via the exchange of marginalia. Li Guangdi ƭjÈ (courtesy Jinqing Ɯ›, alternative name Rongcun ǁƮ, 1642-1718) was once He’s teacher. He Zhuo transcribed Li’s commentaries on various titles, such as the Hanwen chao.16 In addition, he and his brother He Huang KȖ (courtesy name Xinyou Ō¢, 1668-1745), a book collector and textual scholar, also worked together on some texts and exchanged texts via post. He’s colophon at the end of juan 7 of the Hou Hanshu held at the Peking University Library (call number: LSB/7288) reads: In the jiawu year of the Kangxi reign [1714], my younger brother Xinyou got an incomplete Song edition [of the Hou Hanshu] from juan 3 to the middle of this juan held by the Ye family of Baoshan. He sent me the character variants after collation, and I corrected several tens of errors [in my own copy]. įȘȲ”Ō¢ĸŅēʻǤůˆǜôƫɮšʢǕš'•BůƷéĆ“ÂƁǔƉ’ˋ After collating one book, He Huang always copied the character variants and his comments, sending them to his elder brother. Here, the text was transmitted, rather than the book itself. The dissemination of the text was relatively separate from the book. There are several books in circulation today collated by both He Zhuo and He Huang. For most of them, the marginalia were produced in this way. The Ye family mentioned by He Huang was well-known for book collection and production from the Southern Song dynasty through the end of the Qing. He Zhuo established an intimate                                                16 Hanwen chao, manuscript copy held at the National Library of China, call number 06252.  154 relationship with the Ye family and many other book-collection families, which gave him access to quite a few rare editions. For instance, in the Wudai shiji held at the Shanghai Library (call number: XS847873-78), He’s colophon (transcribed by Yao Shiyu æ̾)  states: In the winter of the jiashen year of the Kangxi reign [1704], I borrowed from Chuyin, the eldest son of Qian Zeng (courtesy name Zunwang), the Wudai shiji that was once read by father Dongjian the Minister of Rites. I, therefore, transcribed his marginalia. This book was read by the Minister of Rites in his youth and thus his comments are not refined. But compared with others’, his are already neat and tidy. I also added and subtracted a few parts and am very worried on account of my rash work. Noted by Zhuo. įȘȲȳ{Ƒņˌḗƣ̦ȦiȮ͆èƽǞZŅöGƯȍʐů͋	4A«˟Âʖ`'Ǖơ%öGÔĨ͋ƫƩȐɹ…ȕ˘[;…ɊɆ5ȕvɐL8ċƧÑżǛBâMʡť2 Ȕ˟ He Zhuo borrowed this copy from Qian Chuyin and transcribed the marginalia of the latter’s grandfather Qian Qianyi, who was a great poet and scholar and whose marginalia were filled with profound insights. Qian Qianyi was also a book collector. His library, the Jiangyun Tower (Jiangyun lou ʃ͟。) held a great number of rare books. Unfortunately, the majority of the collection was destroyed in a fire in 1650. Qian Qianyi gave the remains to his nephew Qian Zeng ́ƣ (courtesy name Zunwang ̦Ȧ, 1629-1701), also a noted book collector and the father of Qian Chuyin.17                                                17 See Qian Zeng ́ƣ, Dushuminqiuji jiaozheng ˸ơƄǦ˟Ʒǔ, revised by Guan Tingfen ɱĭʰ and Zhang Yu (Shanghai: Shanghai guji chubanshe, 2007).  155 Qian Qianyi was a prominent poet, the leader of the Yushan School (Yushan shipai ˌēˢǸ ). The Feng brothers, Feng Shu ͱʩ  (courtesy name Jicang ğʲ , alternative name Chanshoujushi ïòďÒ , 1593-1645) and Feng Ban ͱȨ (courtesy name Dingyuan ø̤ , alternative name Dunyinlaoren ̻²ʓ;, 1602-1671), were the backbone of the Yushan School, as well as being book collectors and publishers. Their books and marginalia were transmitted to He Zhuo by the Qian family. For example, Weng Tonghe transcribed He Zhuo’s marginalia on Shitong into a copy of a Ming woodblock edition (now held at the Shanghai Library, call number: 782275-76). In He’s marginalia, there are many comments by Feng Shu that were transcribed from a copy held by the Qian family. He Zhuo’s colophon recorded how he got access to that copy, transcribed the marginalia and knew that the marginalia were written by Feng Shu. It reads: On the ninth day of the ninth month of the jichou year [1709], I borrowed from Qian Chuyin the copy once read by Chanshoujushi [Feng Shu]. Then I transcribed the comments. What was interlinear was transcribed on the bottom margin. [Feng Shu’s] comments are sharp and excellent most of the time, and are greatly valued by scholars in Yushan. Only Ji Cangwei the Attendant Censor has managed to borrow it before. I would have no chance to see it if Chuyin were not fond of me. Initially I thought it was read by Qian Muweng when he just started to work in the Historiography Institute. Therefore, I wrote “Qian’s comment” on the top and bottom margins. After carefully asking Chuyin, [I knew that they were composed by Feng Shu rather than Qian Qianyi, and thus] I made some corrections. Zhuo. ğ̹͔ņ́ƽǞZŅïòďÒ͋ƫẦuˡ˧uÇˎ]ʕ̀'͍˷ˬ8ØʴŐˌēðʕǀɏ̹'bíȅʼTŇ;À̛[Ȝ͢ƽǞàũƩȱ:,ä˩BȐȞʐ‚l«ͮƛů͋ƃ͍ɀĆ́ˡˤ̄'ƽǞ%Ɓǔ2 Ȕ  156 He Zhuo was a big fan of the Feng brothers. He not only transcribed their marginalia, but also collated and commented on one of Feng Ban’s works, the Dunyin zalu ̻²̀͜  (Literary miscellany by Master Dunyin). He’s marginalia on the Dunyin zalu were transcribed by later scholars, and text and marginalia were edited and printed at the same time.18 He Zhuo also established an intimate relationship with the Mao family, one of the best-known families for book collection and production. He exchanged texts with many scholars and bibliophiles such as Lu Yidian ͓́w (courtesy name Chixian ƅi, 1617-1686), Mao Qiling ǣÝͿ (courtesy name Dake Úª, 1629-1713), and Xu Qianxue ł-ð (courtesy name Yuanyi ž, 1631-1694). Exchanging books and transcribing their marginalia was the most common way marginalia circulated throughout the Qing dynasty. In the late Qing and Republican era, despite the penetration of Western powers into China and initial intrusion of Western culture into every corner of Chinese society, many scholars and book collectors still lived in the tradition of this scholarly practice; they established closer relationships with each other through it, and by it they were connected as a cultural unit and endeavored to preserve and maintain traditional texts and culture. Of course, they can be called “reading seeds,” from which the traditional culture could sprout and, one day, perhaps come into being in a new form. For instance, there is a colophon by Zhang Yu in the Sanguo zhi held at the National Library of China (call number: SB14731), stating:                                                18 The Dunyin zalu held at the Shanghai Library (XS771771-72) and two copies at the National Central Library (Taibei) all have He Zhuo’s marginalia. The Dunyin zalu in the Jieyue shanfang huichao ZƦēŮĽ̼, zhihai Źǻ, and other collectanea all has He Zhuo’s marginalia printed as commentaries.  157 The copy that was collated by Mr. Zhu Qiuya was obtained by Fu Yuanshu of Jiang’an. I intended to transcribe [the marginalia] and collate [my copy]. Therefore, Wu Wanbo of Renhe bought it and sent it to Tianjin. … After I finished transcribing it, Shi Zhonglu of Kuaiji borrowed the Jin[lingshu]ju edition to collate. He recorded piece by piece the differences between the Mao and [Jinlingshu]ju editions. His work was careful and refined. Hence, I transcribed his marginalia too. This all happened five or six years ago. On the thirteenth day of the fourth month of the gengshen year [1920], there was rain after a long period of drought. Noted by Mingyi. ƬɞœiȮƷƫȐǧó_Ǭ£ůŅ̾ƧŞʠƷ<¸³Gù%̅Ǖƫ“ǶʠƷȵƫƥɢƏC͵ňZƷ̺τƫ}τʥǣ„ȷʕ̚ƺ˟~ǀȐɹąň€ƒ̀lǕɀ†4rĨ/ĪȳÁƦ’Ƒ̖̇&ƓŅ̿͞ɶ˟ Zhu Bangheng Ƭ̮ː (courtesy name Qiuya ɞĚ, 18th c.) transcribed He Zhuo and Hui Shiqi’s marginalia. The original copies were lost, so Zhu’s transcription is of great significance. This copy was once held by Fu Zengxiang _Ñȃ (courtesy name Shuhe £¸, alternative name Yuanshu Ǭ£, 1872-1949), and then brought by Wu Changshou ³ƕʆ (courtesy name Wanbo ùG, 1867-). Wu sent it to Zhang Yu who transcribed the marginalia into another copy. After Zhang finished the transcription, he transcribed the variants from another copy. Marginalia circulated among these scholars and bibliophiles, sometimes along with the book that carried them and sometimes via transcription. Fu Zengxiang was one of the best-known book collectors in the late Qing and Republican era. Centered on him, a textual “communication circuit”— to employ the term proposed by Robert  158 Darnton19—formed. Scholars who knew or did not know Fu Zengxiang could directly or indirectly get particular texts from him. Through exchanging books and texts, they learned from each other and authenticated book editions with the help of their collections. For the Shitong held at the Shanghai Library (call number: XST01248-53) mentioned above, Deng Bangshu’s colophon reads: I got this book in the Wu region and thought it was collated by Yimen. After seeing the newly bought copy my jinshi classmate Yuanshu [Fu Zengxiang] brought from the capital, I knew that this was a good copy recently transcribed …The handwriting was dignified and vigorous, and could pass for He’s. Besides, the contents were also selected properly. It’s a pity that it does not provide the transcriptionist and the collector’s names, which makes it hard for us to recognize that Yan Biao was not [an offspring of] Lugong [Yan Zhenqing].20 In the winter of the renzi year [1912], written by Zheng’an. ³"ŅǕơ˩BȐʏ͇ɯƷƐ˜Ǭ£¯Ĩůˆƌʡ̳"̅Ǚʕ%ɑǕȐ¯ƛ̠̀'Oƫé̋ɬ̡ͨʜ.K'ɋŸ¤8ǀƋ̷řuʚɇþʕåǤP;ͪ̕Dž͢͵qʗÓè{Ƒǔ͌˟ Wu Zhizhong’s colophon reads:                                                19 See Darnton, The Kiss of Lamourette, 107-135. 20 Here Deng Bangshu used an allusion to Yan Biao ͪDž (9th c.). When Yan Biao took the civil service examinations in 854, the chief examiner made him the optimus (zhuangyuan ȡg), because the chief examiner thought him an offspring of the great calligrapher and loyal governor Yan Zhenqing ͪɋ› (709-785). Contemporaries satirized the chief examiner for “mistaking Yan Biao for [an offspring of] Lugong [Yan Zhenqing]” (̓˦ͪDžM͵q). See Wang Dingbao ȦøW , Tang zhiyan »Ž˝  (Shanghai: Zhonghua shuju, 1960), juan 8, 88.  159 Mr. Zheng’an got this copy and thought it was collated by He Zhuo with He’s own handwriting. This spring, nianzhang (a man of one’s father’s generation) Fu Yuanshu bought a copy once held at the Yuhua Pavillion. This copy has one more colophon which is transcribed above. In the whole book, it is rather clear that for Feng’s comments, the character “feng” was all corrected from “qian.” There are two more colophons by He Zhuo written in the bingxu [1706] and guisi [1713] years respectively and one colophon by Gu Jianping. That copy was considered the genuine version with He’s holographic marginalia, and this a transcribed version…Not only is the imitation of He’s calligraphy in cursive style verisimilar, but also the red and yellow writings are exquisite. It is worth treasuring. I borrowed it from Mr. [Zheng’an] and collated my copies, and then collated it in return based on what I have seen. By Wu Cipei, courtesy name Ouneng, on the twenty-fourth day of the six month of the renzi year [1912].  ǔ͌iȮ‚ŅǕƫBȐKǤŰƷʕ=ƙ_Ǭ£ĨƀŅʹʷ͊ʨˆ̲ƪج̛̲̀̊ͱˡͱéɀ́éÏƁƖȾª̕ ɮšŁØKǤįȘŨȽġͩ˶0…¡ͬȍˉ̊%ąøŀƫȐŰƷɋ̋Ǖ…`ʠ'ƫȟdžDKʶ̞ɋ#ͺɹàªɏ˾,LņiȮZƷ̠Âʧů:ʕ̫BČǔȒÓèrƦĴÁƑ^ʜ³šÌ˶ These colophons tell us that the handwriting of the marginalia in this copy is almost the same as He Zhuo’s, so Deng and Wu initially thought it was written by him. After seeing the authentic copy held by Fu Zengxiang, they finally made it clear that the marginalia were not He Zhuo’s holographs. De Bangshu also wrote a colophon in the copy held by Fu Zengxiang.21 Deng’s                                                21 See Fu Zengxiang, Cangyuan qunshu jingyanlu, 507.  160 colophon was written in the sixth month of 1912, and Wu’s in the winter of that year. At the beginning of this copy, there is a colophon written by the great industrialist and book collector Ye Jingkui ʻƟʽ (courtesy name KuichuŻ‚, alternative name Juanan šɄ, 1874-1949) in the second month of 1940, one year after Deng Bangshu’s death. It reads: This collated edition was transcribed by He’s students and it was once seen by Yimen. Zheng’an [Deng Bangshu] and Ouneng [Wu Cipei] tried to identify [its transcriptionist] successively, but it is a pity that it is unsigned. There is a seal impression at the beginning reading “Held by Jiang Weijun’s family in the Wu region,” and at the end reading “My house is among the nine peaks and three calm ponds.” Mr. Pan Boshan doubted it was transcribed by Jiang Gao, courtesy name Zizun. But Boshan has a Houshan shizhu (Annotations of Houshan’s  Łē [Chen Shidao, 1052-1101] poems) in an early Ming edition collated by Jiang Gao. The style of Jiang’s handwriting in it is simple and plain, not similar to that in this copy. Zizun’s younger brother Bing, courtesy name Zifan, was also Yimen’s student. We are unable to see [examples of] his handwriting, so further studies are needed [to solve this problem] … Written by Jingkui in the second month of the gengchen year [1940]. ǕƷƫȐKǤĸèů`ʠȐʏ͇ů˙:ǔ͌^ʜiŁʔøřžƫƩʌå°šͯƧ³͇ˁʇ̽þˆ˜ščƧþÇ+ėdz͉˜ȉ±—ēȺȐˁè̦ưůʠJ—ēˆƧè̦ŰƷƖ‚ƫ	ŁēˢǴéͲ̍Ȑ§LjʥǕͫè̦'ĸƻéèɴ8ʏ͇ĸèȓņ?ųʑÿyʔǦĪ̖0ƦƟʽơ There is an impression of Ye Jingkui’s seal after this colophon. This colophon and seal show that this copy belonged to Ye, probably after Deng’s death. Ye was an old friend of Deng, and once  161 borrowed this copy, transcribing the marginalia into another copy.22 Since Deng and Wu had already made it clear that the marginalia in this copy were not written by He Zhuo, Ye tried to discover the identity of the transcriptionist. He also discussed this with his friend Pan Chenghou ȉų (courtesy name Boshan—ē, 1904-1943), a printer and book collector.  Throughout the Qing, marginalia were transcribed by different scholars into different copies. Texts, knowledge, thoughts, particular scholarly approaches and lifestyles were transmitted through this practice. Scholars and bibliophiles were bound together by marginalia. In the Qing dynasty, there were also other ties, such as doctrinal affiliations, clan relationships, classmate relationships, local literary societies, and other social relationships, which all played a significant part in binding scholars together, transmitting texts, knowledge, and thoughts and helping to create and consolidate scholarly identity. Marginalia and the practice of transcription strengthened these ties. It was a more efficient way to transmit texts (especially the character variants, collation notes and philological comments) and knowledge across time and space. It not only tied contemporary scholars together but also handed down texts and knowledge from generation to generation. This can also be illustrated by a colophon written by the most famous late-Qing book historian Ye Dehui ʻŋ̑, in a copy of Chunqiu Gongyang zhushu now held at the National Library of China (call number: SB07937). It states: The collated edition of Gongyang zhushu (28 juan in total) was made by He Huang (courtesy name Zhongyou, Mr. Yimen’s younger brother) in the dingyou year of the Kangxi reign [1717]. He Huang recorded the character variants from a Song governmental edition into a copy of Mao’s Jigu Pavilion edition that has both the main text, annotation and sub-annotation. In the                                                22  See Ye Jingkui ʻƟʽ, Juan’an shuba šɄơ̊ (Shanghai: Shanghai guji chubanshe, 2006), 64-67.  162 guiyou year of the Qianlong reign [1753], Hui Songya (Dong), the great recluse, produced a new collated edition based on He’s textual criticism and added the variants from a Song edition held by Cao Yin, Commissioner of the Bureau of Transmission; a Shu large-character edition; and the original annotated-and-sub-annotated edition. One of his students, Zhu Bangheng transcribed [the textual criticism and comments into another copy]. In the guichou year of the Qianlong reign [1793], Zang Yong, courtesy name Zaidong, also transcribed the marginalia into a new copy. In the seventh month of that year, Duan Yucai, courtesy name Maotang, also transcribed the marginalia into another copy. Jiang Yuan, courtesy name Tiejun thus transcribed Duan’s transcription into this copy and gave it to his student Chen Huan, courtesy name Shuofu, and Chen recorded the transmission history at the beginning of this book in the first year of the Xianfeng reign [1851]. The collation notes of one book were transcribed many times; famous scholars and classicists collated it carefully over and over again. Transmitted for more than two hundred years, the book is not only complete in volumes and chapters but also rather refined in the collation of characters and the analysis of sentences. We must keep in mind that the diligence and studiousness of former generations, and their care in reading classics, far exceed that of later young men, who usually write carelessly and do not finish what was started. Ʒôƫ	qʍǴȹ0’pšįȘ̶KC¢Ȗʏ͇iȮ'ĸBô、÷ƫƷƎǣƜǪ§͊„Ǵȹƫ-͕Ƚ̶śƱĚŊ±ƼſKƷÑlƢ̛Ƃ”ůˆôƫˍÚéƫžƲǴȹƫ̹ƷƫuĊ͇ȮƬ̮ːʠƷ'-͕ȽʟÇƯ̈́Í8ʠƷ̲uĨƦǝŤÍȥˑ ʠ̲ǧͅ±Ǭňņǝʠ̠̀ǕƫB́u͇«͒ɔțßƎ¹˻ɺgŰơu`źƎšɬơ'Ʒ̀ʄȓƉ°;ʄģy 163 ÁʀŪǹ`0ȿͭĨ'&͢ȣšxõà8Ʒé͝¨ɹ…ȷĥŜ:ʓ̓àð'ŽŠ˸ʄ'ąŢX͢ŁȮĊèXŰÏŶƧäȓʁʕªǢ, These predecessors, who were industrious in learning and circumspect in reading Confucian Classics were, of course, reading seeds. By transcribing marginalia, they formed a “communication circuit.” Furthermore, in this communication circuit, in the practice of transmitting marginalia throughout the Qing dynasty, not only were scholars and bibliophiles involved, but the role of book merchants and some unknown scribes were also of great importance.  Merchants and Unknown Scribes In addition to scholars, many merchants and scribes also actively participated in the practice of transcribing marginalia. They contributed to both the transmission and proliferation of marginalia. For instance, in the Changjiang ji held at the Fu Ssu Nien Library in Taibei, there is a colophon by Jiang Gao that reads: The Changjiang ji was collated most meticulously by Master Yimen. In the summer of the renyin year [1722], the Master passed away in his house in the capital, and his books were all scattered. Three month later, a person came to me to sell this book, as well as Wang [Wei] and Meng [Haoran]’s works. I was short of money at that time, [so I was not able to purchase them, which] I regretted for a long time. My exam classmate Shu Zizhan said: “If we transcribe the marginalia to other copies, we can regard the new books with marginalia as the original ones.” Therefore, I found three volumes of works by Changjiang [Jian Dao], Youcheng [Wang Wei], and Xiangyang [Meng Haoran], and gave them [to Shu Zizhan to transcribe]. It happened so quickly that I had no time to revise. Afterwards, the Changjiang ji was left in my  164 hands, and yet Wang and Meng’s works were possessed by the Pan family in Wuxing. I was pleased by Zizhan’s idea and felt lucky that the Changjiang ji was still in my possession. During a break from work at the beginning of the spring of the year jiachen [1724], I revised it once and then returned the book. People who are fond of the antiquity understand and do not blame me for my scribbles. Written by Gao of Sanjing. ʏ͇ģůƷ‹	͆ǧ͙ƤȐɹʀÓ”Õģ–Ǝ9̨̰ĢƇʹƦŁƧBƚĢ¡ȦìˢR¼ʕƛǔ)́Řŕ&'¯Ĩʩèđ2Bƒƫ̠~Ȣá:ɋƫ,Â~Ƶ͆ǧ¬ ˔͔ˢx:?‘‘ƢŒɨƠĉƷŁ	͆ǧ͙ȴ1LˋʖȦì͙ĠĒ³ʦȉǤɐL¿èđ'Ŏʖĩ	͆ǧ͙'ȢêȲ̖ƙ‚ƒF'ƠɸƷ̠Þ̫à§ʕɑʋuÏͷ,Ńư˶23 This colophon can shed light on several points: Firstly, books with He Zhuo’s marginalia were rather expensive in the early Qing period. Jiang Gao was an official and his abundant book collection showed that he was not poor at all. His fiscal limit to the purchase of this volume was likely because it was too expensive. Secondly, in another book with He Zhuo’s marginalia transcribed by Pan Zhiwan ȉŎʸ (1849-1899), Pan’s colophon said that He Zhuo’s marginalia was originally written on an edition cut by Feng Ban in the late Ming or early Qing period, which cannot have been highly priced in the early Qing period. Jia Dao’s Changjiang ji was by no means a rare book at that time. So, what made the book expensive was only due to He Zhuo’s marginalia. Thirdly, within three months of He Zhuo’s death, book sellers knew to sell books with his marginalia to his students and other book collectors (such as the Pan family in Wuxing) at a very                                                23 Fu Zengxiang, Cangyuan qun shu jingyan lu, 1083.  165 high price. Books with marginalia must have been popular to a considerable extent, from which booksellers found economic opportunity and plotted their financial gain.  In the Qing dynasty, there were quite a few complaints about the high prices of books containing marginalia. Although scholars were sometimes not able to purchase them, they had chances to read them and even transcribe the text. For example, Weng Tonghe, who was the teacher of two emperors, also failed to buy a book with He Zhuo’s marginalia. His colophon in the Jiayouji held at the National Library of China (call number: SB05397) reads: On the eighth day of the forth month of the bingchen year [1856], a bookseller brought the Jiayou ji with He Zhuo’s marginalia to sell. It was too expensive, so I borrowed it and transcribed the marginalia within one day and night … By Tonghe. ̖ÁƦ‚pƑƧBKʏ͇ŰŲ	¿ɖ͙R¼ʕɇȭƔÂ[ŅɃƑÙ'ˆʠƷ̠¯΁˶ Wu Cipei’s colophon on the Yuanfeng leigao g˻ͫɣ (Collected works of Zeng Gong ƣͣ) held at the Nanjing Library (call number: 118744) also states that he spent two days and one night transcribing He Zhuo’s marginalia onto a new copy. His colophon reads: A bookseller brought the Yuanfeng leigao with Mr. He Yimen’s marginalia to sell. The whole book was collated carefully based on a Song edition, and was filled with red-ink inscriptions. It deserves to be treasured, even though it was a transcribed edition. The bookseller wanted to sell it at a very high price. I was not able to purchase it, so I hurriedly found a copy into which to transcribe [the marginalia]. The bookseller asked me to return the book in a great hurry. I, therefore, finished the transcription in two days and one night. I transcribed all the collation notes. As for the comments and evidential studies, I had no  166 time to transcribe them … By [Wu] Cipei, on the next day of the Shangsi Festival in the third year of the Xuantong reign [1911], when I was living in the capital as a guest. Meanwhile, I bought the edition carved by Wang Yu in the Jiajing reign. It was carved exquisitely. Initially I thought of transcribing He’s textual criticism onto this Wang edition. But I hurt my finger last winter on the road to the Huai Region. It has not recovered yet, nor can I hold a brush, so my handwriting is clumsy and obscure, which would deface a good book. Therefore, I bought this copy to transcribe … Written on the Qingming Festival. ơ̂ŸKʏ͇iȮŲƷ	g˻ͫɣR¼̛̲ȯôƫʀ‹ƬơɁĦ͛~[Ű`̀8̉ć˾ơ̂ͨďÝLˆʜ̅6¤ƫʠƷơ̂ɾžơȭŔZoƑÙ'ˆ–ƿȒ}ǽƷ‹ȵ̀ȓ̨ʢˡ˧ʔ˵…Ơ¡,üʂĨġŁƑšÌ˶ƛû9ģ ¯ƛ̅Ņ¿͉͡Ȧőƫȝ„ȭɹ‚œ̀KƷ1ȦƫŸ{'Ȁ̢"aŰŹʢ=ƩȼʭËɯMơŷȌ̹ǨOɷ%˿Ǖƌ、ƫ̀'ȂƖ ˥ These examples show that it seems to have been a convention for prospective purchasers to keep a book for one to two days. During this time, they could decide whether to purchase the book or not, and more importantly, examine the book closely and even transcribe the marginalia into other copies. In this process, the exchange of physical books failed, but the transmission of texts—marginalia in particular—succeeded. Since marginalia were marketable, booksellers therefore hired scribes to transcribe, and even to forged well-known scholars’ handwriting. By so doing, the booksellers and scribes were also involved in the marginalia culture. They contributed to the duplication and dissemination of marginalia, and yet caused many troubles as well. For instance, there are many books with He  167 Zhuo’s marginalia in transcription, rather than He’s holograph, where the handwriting resembled He Zhuo’s, but the transcriptionist did not leave any information about himself or the transcription process. This caused lots of problems for authenticating He Zhuo’s marginalia. For the Shitong mentioned above, Deng Bangshu and Wu Cipei were able to know the marginalia in it were not written by He Zhuo only when then saw an authentic version. There are two copies of Tangren xuan Tangshi »;̧»ˢ (Tang poems selected by Tang scholars) in the Mao edition carved in the late Ming held at the Shanghai Library. Both have He Zhuo’s marginalia, of which the calligraphy and contents are similar. But one copy (call number: XS839789-96, see Figure 4-5) has the impression of both He Zhuo and the Bao Ͷ family’s seals, which means that this copy was once held by He Zhuo himself and then the Bao family, one of the most famous book-collecting and publishing families of the mid-Qing period. However, the other copy (call number: XS756297-304, see Figure 4-6) was probably a forgery by some book seller. Even decades after He’s death when Jiang Weijun collected He’s marginalia and compiled the Yimen dushuji, Jiang found that there were lots of forgeries of He’s marginalia. He writes: Not only is it difficult to authenticate the books in circulation that purport to contain Yimen’s reading notes, but even in books contain He’s real marginalia errors abound because they have gone through several generations of recopying. ×͉`Ćʏ͇ˡ͋'ƫȟɋ̆ɽá™Vɋƫʖ̼̀Ɖ̠kʪˠ24                                                24 He Zhuo, Yimen dushuji, “Fanli”}S (Guide to the reader), 2.  168  Figure 4-5a  Leaf 1a of the Yulan shi in the Tangren xuan Tangshi, woodblock edition carved by the Mao family in the late Ming. The marginalia in this book were written by He Zhuo. Courtesy of  the Shanghai Library (call number: XS839789-96).  169  Figure 4-5b  Leaf 65b of the Yulan shi in the Tangren xuan Tangshi. The seals here read: “He Zhuo’s seal” and “Held by the Bao Family.” Courtesy of  the Shanghai Library (call number: XS839789-96).  170  Figure 4-5c  Last half leaf of the Heyue yingling ji in the Tangren xuan Tangshi. The colophon in red was written by He Zhuo. The seal was He’s courtesy name, “Yimen.” Courtesy of  the Shanghai Library (call number: XS839789-96).  171  Figure 4-6  First half leaf of the Yulan shi in the Tangren xuan Tangshi, woodblock edition carved by the Mao family in the late Ming. The marginalia were transcribed by an unknown scribe. Courtesy of  the Shanghai Library (call number: XS756297-304).  172 While booksellers did not forge He’s marginalia based on nothing, most of the time they transcribed He’s marginalia anonymously, in imitation of his handwriting. In this sense, they also contributed to the transmission of marginalia.  Who could these unknown scribes be becomes the last question. Scribing was a very old profession. Since antiquity, some scribes worked as low-level government officials. So scribes, both governmental and non-governmental, were elegantly called “chaoxu” ̼ʛ (lit. scribing officials) in the later imperial period. They were also called “chaoshou” ̼Ű (lit. scribing hands) and “shushou” ơŰ (lit. writing hands). In the late imperial period, non-governmental scribes were usually impoverished examination candidates and local teachers who did this work as a part time job, or professional scribes who lived by this occupation.25 For the great bibliophile and publisher Mao Jin ǣƜ (courtesy name Zijin èƜ, 1599-1659), leader of the Mao family in the late Ming, it is recorded: “Walking through [Mao’s] gate, all the servants and house boys were transcribing books” (l͇ɫcɃ̼ơ).26 Since scribes mainly worked for money and their work was the least creative, they were looked down upon and were paid a very low salary.27 For a scribe, there was little difference between transcribing the main text of a book and transcribing marginalia. Most of the time, one kind of marginalia were transcribed by one scribe, but sometimes the marginalia of a voluminous book were transcribed by more than one scribe. For example, He Zhuo’s marginalia on the Hou Hanshu held at the Peking University Library (call                                                25 See Zhou Kan·Q, “Tangdai shushou yanjiu” »AơŰɓɦ (PhD diss., Capital Normal University [Beijing], 2007). 26 See Ye Dehui ʻŋ̑, Shulin qinghua ơƳȂˣ (Shanghai: Shanghai guji chubanshe, 2008), 143. 27 See Ye Dehui, Shulin qinghua, 214.  173 number: LSB/7288) were transcribed in three distinct hands, probably by three different scribes (see Figure 4-7).     Figure 4-7  Last half leaf of the table of contents and leaf 11b of juan 84 of the Hou Hanshu, and leaf 2a of juan 3 of the Xu Hanzhi, woodblock edition carved by the Mao family in the late Ming. The marginalia in this book were probably transcribed by three scribes.   174 Transcription Culture Before He Zhuo’s time, the transcription of marginalia was not a common practice. He Zhuo, his younger brother He Huang, and some contemporaries composed much marginalia and transcribed some prior scholars’ marginalia. Soon after He’s death, more and more scholars were involved in the practice of transcribing marginalia. This practice became a prevalent scholarly culture from the mid-Qing to the Republican period. Marginalia served as a “communication node” that connected various agents—including scholars, bibliophiles, merchants and scribes—who together broke down barriers of time and space. The transcription of marginalia involved various people and was driven by different motives. In He Zhuo’s case, transcription, for his students, was a way to learn from their teacher; for other scholars, it supplied abundant research sources for both collating and making sense of a text; for calligraphers, it was a good opportunity to practice after a masterpiece and improve their skills; for book sellers and scribes, it was a profitable enterprise.  Through transcription, scholars and bibliophiles were connected by marginalia and formed a tight relationship. They exchanged books and, sometimes, only texts; they learned from each other about the authentication of book editions and the collation of texts. They inherited texts, research materials, and a way of life utilizing research from previous scholars; they also handed down to later generations what they had obtained from these texts and practices. Transcription was also a proliferating process. While being transmitted from person to person and from book to book, marginalia were usually not transcribed in a way that was one-hundred-percent loyal to the original copy. They were selected, modified, and distorted sometimes. Something would go missing, and more would be added. Therefore, new interpretations were able to be born therein; new meanings  175 could be generated; new culture and thoughts could be created. In this practice, tradition maintained continuity without being frozen. This practice came into being against a particular political, social, and economic background. It was initiated by some scholars, and followed and disseminated by other scholars and agents. It helped to build up a relatively efficient model of the transmission of information, knowledge, thought, and even scholarly life style. This was a particular scholarly culture, a systematic way, for scholars, of thinking, analyzing, behaving and living. The next chapter will discuss Qing scholars’ thoughts, mental states, and life styles as reflected in the practice of transcription undertaken in particular times and spaces.  176 5 Transcription: Time, Space, and Scholars  Zhao Yi ˼ʄ (courtesy name Xuesong ͊Ě, alternative name OubeiȤ‡, 1727-1814), a great historian, poet and critic of the mid-Qing period, wrote in one of his poems: Busy collating in a curtain of books all day long. / Stepping outdoors, it is already the dry autumn I had not felt at all. ̴ƊƝĤƱƒo̶ˋğɎͣ1 These two sentences describe the scholarly life of a scholar in the mid-Qing. In Zhao Yi’s description, “all day long” and “dry autumn” are the temporal elements; “curtain of books” and “out the door” are the spatial elements; “collating” and “had not feel at all [the coming of the autumn]” manifest the practices and the mental state of a scholar in a particular temporal and spatial context. Time and space are two of the most fundamental elements in pre-modern Chinese writings. Writers intentionally or unintentionally selected particular temporal and spatial elements to encapsulate what they had done and/or what they were thinking, so as to sculpt images in their writings. Therefore, analyzing the temporal and spatial elements and deciphering the hidden meanings underlying these elements can help us to comprehend the writer’s thoughts, attitudes and mental states. This chapter will examine the temporal and spatial elements in Qing scholars’                                                1 Zhao Yi ˼ʄ, “Wan bu cunluo” ƕnjƩʫ, in Zhao Yi, Oubei ji Ȥ‡ͅ (Shanghai: Shanghai guji chubanshe, 1997), juan 43, 478.  177 marginalia, so as to picture their scholarly life, the characteristics of their scholarship, and their attitude and thoughts on scholarship and the world they lived in and perceived.  Temporal and Spatial Records in Marginalia In the marginalia of Ming-Qing China, what is impressive are the detailed records of various temporal and spatial elements in colophons. Colophons in pre-modern Chinese marginalia came in various forms. There were very long colophons at the beginning and/or end of the book, introducing and evaluating its main ideas, stating the history of its circulation, and picturing its physical features. This kind was called “ba” ˾ (lit. postscript) in Chinese. There were also short colophons, usually called “tizhi” ͖˪ (lit. remarks and notes) or “ji” ˑ (lit. records), commenting on the contents of the book, or, most of the time, recording the readers’ reading practices. Let us first read several colophons composed by He Zhuo, so as to get a better sense of what might be recorded in them. There are several colophons by him in the Hou Hanshu held at the Peking University Library (call number: LSB/7288). All of them recorded temporal and spatial elements in detail. At the end of the fourth juan, He’s colophon reads: Written in the Babaixuan in the western chamber of the Qingyuan Auxiliary Palace in the six month of the xinsi year of the Kangxi reign [1701]. Zhuo. ĪȎ̉ĠØfƟǰʡˀʕˇħdƭ̂  ȉ The colophon at the end of juan 9 reads: In the jiawu year of the Kangxi reign [1714], my younger brother Xinyou got an incomplete Song edition from juan 3 to the middle of this juan held by the Ye family in Baoshan. He sent me the variants after collation, and I corrected several tens of errors [in this copy].  178 ĪȎȫ‹њĴľ†ėʬǖŨʺǑûƦɛ“ʓNj“#Œ;ŨƱðĉ”¾ź》Ƃ‰ʿ The colophon at the end of juan 57 reads: From juan 45 to this point, collated by the light of an oil lamp in comparison with an incomplete Northern Song edition, in the tenth month of the guisi year of the Kangxi reign [1713]. Written by Yimen the recluse. ʒ¼‰0“ʓNj;‡ûǑƦȏũƱƓĪȎȶĠ́Ɵʀ̶ǿßˑ The colophon at the end of juan 90 reads: When I first read this book, I was upset at the number of mistakes. Once I had read the pieces on Hou Hanshu in Liu Ban’s [Liang Hanshu] Kanwu (Corrections to the Hanshu and Hou Hanshu), I finally knew that there were few exquisite editions in the Northern Song dynasty, because people did not value this book so much as Ban Gu’s book. Some of the annotation in the Jiajing edition carved by the Directorate of Education was cut off. While this copy is still a complete version, which is one good feature. Its master copy does not have so many mistakes. I corrected the mistakes created while carving in this edition. Written in the western chamber of the Baoding Auxiliary Palace in the Middle Autumn Day of the xinsi year of the Kangxi reign [1701]. Zhuo. sˬNjƝíh˒˦ȅÙ™ˌ|ǖ	r˟ˣƵ"ɅLJû’ɼµƦɵz6̭#äȜƝŻ'º͎ Ž4ÃïȻrʇǢɯuyNjȖüƝŻƒ̵hʗƦĞNjŁǸ˦#ðȯȅ̈́ƃź‘.ĪȎ̉Ġ Ɏļ͖ƇI‘ˀʕˇħ  ȉ  179 The colophon at the end of juan 22 of the Xu Hanzhi reads: From the nineteenth to the twenty-second juan, in the dingyou year of the Kangxi reign [1717], when I worked in the Wuying Publishing House, I saw by accident an incomplete copy of the big character edition carved in the Yijing Hall of Cai Qi (courtesy name Chunfu) in Jian’an in the wuchen year of the Jiading reign in the Song dynasty. It has Congshu Hall’s seal impression. I thought it must be good. Therefore, I borrowed it from the keeper and collated [my copy]. However, the mistakes could provoke anger …  I recorded [them] to show that Song editions can also be so unreliable, without presuming to defame them. Written by Yimen the old man, on the sixteenth day of the seventh month. ʒ‰%“ʓ,‰,“ĪȎ̦ɊĺǍʣƝĒQˉbûº‘š̌İúʱȟɧȑɯÍ̸͆ÜðƦƠʧƝ͐Ń;ȅńC¾Ŀjűʇ&;Ʊčxʙ˟ ȅŞ《˪#;ˉûƦ3Ơ˽ŸKäNj͏ſȅˏȃ'ƟƉƣʀ̶ʅǗƝ The colophon at the end of juan 23 of the Xu Hanzhi reads: In the sixth month of the guiwei year of the Kangxi reign [1703], I attended the eighth prince in the Nanxun Palace. There was a copy of [the Hou Hanshu] carved by Wang Wensheng, which I used to collate this juan. The Wang edition still followed the mistakes just as in the “Treaties on Geography” of the Hanshu. I am afraid of comparing them critically. Is it because of people’s carelessness that there are no good editions? Noted by Zhuo.   180 ĪȎȶƤfƟEd˰-ŽʷǒƬƠǝƃȹrƦ¾œ;ƱNj“ǝǖ38˒ˆʙä	zƝÈȞ3ŝƇ/ƝȈµƦ˯͏ʢɡ#̖±  ȉˑ The colophon at the end of juan 30 of the Xu Hanzhi reads: In the early summer of the xinsi year of the Kangxi reign [1701], I finished reading the thirty juan of the Xu Hanzhi in Shaobo’s boat. The master copy of the Hou Hanshu carved by the Mao family was far from good in comparison to that of the Hanshu. The boat was sailing so that I have no chance to collate with other editions. For the time being, I must wait and reread it [the Hou Hanshu] after going back to the South. Written by Zhuo. ĪȎ̉Ġ͞Ø-Ÿ=ʛ ̻ü	ɻǼņ‰“Ǖǖ	ļǼƝŨŸ#Ʀ̑̚ȜƝʛˀ˜ȈĿP9Ʀ/ƱêJŽǐk̻.  ȉ˪ From the twenty-third juan to this, collated with an incomplete Northern Song edition in the winter of the guisi year [1713]. ʒ,‰“ʓNjȶĠlƊľ‡ûǑƦƱ According to these records, we can know that He Zhuo read and collated the Hou Hanshu from 1701 to 1720, about twenty years. The spatial elements he mentions include Qingyuan, Baoding, the Imperial Palace, siting under an oil lamp in a study, and even reading on a boat. Once he had access to a new edition of the Hou Hanshu, he would borrow it and use it to collate the text of the Hou Hanshu in a copy of his own. It seems that He Zhuo never forgot to read and collate books at any time and in any place during these twenty years. He Zhuo’s biographies tell us that he was never appointed to a position of real political power or involved in any administrative practical affairs. He worked as an academic advisor in the Southern Study and as an editor in the imperial  181 printing office in the Hall of Military Glory for decades, devoting himself to reading and editing and occasionally supplying some opinions to the emperor and princes. He was a professional scholar who lived on a salary supplied by the imperial government. His job was to read and collate books, i.e., to produce new (reliable) texts. One of the most essential features of the scholarly culture of late Imperial China can be illustrated by He Zhuo’s case, that is, “to preserve the old while making it nevertheless new—to maintain continuity with a tradition without freezing it”—using an argument made by Daniel Boyarin about midrash.2  “To preserve the old while making it nevertheless new,” as emphasized by Boyarin, was a task of every successful culture.3 As discussed in chapter one, in Chinese history every generation of classicists proposed new ideas out of new interpretations of the Classics. Nevertheless, in the Qing dynasty, there were some new characteristics of the interpretation of ancient Classics, of which one of the most prominent is that scholars paid more attention to the text. Qing scholars were prompted to search for the Way (dao ̗), or recover the Way of antiquity, through carefully studying the text of the Classics.4 In the scholarly culture of the Qing dynasty, new ideas were generated from new interpretations of old texts; new interpretations came into being according to newly “collated” and “edited” texts. Catalogues and other sources show that, it was very common for scholars to collate, transcribe, and analyze texts and marginalia. There were many more textual experts in the Qing than in any other periods. This scholarly culture was created by scholars in the late Ming and early Qing period in a particular political and social context, guided by a particular                                                2  Daniel Boyarin, Intertextuality and the Reading of Midrash (Bloomington & Indianapolis: Indiana University Press, 1990), 22. 3 See Boyarin, Intertextuality and the Reading of Midrash, 22. 4 See also Ori Sela, China’s Philological Turn: Scholars, Textualism, and the Dao in the Eighteenth Century (New York: Columbia University Press, 2018), 4-5. Sela refers to this approach as “textualism.”  182 zeitgeist. It generally spread all over the empire, and was, perhaps unconsciously, accepted by scholars and practiced by them. When this scholarly culture was formed, it became an invisible power, shaping all scholars’ thoughts and practices. While devoting themselves to this textual enterprise, scholars’ ultimate goal was to search for the Way or recover the Way of antiquity, and build a better world. But they had to live in this world (cishi Nj) and deal with annoying, sometimes even dangerous, affairs. Still taking He Zhuo as an example, his colophons at the end of juan 16 of the Xu Hanzhi read: At the shen time [3-5 pm] on the twentieth day of the forth month of the bingshen year [1716], while I was going out from the Royal Palace, a heavy wind blew. The bolt of the Duan Gate was destroyed and the Wu Gate, opened wide, was seen in the distance. I was shocked by this so recorded it here. Ȭ¼Ɵ,‰ƊȬƓʒaȿo̜Ü͚¬ɘ̶ȔÔ‹̶̙ƣǤ̸ȱʈ˪# During the twenty-sixth to -seventh day of the sixth month of the gengzi year [1720], there were billions of turtles blocking the Lu River and going into the sea from Tianjin. Transport ships could hardly sail. [People] were seated in the barn and sacrificed a lamb and pig with music, [and] afterwards, they opened the sluice to send them out. In this way, they were gone. From the seventeenth to nineteenth days of the seventh month, it happened again. ĨïfƟıfƊʓƊ,ƊƠͭYʪƴÏȀǟȪÝǥ`ǩǻʜʓ ˀÊɤĮi ȕͮǂʔɋ̸Ȓ̎#éŁˉƟ‰Ɗʓ‰%Ɗ˜äƒ These two colophons recorded exceptional events: a damage caused by an odd wind and inconvenience brought about by billions of turtles. This can reveal some hidden feelings of He Zhuo. Damage to and disappearance of a gate bolt was long considered an omen of internal  183 disorder in Chinese history. The “Wuxing zhi” 0ˀņ (Treaties on the five elements) of the Hanshu states: In the first month of the first year of Emperor Cheng’s reign, the bolt of the Zhangcheng Gate of Chang’an disappeared; so did the bolt of the second gate of the Hangu Pass. Jing Fang’s Yizhuan (Annotation on the Book of Changes) states: “If [the government] announces that peace reigns over the land without relieving the famine, there will be flooding and the bolt of the gate will be gone.” The Yaoci (Demonic words) reads “The gate moving and bolt going missing are omens of malevolent ministers carrying out misdeeds, and traitorous ministers will usurp the throne.”  ţĢZįZĦ》Ɵ̵úɖÌ̶̶Ȕʒ1pˮ̽。̶Ȕ3ʒ14ŧ	ƏTƜ͜ʈŵʤ˥ǣ–ȄǙ–°Ȕ1	å̋Ɯ̽€Ȕ͛̊ȅ1̗ʐȅ͏–°)ʐˤɠ5 As an expert on the Hanshu and the Hou Hanshu, He Zhuo would be very familiar with this omen and its meaning. During the late period of the Kangxi reign, nine of his sons fought for the throne. He Zhuo was also involved in these events. In his lifetime, he was accused falsely and put into jail; after his death, he was defamed because he was once the tutor of the Eighth Prince and helped him fight for the throne. In this sense, this piece of marginalia manifests his hidden worry about his life and reputation. Scholars who participated in political affairs could hardly escape calamities and preserve their bodies and reputations whole. Was devoting oneself to antiques and hiding in books most scholars’ best choice? For He Zhuo, a scholar who worked in the royal palace and was                                                5 Hanshu, juan 27, 1401.  184 intimate with political figures, the answer seems to have been negative. But scholars sponsored by local government, merchants, or family resource had fewer misgivings of this kind. They concentrated on all kinds of scholarly affairs in different times and spaces. Let us first take Lu Wenchao’s case as an example. Lu Wenchao Ƚƃĵ  (courtesy name Shaogong Ÿij, 1717-1796) was one of the most prominent classicists, textual experts, and bibliophiles in the mid-Qing period. He collated lots of books and wrote and transcribed enormous amounts of marginalia. The Yili zhushu XɍǢȳ (Rites and Ceremonies with annotations and sub-annotations) held at the Shanghai Library (call number: XS797827-36) is filled with Lu Wenchao’s marginalia, which manifests his painstaking work in reading and collating this book. Lu’s work, especially his brief yet diligent studies, was influential for later scholars. At the end of the preface there are colophons written by two scholars in the late Qing, Huang Sidong ͫ¹ƪ (courtesy name Xiaolu Ďͥ, 1846-1910) and Huang Pengnian ͫĹĦ (courtesy name Zishou ï×, 1824-1890). Huang Sidong’s colophon reads: This is the original manuscript of Lu Wen’s Yili xiangjiao (Complete collation notes on the Rites and Ceremonies). Lu is a scholar of my home town. This book has ten volumes and seventeen juan in total. The date for reading and collating is at the end of each juan recorded by Master [Lu] himself. [Master Lu] devoted [himself] to this book for forty-four years, starting at the gengwu year [1750] and ending in the jiayin year [1794] of the Qianlong reign … [Master Lu’s] collation and evidential research were careful and refined, far beyond the capacity of recent classicists and scholars. I acquired this book from a bookstore in Chang’an. After reading it once, I felt like I possessed an extremely valued treasure. It will be good if my sons and grandsons can keep it for generations. Otherwise, I will send it to those who are able to read it to avoid its loss and dispersal. In so doing, I  185 can also be Lu’s meritorious servant. Written sincerely by Huang Sidong, a later scholar, on the fifteenth day of the second month of the dinghai year (the thirteenth year) of the Guangxu reign [1887].  Nj­̣Ƚijȑ\Ȧ	Xɍ˜Ʊ•Ʀ'Ɲn‰Ʀ·‰“ǔ“ƥȸƠ\ȦʒˑƱ̻ĦƊé(͂Ĩ‹ːȫ…Ȩ}Ƈʤʇn¼‰Ơ¼ĦʆƱɣă˞͏̍ɯȦ“Ũ™@ľ#Ƈ̵úġ ɫˬ̖äȗŰȢ­ïõʎù#ÀCªx;̝#ʎˬƒƝʇM^ƀA3Ƚǖ#~ʐ']ɴ‰Ħ2,ƟƣƊļ÷ͫ¹ƪ˧˪ Huang Pengnian’s colophon reads: Seeing Lu Wenchao’s careful and hard work expended on this book even when he was old, I hurriedly borrowed it from Xiaolu and transcribed [the marginalia] to show my respect. Written by Huang Pengnian a scholar of later days, in the eighth month of the thirteenth year of the Guangxu reign [1887]. ˌNjƦȨ}#ɣ„ʅʈOŋĿĎͥNʑ̖;ņ»Ļ]ɴ‰ĦdƟļ÷ͫĹĦ˪ In this book, Lu Wenchao recorded in detail the reading and collating time at the end of every juan. Owing to these temporal records and the enormous amount of marginalia, Huang Sidong and Huang Pengnian got the whole story of Lu Wenchao’s reading and collating of this book. Moved and motivated by Lu’s sincerity and diligence, they also devoted themselves to the transcription of marginalia. From Lu Wenchao to Huang Sidong and Huang Pengnian, what was transmitted  186 was not only research materials but also mental fortitude and faith that meaning resided in books and the study of the classics. All of this is encapsulated in the detailed records of time spent.  Much like Lu Wenchao, another prominent textual scholar in the Qing, Guan Guangqi ͙ĭÉ  (courtesy name Jianpin ȁ! , Alternative name Qianli Š̬ , 1770-1839) also habitually recorded the times he spent reading and collating books. The Huayang guo zhi ʨ́Ãņ (Chronicles of the state of Huayang) held at the National Library of China (call number: 6226) has a great many of He Zhuo and Gu Guangqi’s marginalia copied by an unidentified transcriptionist. The following table summarizes Gu’s colophons at the end of some juan:  Location Gu Guangqi’s Colophons juan 1 Reread in my residence in Jiangning, in the guiyou year of the Jiaqing reign [1813]. By Jianpin. ºŚȶ̦kˬƇǛćĄ   ȁ!ˑ juan 4 Collated in the tenth month of the guihai year of the Jiaqing reign [1803]. By Jianpin. ºŚȶ2‰ƟƱ ȁ!ˑ Recorded again after rereading in my residence in Jiangning in the guiyou year [1813]. ȶ̦0ƟǛćĄ kˬ˜ˑ juan 5 Reread in the fourth month of the guiyou year [1813]. ȶ̦¼Ɵ̭ˬ juan 7 Reread in my residence in Jiangning in the fifth month of the guiyou year [1813]. ȶ̦0ƟkˬƇǛćĄ  juan 8 Reread in my residence in Jiangning in the third month of the guiyou year [1813]. ȶ̦ƟkˬƇǛćĄ  juan 9 Collated on the twenty-seventh day. Jianpin. ıƊƱȁ!  Reread in Jiangning in the fourth month of the guiyou year [1813]. ȶ̦¼ƟkˬƇǛć juan 10 Collated in the tenth month of the guiyou year [1813]. ȶ̦‰ƟƱ  187 Location Gu Guangqi’s Colophons juan 10 Reread in my residence in Jiangning in the fifth month of the guiyou year [1813]. Recorded by Jianpin. ȶ̦0ƟkˬƇǛćĄ ȁ!ˑ juan 12 On the first day of the eleventh month. By Jianpin. ‰ƟƢȁ!ˑ By Jianpin in the light of the lamp on the twenty-first day in the guihai year of the Jiaqing reign [1803]. ºŚȶ2ıƊȁ!ēÕȏˑ After ten years, in the guiyou year [1813], I collated this book for Sun the Surveillance Commissioner in Jiangning. All things have their destiny like this. Written again. ̻‰Ħȶ̦ȅõˌąƱƇǛćn+ʒƠ‘ƂäNj˜ˑ Table 5-1  Gu Guangqi’s colophons in the Huayang guo zhi held at the National Library of China  This is a detailed record of his reading and collating history of the Huayang guo zhi, and of his special relationship with this book. This kind of record does not make much sense to readers of the main text. But for Gu Guangqi himself, recording the temporal sequence in detail made this book his “reading diary.” What was encapsulated within these temporal records were the events and emotional states he experienced at that time.   Weather and Art Some marginalia-composers and -transcriptionists also had the habit of recording the weather and other natural conditions they were experiencing during the time they read a book, such as He Zhuo’s record of the odd wind in the Hou Hanshu mentioned above. These records can reveal such things as their emotional state and attitude to scholarship. For instance, the Zhongwu jiwen «ɥʊ (Record of events heard in the Wu Region) held at the National Library of China (call number:  188 On the nineteenth day of the gengchen year of the Kangxi reign [1700], the snow ceased and the window was clear and bright. By Zhuo while blowing away the chill. ĪȎĨ̌‰,Ɵ‰%Ɗ͊͌ɔƎ®mƝȉ Interestingly, another copy of the Zhongwu jiwen at the National Library of China (call number: S2195) has He Zhuo’s marginalia transcribed by Wu Zhizhong «ņŇ  (alternative name Miaodaoren æ̗6, a scholar in the later era of the Daoguang reign). For this particular colophon by He Zhuo, Wu’s transcription lacks these last sentences. Yet at the end of juan four, Wu’s colophon reads: In the eleventh month of the renchen year of the Daoguang reign [1832] … on the nineteenth day, blowing away the chill. Miaodaoren. ̗]Ö̌‰Ɵ‰%Ɗ®mæ̗6 It’s difficult to know why Wu Zhizhong “forgot” to transcribe the last sentences. However, to be sure, Wu must have read the whole colophon and even been impressed by the last sentences, and then he imitated He’s expression in his own colophon.  The Sanguo zhi held at the National Library of China (call number: SB06264) is filled with He Zhuo and other scholars’ marginalia transcribed by Weng Tongshu. At the end, Weng’s colophon states: Finished reading amidst the wind and rain on the eighteenth day of the seventh month. Ɵ‰dƊ͚͉ ̻ɕ  189 The Wudai shiji at the National Library of China (call number: SB14710) has He Zhuo’s marginalia transcribed by Zhang Yu, as well as some marginalia composed by Zhang Yu himself. At the end of juan 22, Zhang’s colophon states: Collated on the eighteenth day, when the rainy season began and it suddenly became so cold that we needed to wear cotton-padded jackets. ‰dƊƱƳ͉͡ǭ ŀʵƷ˂ The Gengzi xiaoxia ji held at the National Library of China (call number: SB15338) has He Zhuo’s marginalia and colophons by Zhu Yun Ƨɞ (1729-1781), Yu Ji @ͅ (1738-1824), and Xia Huang’s Øȡ (jinshi 1809). Xia’s colophon at the end of the book states: Collated on the fifth day of the second month of the bingzi year [1816]. It was raining and a little cold. I tried the old ink Yeting sent me, writing [this] casually at the end of the book. The sparse raindrops knocked my window, and the plum blossoms in the vase seem to smile. ï,Ɵ0ƊƱčȮƒƊ͉Ýǘłǭ˘Ǭ5ŨŕžÓǽƝ“đĎ͉ͬȓȣƳDžə These poetic writings about the weather resemble the artistic description in Zhao Yi’s poem, “Busy collating in a curtain of books all day long. / Stepping outdoors, it is already in the dry autumn that I had not felt at all.” They manifest scholars’ aesthetic mental states when viewing the physical world and engaging in self-reflection within the realm of their studies. Textual criticism and evidential research in marginalia are both rationalistic scholarship, but Qing scholars treated them in a very perceptual and poetic manner. Comparatively speaking, in modern scholarship, it is  190 common in prefaces for books (and sometimes acknowledgments to articles) to take a much more personal tone than in the main body of academic writing. This predisposition in Qing scholars was shown not only in their writings of time and weather but also in the way they did their writing and treated their books.  He Zhuo was a well-known calligrapher and good at regular small style and semi-cursive style script. Books that had He Zhuo’s marginalia written by himself were thus treated as artwork and treasured by bibliophiles and calligraphers (see Figure 4-5). The Ming manuscript edition of Zhuozhong zhi ̧ ņ (Impartial records of Palace events) held at the National Library of China (call number: SB11564) has He Zhuo’s holographic marginalia. There is a short colophon written by an unknown collector on the cover, reading: Fifteen juan of the manuscript are still in existence. The seven comments on the top margin and the red characters in the interlinear spaces were written by Yimen’s own hand. They are exquisite and endearing. ñ•ū‰0“͓Ūʿ™ˀ̺ɦð"ʀ̶ˊɜɣɷ ř The Zhuozhong zhi has twenty-four juan in total. This manuscript copy has fifteen juan, i.e., more than half of the whole book. There are just seven pieces of He Zhuo’s eyebrow marginalia and several pieces of his interlinear marginalia. However, the collector emphasized these marginalia, which shows their special significance for him. In pre-modern China, people learned calligraphy by imitating previous calligraphers’ handwriting. He Zhuo’s students, later scholars, and calligraphers all practiced their calligraphy based on marginalia written by him. Sometimes, “transcribing marginalia” was also called “imitating (lin ʑ) marginalia.” In this practice, transcription of marginalia was not only a process  191 of transmitting research materials, but also a way to practise calligraphy. As mentioned in the last chapter, Jiang Gao, one of He Zhuo’s students, was known for transcribing and imitating his teacher’s marginalia. Similarly, Yao Shiyu ë̱ (courtesy name Yucai ș(, alternative name Yitian ʶȩ, 1695-1749), who personally learned from He Zhuo, was also skilled in imitating He’s handwriting while transcribing his marginalia. A copy of a Ming edition of Wudai shiji carved in the Mao family’s Jigu Pavilion held at the Shanghai Library (call number: XS847873-78, see figure 5-1) has He Zhuo’s marginalia transcribed by Yao Shiyu. Yao’s handwriting in regular small style script is quite handsome. Gu Guangqi’s colophon at the beginning of this book reads: This is a collated edition with Junior Compiler He Yimen’s marginalia transcribed by Mr. Yao Yitian. The Junior Compiler collated it with a copy read by Dongjianweng (Qian Qianyi) in the Jiangyun Tower. The original copy [with He Zhuo’s own handwriting] was formerly held at the Congshu Tower of the Ma family in Weiyang, and is said to have already been scattered to the north of the Yangzi river. What is recorded in the Dushuji [of He Zhuo] are just some important points.  Mr. Yitian once worked as a teacher for the Ma family. He transcribed He’s marginalia [into this copy] in detail without any omissions. There is no duplicate south of the Yangzi River. I expect readers treasure this book. Written by Pinzhou. Njëʶȩ\Ȧũūʀ̶?Þ¡ƱƦ'Þ¡Nɮ͋、ƪȁʁ̻Ʀͬ‘•ƝƐÇɲŴ͠ǖƝ、ʊğƀǐǛ‡	ˬƝˑŨ̅VŷǫȰʶȩ\Ȧ͝Ƈ͠ǖĿ•Ɲ̳o˜ĈȈ̝ÜǛ;ŽtȈ{Ʀˬʇhțɏ#!Ǧˑ   192  Figure 5-1a  First page and Table of Contents of the Wudai shiji, woodblock edition carved at the Mao family’s Jigu ge in the late Ming. Colophon by Gu Guangqi. Courtesy of the Shanghai Library (call number: XS847873-78).   Gu Guangqi stated how Yao Shiyu did the transcription and pointed out their value. Decades later, Zhang Yu transcribed all the marginalia from this copy into another copy (the one held at the National Library of China mentioned above). Zhang Yu’s colophon at the end of that copy states: The copy that bears Mr. Yitian’s transcription of He Yimen’s marginalia is now held at the Tianjin Library. 6  [The marginalia] are so exquisite that they are unrivalled in the world.                                                6 This copy is now held at the Shanghai Library. Its call number is XS847873-78.  193 In the seventh month of the renxu year [1922], I borrowed it and transcribed the marginalia into another copy in two days, which was done carelessly and for which I would be reprimanded by the wise men of the past. By Shizhi. ʶȩ\ȦũT?ʀ̶\ȦƦ7ʺÝǥÅƝ͝ ɣƁȈˆÖŢƟNTƦɗ,Ɗ#}+ʥȘȲȅz˶Ũ˓IJ# Figure 5-1b  Last leaf of the Wudai shiji. Colophons composed by He Zhuo and transcribed by Yao Shiyu in imitation of He’s handwriting. Courtesy of the Shanghai Library (call number: XS847873-78).     194 Some of Yimen’s marginalia [in this copy] are not included in the currently-circulating Dushuji. Yitian’s [Yao Shiyu] handwriting greatly resembles that of He Zhuo’s. Further noted by Yu. ʀ̶˪˝ȝˀ	ˬƝˑƠƤ™̅ʇʶȩƝ˿ƻ>?ǖ̱˜ˑ It was not rare in the Qing dynasty that a scribe, like Yao Shiyu, was capable of imitating He Zhuo’s handwriting to the degree that it passed as genuine. Generally, He’s students and most scholars would be perfectly happy to record in their colophons their transcription and all other encounters with that book in detail, with specific reference to temporal and spatial elements of their experience. The benefits of this practice—a habit widely practised in paintings, calligraphic works, and books—are manifold. It was a way to actively interact with the book, especially the text borne by this book. By so doing, they were involved in the history of the circulation of that book and even the history of the transmission of the text beyond the limitations of the physical book. This practice was a way to show not only that they once possessed and read the book, but also that they contributed to the text. In short, by writing in the book, a scholar made himself present with the text of the book over temporal and special span of its history. “Exquisite and neat” (jingzheng ɣƁ ), “exquisite and refined” (jinggong ɣĝ ), and “exquisite and good” (jinghao ɣã) are words that Qing scholars employed to praise the beauty of the calligraphy of marginalia. These words manifest the extreme care exercised by marginalia writers and transcriptionists while writing on a book, regardless of the calligraphic style. Transcriptionists sometimes needed to transcribe more than one kind of marginalia. In order to tell them apart, multicolour inks were employed—red and yellow being the most widely used colors. Therefore, expressions such as “eyes filled with red and yellow” (!ͫǶȾ) and “exquisitely and  195 carefully written in red and yellow” (!ͫɣ˧) often appeared in colophons. In this regard, for Qing scholars, the significance of a text lay not only in its content but also in its form. Scholars, bibliophiles, and calligraphers not only copied the contents of the text, they also tried to improve their calligraphic skills and incorporate some aesthetic values during this process. Transcription supplied new possibilities for using—not only “reading”—the text. It is worth mentioning that there were also lots of anonymous transcriptionists, who imitated He Zhuo’s handwriting but left no information about themselves, making it difficult to authenticate He Zhuo’s marginalia. As discussed in the last chapter, some of them were produced for commercial purposes. That is, merchants hired professional scribes to forge He Zhuo’s marginalia so as to raise the price of the book.  Transformation of the Textual Space of Marginalia “Textual space” here refers to the location of the text in space, including its location in a book or other media, and that media’s location in a particular physical space. People’s location can change; the location of the text can also be changed. Once the owner of one book was changed or the marginalia were transcribed from one copy to another, the textual space of marginalia would undergo a transformation, which would then cause differentiation of their features and significance. For instance, the Changgu ji ƍˮͅ (Collected works of Li He ƨ˵) carved in the Meicun Study (Meicunshuwu Ƴ̠ƝĔ) in the early Qing era held at the Shanghai Library (call number: XS799517-18) has He Zhuo’s marginalia transcribed in red and yellow ink by a Qing scholar, Chen Benli ̾Ʀɍ (courtesy name Jiahui ºŕ, alternative name Sucun ɨƩ, 1739-1818). He Zhuo’s colophon at the end of the book reads:  196 In the winter of the gengwu year of the Kangxi reign [1690] when I lived in the capital, I wanted to read [Li] Changji’s poems but did not have one [copy of his anthology]. Therefore, I bought this bad edition from the market. After reading several times, I could not bear to abandon it. Those living in later times, thinking of my difficulty in getting books, should derive from this a determination to study. Noted by Zhuo after twenty years. ĪȎĨ‹lĄ4ģDžˬ̵¤˙Ȉ#¾Ŀʌ ˲ľNjŖƦĕɯȾFŅƶ—ļ6ň@ˉƝ#͈͗…ņ§÷'ļ,‰Ħȉˑ As a matter of fact, “this bad edition” does not refer to Chen Benli’s transcribed copy, but to a copy of “that edition” that was once read and had marginalia written on it by He Zhuo and perhaps has already been scattered and lost. So, when reading this colophon, we should keep in mind that the adjective He Zhuo used to describe the quality of the book, “bad,” may not suit this new copy. At the same time, He Zhuo’s description and comments on the textual and physical features of the old edition may not suit this new edition, either.  Actually, while transcribing marginalia, transcriptionists usually made some adjustments to adapt the marginalia to the new edition. There is another copy of Changgu ji held at the Shanghai Library (call number: 764582-83), a printed copy of a Ming edition. It is filled with He Zhuo’s marginalia and emphasis marks in red ink, and the quantity of marginalia is greater than that in Chen Benli’s transcription. Moreover, the handwriting resembles He Zhuo’s. If this is the real original copy that He Zhuo read and in which he wrote marginalia, we then can know that Chen Benli cut out some contents and changed the color while transcribing. If this is only another transcription, either the marginalia on this one or that transcribed by Chen Benli was modified in content and form. Or, more likely, both of them were changed in some way.  197 Generally speaking, marginalia transcribed by professional scribes hired by scholars, bibliophiles, and merchants were much closer to the original copy. For instance, the marginalia on the Hou Hanshu held at the Peking University Library, that we discussed in the third chapter, were transcribed by three scribes. There is much meaningless content and lots of instructions on composing rhymed couplets and poems in the voluminous marginalia. These contents are in line with the characteristics of He Zhuo’s marginalia in their original form. Except for their handwriting, scribes never left any personal information. Their self-effacement causes an absence of temporal and spatial records, and thus an absence of a particular time and space associated with the text, and therefore ambiguity in the relationship between people and people, people and the text, and text and the text.  Few scholars would copy all the marginalia into a text without making changes by their own hand. Some scholars only transcribed the collation notes and ignored all the comments; some would merely transcribe the comments. As for transcribing the collation notes, when the new copy and the old copy were of different editions, the variants, collation notes, and description of the layout would vary. Besides, the color and calligraphic style could also change while transcribing. In addition, transcriptionists would alter the location and color of the marginalia based on the physical forms of the new edition, and select the contents of the collation notes and comments according to their personal academic tastes. All of these changes were often described in the colophons. For example, the content of the Geshi bian Lj˙ɶ (Anthology of poetry), another anthology of Li He’s poetry carved by the Mao family and held at the National Library of China (call number: SB14622), is almost the same as that of the Changgu ji, so much so that Zhang Yu directly transcribed He Zhuo’s marginalia on the Changgu ji into this Geshi bian. Zhang Yu transcribed the marginalia from a copy transcribed by Deng Bangshu, who transcribed them from  198 a copy transcribed by Chen Ge. After several generations of transcription, the text of the marginalia transcribed by Zhang became rather complicated, consisting of layers of information accumulated by different scholars. In the colophon at the beginning of the book, Zhang Yu described the changes he made while transcribing: In the sixth month of the renzi year [1912], I borrowed Deng’s copy to transcribe. The previous history of transcription and transmission is stated in detail in Deng’s colophon above. Deng’s copy is an edition that has Xu Wei and Dong Maoce’s commentaries and was carved in the guichou year of the Wanli reign [1573]. This copy is considered one copy of the Kuaiji edition cut by the Mao family as claimed in Deng’s colophon. If there are any differences between these two editions, I have recorded “carved in a certain way in a certain edition” at the bottom [of this Mao edition]. If the collation by Mao and He, both according to the Song edition, are the same, I do not record it. If the characters [in this Mao edition] are different from that of the Song edition, I record He’s collation. He’s collation notes always state that “the Song and Jin editions write XYZ.” Since this Mao edition was carved based on a Song edition, I changed He’s expression into “the Jin edition write XYZ” to differentiate them. The marginalia in this book was transcribed many times. The editions were different from each other. The marginalia must not be in line with He’s original version, but it cannot be said that this is not a good copy of the Changgu ji. By Zhang Yu of Changzhou in the rented Tinghe House in Jinmen on the first day of the seventh month. ÖïfƟN̥Ʀ̖̳;zTƱǵǧ˜̥˾̥ƦGƎʪƚȶrĽDzʭşɝŪ˔Ʀ’NjƦǕ˾Ũ˥ƞɑƦ'ʖNjƦ¥ʿǢƎƮƦƮƇƆǕŸûr?ŸûƱ¥ʇtoʖû¥x8Ÿ?Ʊ̳`?Ʊǔ.û̯ƦBƮ;NjƦƉoûƦxźBƮ̯Ʀ¥;t#NjƝ̇̈T̳ƫƦ˜¢¢¥ 199 Ƈ?ǖƦD͐ȾńƠɚʿȊ3 ˥͏	ƍˮͅ#µƦɄƟƢ̵Ǧɖ̱ˑƇǥ̶ʋͩWʘ The edition He Zhuo read and in which he wrote his marginalia was unknown, but He mentioned a Jin edition and a Song edition and recorded all the variants from them into a Ming edition. Zhang Yu used a Mao edition to transcribe He Zhuo’s marginalia. This Mao edition was a reprint of the Song edition He mentioned. For this reason, Zhang Yu did not transcribe the variants of the Song edition. According to Deng Bangshu’s colophon, Deng used a Ming edition to transcribe He’s marginalia. Zhang Yu recorded the variants of this Ming edition, too. Since it combined character variants from various editions, Zhang’s copy became a very good and special edition.  As Zhang Yu had done, scholars and bibliophiles in the Qing dynasty usually transcribed the marginalia of more than one scholar into one copy. In order to distinguish them, transcriptionists employed various strategies. Sometimes, they marked the origin at the beginning of every piece of marginalia. Most of the time, they used different colors to differentiate them. For instance, the Li Yishan shiji ƨʀė˙ͅ (Collected poems of Li Yishan) printed in the early Qing period and held at the Shanghai Library (call number: 6889) has He Zhuo, Chen E ̾ě, Ji Yun ɥƌ, and some unknown scholars’ marginalia respectively transcribed in red, green, blue and black (see figure 5-2). The colophons in the book state: The marginalia in black ink were composed by Chen E, courtesy name Zuofeng. / The marginalia in green ink were composed by He Zhuo, courtesy name Qizhan, alternative name Mr. Yimen. Óɜʨ5̾ęĜěŪ/ɰɜ?ĘɃȉʀ̶\ȦŪ The marginalia in black ink were composed by Ji Yun, courtesy name Xiaolan.  200 ɥƌƛĜɩɜ Sometimes, the transcriptionist also transcribed multicolor marginalia onto a new copy in a single color, usually black. For instance, the Suxueshi ji ʼ÷Õͅ (Anthology of Academician Su, Su Shunqin ʼʚdž 1008-1048) printed in the early Qing period held at the Shanghai Library (call Figure 5-2  Leaves 5b and 6a of juan 1 of the Li Yishan shiji, woodblock edition printed in the early Qing. In this copy, He Zhuo, Chen E, Ji Yun, and an unknown scholar’s marginalia were respectively transcribed in red, green, blue and black ink. Courtesy of the Shanghai Library (call number: 6889).     201 number: 753558) has Gu Guangqi’s marginalia transcribed by the well-known bibliophile, Huang Pilie ͫȆ (courtesy name Shaowu ɬǍ, 1763-1825). Gu’s colophon reads: The unknown compiler’s Lize jishi has 35 juan in total. It is a Song edition. I collated [this Suxueshi ji] according to [Su’s poems] recorded in it and wrote the variants in black ink. Jianpin by the light of the lamp on the twenty-fourth day of the ninth month of the yimao year [1795]. Ȉ¦ǖ	ͪȂͅ˙n‰0“ûǁ'ĐhŨ̅Ʊ#;Óɜȅt$%Ɵ,‰¼Ɗȏȁʸˑ After this, Huang Pilie’s colophon states: This collated edition was made by my friend Jianpin. The marginalia in the whole book are in red ink, while what was collated according to the Lize jishi is all in black. In my transcribing of it, I changed the red into black. What was originally in black I labeled with the two characters, “Lize.” If the characters in the Lize and He’s collation are the same, there were black circles at the characters in the original copy. I followed this pattern without any change. If the characters [in the Lize] were the same as the main text of this printed edition, there were also black circles at the characters. I followed this, too. As for characters in the Lize that were different from that of He’s collation and this printed edition, I recorded them as “X is written into Y in the Lize.” Written by Huang Pilie, who recently mourned his parents, in the Study for Reading Books not Seen Before in the sixth month in the summer of the thirtieth year (the wuwu year) of the Jiaqing reign [1798]. NjƱƦȅ@šȁʸŨT̳̏͢ȸȨɇɜ²ŨƱ	ͪȂͅ˙ȸ;Óɜ@ʑƱźɇɜȅÓɜƇ•ƦÓɜʇȸ;ͪȂ,ðt#	ͪȂʖ?Ʊ¥ʇ• 202 ƦȸƠÓÂ7Ŏ8#ƠʖNjw¥ʇ3ƠÓÂ738#Ŕ	ͪȂʖwƱȸ¥ʇ7ź͖ƜͪȂƮBƮȇºŚ‰Ħš‹ØfƟƸ6ͫȆ˪ƇˬƤˉƝͯ From these colophons, it can be known that Gu Guangqi transcribed He Zhuo’s marginalia and then recorded in black ink the character variants of Su Shunqin’s poems taken from a Song edition of a poetry anthology. While transcribing Gu’s marginalia, Huang Pilie changed it from red to black and labeled the variants from the Lize jishi with “Lize” in order to distinguish the two kinds of marginalia. There is another copy of Suxueshi wenji in the same edition held at the Shanghai Library (call number: XST02612-15) bearing marginalia in red, blue and black. The red marginalia are a transcription of He Zhuo’s marginalia. The composer and transcriptionist of the black marginalia is unknown. The blue are collation notes made by Ye Jingkui ʬƗʮ (courtesy name Kuichu ųs, alternative name Juan’an “ȼ, 1874-1949) based on a comparison of his own copy with Huang Pilie’s transcription of Gu Guangqi’s marginalia. Ye’s colophon reads: In the past, I got a copy [of the Sushunqin ji] in the Baihuashuwu edition, which has He Zhuo’s marginalia transcribed in red ink. There were lots of erroneous characters in He’s collation. There were also three pieces of textual criticism in black. The writer did not sign his name. It was not composed by He Zhuo. In the late spring of the wuyin year [1938], I managed to borrow a collated copy from my old friend Pan Jiru. It was collated by Huang Pilie who transcribed Gu Qianli’s transcription of He Zhuo’s collation notes. He Zhuo also collated the poems according to a Song edition of the Lize ji. I collated my own copy with that, and recorded the variants in blue so as to differentiate them from the old red and black marginalia in this copy. I corrected all the errors and filled in all the dropped characters.  203 ʗľȷʨƝĔƦƠƧɜT̳?Ʊ͕Ơ˩ð˜ƠÓɜƱ˝ƵƤɾ¦͏?Ʊš…ƑƙPľʅšǾôøŨʺͫʴÁT̳͙Š̬ʑ?ƱƦ˜;ûr	ͪȂͅƱ˙¾čƱ̖ƼȨʹɜ;t-ʗƠ#ƧÓɜn˒áʿŎʖź》 It was both for correcting the mistakes of the main text and then making their own copies “good editions” (shanben µƦ, here referring to an edition that is well collated and has reliable texts) that Ye Jingkui transcribed He Zhuo and Gu Guangqi’s marginalia, and that Zhang Yu transcribed Deng Bangshu’s transcription of He Zhuo’s marginalia after several other scholars’ transcriptions. A well-known late Qing bibliophile, Jiang Fengzao (courtesy name Xiangsheng ͟Ȧ, ca. 1845-1908) once transcribed various scholars’ marginalia on the Dushu minqiu ji ˬƝżǚˑ (Notes on reading and diligently in investigating the past), an annotated catalogue composed by a celebrated early Qing bibliophile, Qian Zeng. While transcribing various marginalia, Jiang not only employed multiple colours but also differentiated them according to calligraphic style (see Figure 5-3).7 This exquisitely transcribed edition, bearing various collation notes and comments on the main text, is of course a “good edition.”  All these “good editions” produced from the transcription of marginalia were actually “established” by multi-layered texts and multiple scholar’s opinions in different temporal and spatial dimensions. As mentioned above, the marginalia culture was a hybrid of print culture and manuscript culture. When marginalia were transcribed from one printed copy to another, their textual space changed, so the relationships between marginalia and the main text, and between marginalia and the physical book that bore the main text would also change. With the                                                7 See Zhongguo guji gaochaojiaoben tulu  Þɢɒ̰ƱƦÅ̳, Chen Xianxing ̾\ˀ, et al., eds. (Shanghai: Shanghai shudian, 2014), 909-910.  204 transformation of the textual space, the contents of marginalia would be altered; the color, style, location on the page would also be adjusted to adapt to the new textual environment. These changes Figure 5-3  Jiang Fengzao’s transcription of various scholars’ marginalia on the Dushu minqiuji. In his transcription, Jiang not only employed multiple colors but also used different calligraphic styles to differentiate different scholars’ marginalia. Picture from Chen Xianxing, etc., eds., Zhongguo guji gaochaojiaoben tulu (Shanghai: Shanghai shudian, 2014), 909.    205 seem minor, but they are an important part of the evolution of the text. Cumulatively, these minor changes can produce huge effects: they directly determined how the text appeared to a reader, and thus influenced how the text could be read and used. For a scholar in the Qing dynasty, the quality and quantity of the marginalia, especially collation notes, determined the depth and breadth of his study to some extent. For a common reader, “seeing” a book filled with eye-catching writings and reading the main text along with other readers’ erudite and interesting comments must have given them a very different reading experience. Texts are created in a particular context; in return, texts also create a particular context. The accumulation of marginalia changed the context whereby the main text made sense, and thus transformed the meaning and significance of the main text. Marginalia, their transcription, and other scholarly practices involved offer a key to accessing Qing scholars’ reading habits and intellectual world.  Scholars’ Mental World Sometimes, the marginalia writer’s brush reached beyond the space of his study to the much broader external world, revealing the extension of his gaze from one room (yishi ’) to all under heaven (tianxia Ý). Outside of the study, there was not only the objective natural environment but also a variety of more complicated human affairs. For instance, the Nanshi Ž¡ (History of the Southern Dynasties) held at the National Library of China (call number: SB04309) has He Zhuo and Wang Mingsheng’s marginalia transcribed by Liu Lüfen |Ėʟ  (courtesy name Yanqing ĸ͍, 1827-1879). There are three colophons composed by Wang Mingsheng at the end of this book. The first states that he read this Nanshi and wrote some marginalia in it in 1772. The second and third colophons read:  206 On the twenty-third day of the first month of the thirty-eighth year [of the Qianlong reign], the guisi year [1773], I transcribed the comments from other copies again, starting from the thirty-fourth juan. This work was finished in the afternoon of the second day of the second month. In the morning of that day, another daughter of mine was born, when I was fifty-two sui. ‰dĦȶĠ》Ɵ,‰Ɗ̭ʑˣƦŪ:ʒ‰¼“˻ʓ,Ɵs,Ɗ‹ļȮĝƒƊǰƖ˜ȧâƓ*Ħ0‰Ơ, From the eighth month of the renchen year [1772], my two concubines, whose surnames are both Chen, went back home. Their fathers harbour malicious intentions in their hearts. I was depressed. What was worse was that I sold houses at a loss in the eighth and ninth months, which wasted a lot of money and energy, beyond expressing. Worse yet, my son was stricken by a serious disease and the money I spent was incalculable. In that half year, I was almost buried by depression, so that I neglected my enterprise. In the spring of the guisi year [1773], when I consented to my fourth daughter’s marriage to the Yao family, I was still upset. I snatched a moment of leisure and finished collating this book. Amidst all this chaos, I did not stop learning from the ancients. Perhaps it is because this is what I love. ʒÖ̌dƟc̾è˜ʼn½“—hȑ͕ŠŖŃ*ŒɴƉC˜d%cƟ̐。˴˷ŧĔ˳˱ƒ͔ ‚ˍ˜_¹ǀȴ‘̫ʻ#˳˕Œ̅# ƊÇŗÌ̔3ĬƺȞȶĠsƑƓƆ̛ìɛ¼â-ëǖŒɴ3ȍŶR̹ƱȮNjƝvH# ̆ɑžʰ*#ŨãÇNj'  207 Wang Mingsheng Țͧȹ (courtesy name Fengjie ͦ¶, 1722-1797) was one of the greatest historians and classicists in the mid-Qing, who took second place at the palace examination (bangyan ƾɂ) in 1754 and was the author of the masterwork, Shiqishi shangque ‰¡³ƿ (Discussion of problems on the Seventeen Histories). Among the enormous number of objective historical comments in his marginalia in the Nanshi, these two colophons stand out: the former recorded the event that he had a daughter at the age of 52 sui; the latter recorded various annoying family affairs. During all these pleasant or uneasy times, Wang never “stopped learning from the ancients.” At the very moment when his brush touched on external matters, it withdrew to his study where he could find peace and happiness. Here, learning and reading were not only an job or hobby; they became a mental pursuit, a habit, and an important part of his daily life. Reading books, collating them, and many other scholarly practices played a crucial role in his life.  Another relevant example of such movement between external spaces and the spaces of scholarship is the case of Weng Tonghe ʁ¥ͱ (courtesy name Shuping ›ĥ, 1830-1904). The Han wen chao held at the National Library of China (call number: 06252) has Weng Tonghe’s transcription of He Zhuo’s marginalia. At the end of this book, Weng’s colophons read: On the sixteenth day of the second month of the wuwu year [1858], I saw a copy of Han Yu’s classical prose with Li Rongcun’s [Li Guangdi] marginalia. I borrowed it and transcribed the marginalia. There were just a few comments by Li. Most of them were He Yimen’s comments. At that time, my wife had been ill for several years and her breathing was feeble. I was reading by the light of a lantern at night and my mood unbearable. Remote in time and space, who could understand my sadness! Finished on the twentieth day and recorded here. By Tonghe.  208 š‹,Ɵ‰fƊƇīʌˉƨƽƩ\ȦŪƦ͑ƃPǐʑƱ̖ƨǖˠ̖ƂƵÜŭʀ̶?ǖ#ˠēÙƓ@çȴȵɪĦʓƒǘVäɹɟȏÚˬŘÑÎŏŏNj öɅ@Ő',‰ƊʑȮ¾˪¥ͱ In the ninth month of the wuchen year [1867] of the Tongzhi reign, I escorted home the coffins of my deceased father whose posthumous title was Wenduan, and my deceased older brother whose posthumous title was Wenqin. We started south of the Lu River and stopped in Linqing. There was no water in the river there, so we took the land route thenceforth, and returned to the boat again in Zhangqiu. I read and punctuated this book in great sorrow and depression. I took another boat carrying my wife’s coffin and watched from the distance the limitless mist covering the water. Recorded by Tonghe, on the lake at Mount Wei on the fourth day of the month. ¥Ǡš̌%ƟŬ˫\ƃɘe™\[ƃ„#¸ȪȀǟŽ̟ʑǰȈǙ"òŁʒĶ`ʛśUŜő ͬˬ̖t;Ďʝ̅》çưɀƣƇȋǡǨdz̺'ƒƟ¼ƊłėǴ ¥ͱˑ These two colophons record domestic tragedies Weng experienced. In this depressed mood he could only take slight comfort from reading and put down his sadness in the margin of the book he was reading.  Marginalia’s reach beyond the space of the scholar’s study was not always into a space of grief. The Jinshilu ̯Ɇ̳ (Catalogue of ancient bronzes and stone tablets) held at the Shanghai Library (call number: XST03117-122) has a colophon at the end by Weng Tonghe, stating that he was fortunate to acquire this book in a book market and that He Zhuo’s marginalia in it were  209 transcribed by a noted scholar-official and calligrapher, Dong Chun ʭ̪ (1810-1892). After this, another colophon by Weng reads: In the past, my friend Pan Boyin (Pan ZuyinǾɉʳ, courtesy name Boyin =…, studio name Pangxi Study [Pangxi zhai], 1830-1890) got an incomplete copy of a Song edition formerly held at the Yeshi Garden [Yeshiyuan]. Pan was extremely surprised and considered it a very rare and valuable treasure. He carved a seal, reading “the family that has ten juan of the Jinshilu,” and invited friends to drink and compose poems for it. Pan declared, “The lost treasure stopped wandering.” This certainly deserves our cheers. Noted by Pingsheng. Ɛ­šǾ=…ľ'ƒÄûǁǑƦ˚ȅàĊw̯Ɇ̳‰“6“ϐɽ̨̞6˸˙Ƿ·.1͚ǧLJɭ ‚《±ȣȦˑ Pan Zuyin was a high-ranking Qing dynasty official, a noted art collector, and a prominent calligrapher. The Jinshi lu, co-written by the famous epigrapher Zhao Mingcheng ˼Ǝ˞ (1081-1129) and his wife, one of the great female poets Li Qingzhao ƨǰȌ (1081- c. 1141), was considered one of the earliest and most important catalogues and study works on ancient Chinese bronzes and stone tablets. It enjoyed a great reputation from its publication in the Southern Song dynasty. But in the Qing dynasty, scholars believed that there were only ten juan of the Song edition (one third of the whole book) left. This incomplete Song copy (can Songben ǑûƦ) thus enjoyed a very special reputation. Every collector who held it would carve a seal reading “the family that has ten juan of the Jinshi lu” (̯Ɇķ‰“6“), and so did Pan Zuyin. His seal was carved by Zhao Zhiqian ˺#˭ (18291884), a renowned calligrapher, seal carver and painter  210 in the late Qing.8 Here, Weng recorded this anecdote, reflecting his admiration for his friend Pan Zuyin who held a Song edition, albeit incomplete, of the Jinshi lu, his regrets that most rare books were scattered and lost, his happiness that Pan stopped this incomplete Song copy from wandering, and his satisfaction from managing to get his own copy filled with previous scholars’ marginalia. In their colophons, scholars did not talk much about theories, principles, the Way of former sages, or any kind of concrete knowledge. What they recorded was their life style, various practices with books, and their attitude toward these practices. Through transcribing previous scholars’ marginalia, the colophons in particular, they learnt from their predecessors a way to do their research, a way of life, and a way to think about their lives. They inherited from their predecessors not only concrete knowledge, but also faith in—or at least an attitude toward—books and ancient texts. That is, they were living in a particular kind of culture, shaped by it and maintaining it. I am not arguing that one particular cultural form will continue in a society without any changes. On the contrary, I’m trying to shown that the condition of a culture is influenced by political, social, economic, and other factors and undergoes a very slow, but certain, change.  In the late Qing and Republican periods (during the late nineteenth- and early twentieth-century), when China was going through a transformation from pre-modern to modern, the scholarly culture of the Qing dynasty was still intact. Many scholars continued the scholarly practices of their predecessors. They occupied themselves with reading and collating ancient books, transcribing marginalia, and discussing books with friends. Zhang Yu was one of them. The Wudai shiji held at the National Library mentioned above has a colophon by Zhang at the end of juan seven, reading:                                                8 See Xu Ke , Qingbai leichao, 4246; Pan Zuyin Ǿɉʳ, Pangxi zhai cangshuji Ƿ·ͯʺƝˑ, in Xuxiu Suku quanshu, vol. 926, 429-435.  211 In the sixteenth day, we held a celebration for my little son Yuanyi’s first full month. I had my son’s hair cut and talked sweetly with my relatives. I snatched a moment of leisure and collated this book. ‰fƊɐïZʀǹƟʃͤʖˊŦŒ?ŮƘƱNj The colophon at the end of juan 62 states: Collated for thirty-three days. I did not do anything else or contact any friends, so I read and collated a lot. ‰ƊƱǠ9+ơĿɭ˿ŻƱˬ̄Ù While busy with family affairs, he competed against time to read and collate; in his leisure time, he read and collated more. When he was happy because of his son’s full-month celebration and the chatter with his relatives, he did not forget to read and collate; in a lonely time when he cut off the contact with his friends, he devoted himself to reading and collation. At the end of juan forty-two, another colophon reads: Collated on the twenty-first day, which was my forty-ninth birthday, twenty-seven years after my father’s death, seven years after my mother’s death. Ten years ago, I became an official. On the same day, when I lived in the capital, I went to Father Wang’s and tried to write a policy essay. Another decade earlier, I was in my home (in Suzhou). [My son] Yuanshan was less than one-year-old at that time, but now he has already gone to the U.S. to study and will graduate soon. Another decade earlier, I was in Xumen (in Suzhou) and studied in an elementary school named Sixian. One more decade earlier, I learned to read the Mencius from master Ding. Recalling past events, I felt everything vividly [in my mind]  212 as if there was a painting. Qu Yuan (courtesy name Boyu) a senior official [of the Wei in the Spring and Autumn Period] said that when he was fifty sui he knew the mistakes he made at forty-nine. I will say that what I did in the last decade was perhaps right, in comparison with what I am doing now. About the reasons therein, with whom can I talk? ,‰ƊƱƒƊȅ@¼‰%ǎɌĩ—\¨#1,‰Ħ\Ǔ#1Ħz‰Ħȅ̏ɢ#ĦƒƊĄ4̘ïȐǝʿ˘ĉɝƦ˜z‰ĦÇ̶̬ƒƓZµďƤ˘̒7ğ̓̚ɿÃǚ÷ċ。Ȯƺ˜z‰ĦƓÇʍ̶Ȫƅ̲ʿʯ͝˜z‰ĦĿģˬ	óï½ƣzÐǏǏäɺʾÜß¦Ƞð=șƑɎƓˁÃÜʐ˥Ʌ¼‰%Ħ#͏@x˥;7Ʊ#zƂ‰Ħť̄ƒʉNj Ǫōċʖˡ˝̡ Here, on his forty-ninth birthday in 1913 when he lived in Tianjin,9 Zhang Yu looked back at his life in segments of ten years. In these forty-nine years, Zhang moved from his home to Xumen in Suzhou, then to the capital (Beijing), and then to Tianjin. In any particular time and space, reading and learning were always cornerstones of Zhang’s life. For a reading seed like Zhang Yu, external time and space transformed, but his reading life did not. Nor did his relationship with books.  Two years before 1913, the Xinhai Revolution ended the Qing dynasty, the last empire in Chinese history. The new regime was established. In this colophon, Zhang Yu also mentioned his son going to the U.S. to study. The world had changed, but he was confused. Perhaps it was time for him to transform how he saw the world, perceived the world, and responded to it. But Zhang continued to read the world according to texts that had been read and altered by past wise men. He was defending the tradition in his way. His colophons, sometimes, are inundated by confusion,                                                9 See Su Jing ʼɣ, Jindai cangshu sanshijia ̍:ʺƝ‰“ (Beijing: Zhonghua shuju, 2009), 46.  213 bewilderment, and depression about the world and his destiny in it. At the end of juan fifteen of the Wudai shiji, Zhang recorded discussions with one of his friends about one of Su Shi’s ʼ̃ (1037-1101) song lyrics, the “Shui long yin (Ci yun Zhang Zhifu yanghua ci)” ǙͰ©。͒ɖ˹ßƹʠ; (Tune: “Water Dragon’s Chant,” After Zhang Zhifu’s Lyric on the Willow Catkin, Using the Same Rhyming Words). This colophon reads: Collated on the seventeenth day. That night, Shi Zhonglu came to discuss Pogong’s [Su Shi’s] song lyric, the “Willow Catkins” (Tune: “Water Dragon’s Chant”) with me. He said: “The two sentences that start with ‘bu hen’ (do not grieve) in the second part mean that the willow catkins are not worthy of pity, yet what went with them is to be greatly pitied.” When I ask him why, [he said that] the willow catkins and what were not willow catkins both returned to dust. The six sentences from “When dawn comes” to “flowing water” have the meaning that the myriad things all perish together. This [reading] grasped the profound meanings [of this poem]. In my opinion, the first part of this poem talks about this life, the second part discusses this world. These sad words throughout all ages find someone to corroborate them. The second day when I got up in the morning, I briefly sketched this. ‰ƊƱÚƈ<ͥDʖˢËe	ǙͰ©ƹʠ̼DZ.ŒŌ,˝˥ƹʠ˽œ̈́;L1ʇ"Ü œ™´hɓɕxƒƹʠ͏ƹʠ¥ǐƇÐÆǧǙƛDf˝Ơʪ͘¥ǐƇȺ#ŘƻľłƋ@ƇNj;ǔ˥ŒȅŤȦˠŒȅȭˠŠžUŃ#̋Ơ6ȅ#˨ʂƊƖ˻ŹƝȅˑhȰ Su Shi’s entire lyric reads:  214 It seems to be a flower, yet not a flower, / and no on shows it any pity: let it fall! / Deserting home, it wanders by the road; / When you come to think of it, it must / have thoughts, insentient as it may be. / Its tender heart twisted by grief, / its delicate eyes heavy with sleep, / about to open, yet closed again. / In its dream it follows the wind for ten thousand miles, / to find where its lover has gone, / but then it is aroused by the orioles’ cry once more. >ʠ̟>͏ʠ'Ȉ6œĿžÒů“S̀Ŋ̮”ƒȈŒƠŊɸŵƯʏ¿̩îɂDž̸̷̟Û͚̈́ʪ̬Č–—ʿ˜̟˃ͨ¯˻ I do not grieve that the willow catkins have flown away, / but that, in the Western Garden, / the fallen red cannot be gathered. / When dawn comes and the rain is over, / where are the traces they have left? / A pond full of broken duckweeds! / Of all the colors of springtime, / two-thirds have gone with the dust, / and one-third with the flowing water! / When you look closely, / these are not willow catkins, / but, drop after drop, parted lovers’ tears.  ŌNjʠ͛ȺŌˇÄʫɦ͈ɳƛD͉̖̝́?ÇǜʩɈƑʞq,qÐÆqǧǙɫɁDƒƹʠͬͬƒ͇6Ǯ10                                                10 Su Shi ʼ̃, “Shui long yin” (Ci yun Zhang Zhifu yanghua ci) ǙͰ©。͒ɖ˹ßƹʠ;, in  Zou Tongqing ̤¥Ś and Wang Zongtang ȚýÍ, Su Shi ci biannian jiaozhu ʼ̃;ɶĦƱ˔ (Beijing: Zhonghua shuju, 2002), 314. Translation by James J. Y. Liu, in Sunflower Splendor: Three Thousand Years of Chinese Poetry, Wu-chi Liu and Irving Yucheng Lo, eds. (Garden City, New York: Anchor Books, 1975), 349-350.  215 The first paragraph of Su Shi’s song lyric writes that the willow catkins seem to be have no feelings, but they actually have deep thoughts. The second stanza states that the willow catkins, as well as all the flowers, have gone with the passing of spring, which triggers in Zhang the feeling that “the myriad things all perished together.” Zhang thought that “the first part of this poem talks about this life, the second part discusses this world,” which implies that everyone under heaven will go to an end in the political and social transformation at that time, when the order of the world was lost, even though one still has affection for the world, people and the myriad things. This kind of helpless, confused and depressed feeling is vivid portrayal of the minds of many “reading seeds” of that time.  These reading seeds, who had carefully read the Confucian Classics and other ancient works, more or less possessed a mind of romanticism. They dreamed of building up a utopian society and renewing the glorious customs recorded in ancient texts. In their opinion, all the principles of the natural world and human society were encapsulated in the classics, so that they could understand and establish a relationship between themselves and the external world according to the doctrines in those classics. They derived knowledge from various texts and tried to apply it to the practical world. However, there was always a great gap between the ideal and the reality, between what ought to be and what was so. Because of this great gap that could hardly be crossed, they withdrew from the external world to their study and tried to rebuild their ideal world in texts. This is the faith they inherited from their predecessors and would hand down to later generations. They accepted this faith, and the practices of enacting this faith, out of a kind of cultural inertia; i.e., they grew up in that culture, lived in that culture, and were shaped by that culture.  The text had rational contents and artistic features. In this regard, what was built up by the scholars was actually a rational and artistic intellectual world. In textual criticism and evidential  216 research, scholars rigorously complied with rational principles. Their works relied on observation and rational deduction. This “textual rationality” was never truly and successfully applied to the real world, yet it supplied a mental realm to which scholars could return, a secret getaway (taohua yuan Ʋʠǵ, lit. a Peach Garden, referencing the story of the same title) established in the scholars’ collective imagination. Therefore, in the real world, they chose an artistic life style: they had a poetic eye for all the things in the natural world; they appreciated and held on to books, and all the objects in their studies as well. They paid a lot of attention to the calligraphy of marginalia, wrote poems to eulogize rare books, and more. Generally speaking, some Qing scholars leaned toward the rational and practical aspect of their mode of reasoning; some leaned toward the artistic aspect, but neither group abandoned the other side completely. They managed to find a balanced point in between. Even when building up their ideal world in the text, their purpose was never to simply “finish” a project (compose a book or article). The scholarly practices themselves mattered. The process was more important than the result; the experience and the comprehension were a more important part of the purpose. Scholarship was not only their occupation, but also part of their life. These “reading seeds” had a kind of rational aesthetic personality. On the one hand, they had gained the capability and inclination to pursue the truth via extensive reading and training in how to be a scholar (critical thinking within the classics); on the other hand, they pleased and/or comforted their hearts by ways of finding and appreciating the beauty of books and the practices of their scholarly life. The Zhongwu jiwen held at the National Library of China (call number: S2196) has He Zhuo’s marginalia transcribed by Wu Zhizhong «ņŇ  (alternative name Miaodaoren æ̗6, abb. Daoren). After the table of contents, there is a colophon composed by Wu, reading:  217 [I,] Daoren have collated this book according to four copies: an old manuscript copy collated by Mr. He Yimen and transcribed by Mr. Shen Qiutian: a copy of the Mao edition that was collated by Mao Zijin and Lu Chixian; the copy that was collated by Wuyuan according to a Song edition; and the copy compiled in the Shuofu (Persuasion of the suburbs) in a Ming manuscript edition. In addition to that collated by Shoujie according to the Tao and Jiang editions, there are six editions. Now I have come across the copy held by Ye Wenzhuang and collated by Mao Fuji, and know that it was the master copy used by He Yimen. After comparison, I corrected a lot of mistakes. How deep is Daoren’s special tie with this book! ̗6ƇƒƝn¼Ʊ?ʀ̶\ȦʗūƦǞɎȩ\ȦʑƱʇ'ǕïƔw̳•ƦɯïƔ™̀Ž\Ũˎ‘ʇ'Ǎǵǖ;ûƦƱʇ'Ǝ6ū	ˠòŨ̳Ʀ'£#ʗƠɱ̓ŨƱ̿ʲ,ǖƦʈfɄ7˜̕ǕƄôũƱʬƃʦʺƦ"Ʌ’ȅ?ʀ̶ƦŨɉ,Ć)ŲǺȥÙ̗6ƇƒƝ'3?ɵ#ǯ̡ Wu believed that he has a “special tie” with the Zhongwu jiwen. Many scholars in late imperial China believed that they had a special relationship with certain books. These “reading seeds” were devoted to scholarly practices related to books and texts all their lives. It seems less important whether there is an ultimate answer to their questions about their life and the world. What they treasured and were obsessed with, as least as expressed in marginalia, was this special tie with books.   218 6 Conclusion: Marginalia, Scholars, and the Intellectual History of the Qing  A Kind of Text and a Kind of Culture This study of marginalia supplies an entry point into the history of texts and the history of scholars in late imperial China. Two key concepts form the foundation of this study: marginalia and scholars. Marginalia are a variety of writings and symbols drawn by readers on the margins of books, mostly printed books in the Qing dynasty. Before the late Ming, marginalia were rare in both records and physical books. In the late Ming and the early Qing period, print became the dominant mode of the circulation of texts; books became more and more accessible; and scholars had ample opportunity to compose marginalia. In this context, He Zhuo, a textual scholar and calligrapher, one of the tutors of the emperor’s sons, and friend of the most famous bibliophiles in the Jiangnan region and the capital Beijing, devoted himself to reading and collating books and composed marginalia on hundreds of titles. He inherited the Song-Ming tradition of commenting on texts, appreciating their literary features and discussing their historical and philosophical contents. At the same time, he conducted a very careful and in-depth project of textual criticism and collated a great number of texts. He was praised as a “reading seed” by the Kangxi Emperor, and regarded as one of the initiators of textual criticism, a sub-discipline within the methodology of evidential research that flourished in the mid-Qing period. After his death, his marginalia quickly became popular, and books containing his marginalia sold at a rather high price. Bibliophiles sought books with his marginalia and collected them, treasuring them to the degree that they would not show them to others. Merchants forged them. More importantly, more and more scholars transcribed  219 them and composed marginalia of their own. In the mid-Qing period, composing and transcribing marginalia became a scholarly practice conducted by a great number of scholars.  When marginalia were transcribed from one printed book to another, their textual space was altered, as was their relationship with the main text; they were also edited in both form and content to adapt to the new textual environment. In this process, how a text was read could be rather different from how it is read today: the colorful comments, symbols and collation notes were difficult to ignore while reading. Qing scholars derived thought and knowledge, and built up their worldview by drawing on various texts. Therefore, how texts were read and interpreted could affect how they perceived both their external and inner worlds and the relationship between them. Marginalia and their long circulation history supply sources whereby we can explore these scholars’ reading practices, life styles, thought, and mental states. Composing marginalia was by no means an absolutely new practice. Nevertheless, under the influence of a publishing boom, changes in the way scholars were sponsored, and the cumulative development of research on ancient texts during the late Ming and early Qing periods, this practice developed to a new stage, becoming a new fashion practiced by scholars empire-wide. In this process, a particular scholarly culture, “marginalia culture,” emerged. This marginalia culture was a hybrid of printed culture and manuscript culture, and was created by a series of social, political, economic, and philosophical factors. In its mature stage, it developed into a systematic way of perceiving, analyzing, and acting, and had a meaningful and powerful impact on its practitioners.  In chapters four and five, I discussed the transcription of He Zhuo’s marginalia by various agents, including He Zhuo’s disciples, later scholars, bibliophiles, calligraphers, book merchants, and so forth. This practice meant different things to different agents. For He Zhuo’s disciples, transcription was a way to learn from their teacher; for scholars, it supplied abundant research  220 sources for both collating and making sense of texts; for calligraphers, it furnished an opportunity to imitate a masterpiece of calligraphy and improve their skills; for book sellers and scribes, it was a profitable enterprise. In this cultural landscape, reading was never merely a way of obtaining knowledge; it was a habitual lifestyle, a practice used to cultivate oneself. This practice, for these “reading seeds,” entailed not only reading the words of a book, but also giving attention to the authentication of the text and form of the book, consulting the various marginalia surrounding the main text, transcribing previous scholars’ marginalia and composing their own, and appreciating the calligraphy of the marginalia and sometimes imitating them.  Modern scholarly literature holds the idea that Qing evidential researchers proposed to search for the Way (dao #), or recover the Way of antiquity, through carefully studying the texts of the Classics. 1  This is correct. Searching for and/or recovering the Way of antiquity was always Confucian scholars’ ultimate goal. Nevertheless, I argue in this study that the practice of searching itself also counted. To some extent, it had more meaning and significance than the ultimate goal they were searching for yet could hardly achieve. Reading, collating, and looking for the “authentic” texts were their way of cultivating themselves, perceiving this world, and living in it.  Marginalia Culture and the Intellectual Trend When composing and transcribing marginalia became a common scholarly practice in the mid-Qing period, marginalia were given more attention. Some of them were collected, edited, carved, and printed. Printing made marginalia accessible to a much larger reading public. This process can                                                         1 See Sela, China’s Philological Turn, 4-5. Sela refers to this approach as “textualism.”  221 also show that when marginalia culture came into being, its influence proved pervasive, permeating people’s practices in both scholarly and daily life.  Furthermore, as discussed in the previous chapters, marginalia usually drew upon a very large range of sources, including literary and historical comments, collation notes, and various colophons, to name a few. However, publishers in the Qing dynasty did not publish all of them. They chose what met their academic tastes and the supposed tastes of their readers. They also edited the text of the marginalia they published. Let us take the printing of He Zhuo’s marginalia as an example of this editorial manipulation. Generally speaking, He Zhuo’s marginalia were printed in three ways in the Qing dynasty and the Republican period: 1) His marginalia on eighteen titles were collected and compiled as a book named Yimen dushu ji (Master Yimen’s reading notes); 2) some marginalia, mostly literary and historical comments, were printed alongside the main text on the margins of the book or within the main text in double-columned small characters; 3) some of his collation notes were extracted from his marginalia and edited, then attached at the end of the main text or included in scholarly collectanea. He Zhuo’s Yimen dushu ji was a well-known scholarly biji ! (lit. brush notes) in the Qing dynasty. But unlike what had been written before it, it was compiled from the marginalia and its contents were arranged according to the sequence of the main, whereas previous scholarly biji were usually compiled from notation books and their contents arranged according to different topics.2 He Zhuo’s marginalia covered a very wide range of topics, including historical and literary criticism, collation notes, variants from other editions, comments on the visual appearance and                                                         2 About scholarly biji, see Liu Yeqiu , Lidai biji gaishu !" (Beijing: Zhonghua shuju, 1980), 1-10, 33-36, 71-78, 106-112, 132-136, 158-166, 193-206.  222 physical conditions of the book, short colophons on the circulation of the book, and other amorphous thoughts while reading. Usually they were not deep research on specific topics but close reading of the main text, including anything that might have something to do with the main text. While compiling the Yimen dushu ji, the compilers left out most of the collation notes and omitted almost all of He Zhuo’s colophons. Most of what they included were He’s historical and literary comments, and these comments often were edited in some way.  Taking He’s marginalia in the Hou Hanshu as an example, the copy of Hou Hanshu held at the Peking University Library (Call No.: LSB/7288; abbr. PKU copy) contains 3,726 of He Zhuo’s marginalia, but the Yimen dushuji only has around 700 of those found in the PKU copy. Among these 3726 pieces of marginalia there are 1,625 pieces concerned with textual criticism, of which the Yimen dushuji included just ninety-four. More than nine tenths of the collation notes were left out. There were more than ten colophons in the PKU copy of the Hou Hanshu, discussing the editions of the Hou Hanshu and introducing the way that He Zhuo the Hou Hanshu, and recorded his private life, thoughts and mental states during the days he was engaged in the task. None of these were included in the Yimen dushuji. For the Hou Hanshu, most of the items in the Yimen dushiji are historical comments. The compiler appears to have favored the longer ones He Zhuo wrote, omitting most of the short ones. These longer comments can be seen as scholarly notes, suggesting that the compiler wanted to make the Yimen dushuji a traditional scholarly biji, regardless of  He Zhuo’s efforts in textual criticism.3 In the early Qing period, just at the time when the Yimen dushuji was compiled, some of He Zhuo’s marginalia were printed alongside the main text, of which the best-known were He’s                                                         3 About He Zhuo’s marginalia on the Hou Hanshu, see Wei Yinzong, “He Zhuo Hou Hanshu pijiao zhi zhengli yu yanjiu”  (MS diss., Peking University, 2014).  223 marginalia on the Kunxue jiwen and the Wenxuan. For the former, other scholars’ exegeses and comments were printed within the main text in double-lined small characters, following the layout of annotations of Confucian Classics and histories. However, He Zhuo’s comments on the Wenxuan were printed mainly on the top margin, sometimes in red ink, similar to the printing of literary commentaries for classical proses, poetry, and fictions in the late Ming period. The idea of printing reader’s comments was not entirely new. As discussed in chapter two, printing classical prose, poetry, and works of drama and fiction with literary commentaries and emphasis marks was a widespread practice in the late Ming. These were crafted commentaries rather than truly spontaneous reactions, but rhetorically they retained the aura of spontaneity and sprezzatura that was desirable to the late Ming readers. Most, if not all, of them were commercial imprints produced for students to master skills for the civil service examinations and for common readers to appreciate the aesthetic values of the text. Printed marginalia, to some extent, inherited this practice and shared the characteristics of commercial printing, themselves being commercial imprints for the most part. Furthermore, while printed, marginalia were edited to look like commentaries: almost all collation notes were left out and the contents were altered to make them more easy reading. The production of Yimen dushuji and the printing of He’s marginalia within the main text both show the publishers’ intent to produce easy-reading texts that imparted relatively solid and reliable knowledge to a reading group that extended well beyond scholarly circles. The prevalence of these books shows that most scholars in the early- and mid- Qing period still followed the reading habits and even the scholarly methodology of Ming scholars. There was no drastic transformation from Ming to the Qing in the practice of reading. The development of new approaches, eventuating in  224 evidential research, was a rather long process that extended from a small group of scholars to larger groups and geographical areas. It was in the mid-Qing era that collation notes began to be collected from marginalia, edited, and printed. This practice flourished in the late Qing and into the Republican period. For instance, the Su Xueshi wenji  % (Collected works of Su Shunqin   [1008-1048]) in the well-known collectanea Sibu congkan $	  (The four branches of literature collection) includes He Zhuo’s collation notes printed at the end of the book (see Figure 6-1). Many scholars in the mid- and late-Qing period also edited their own collation notes and printed them, creating a new kind of scholarly biji—collation biji. Collation notes were very important scholarly sources for authenticating the texts. The flourishing of collation biji and the printing of various collations notes indicate that textual criticism (jiaokanxue ; lit. studies of collation), as an independent scholarly discipline, had already reached maturity in both theory and practice in the Qing dynasty.  The editing and printing of  marginalia is thus an important area which can supply a great deal of information concerning the evolution of scholarly culture and Qing intellectual trends. It deserves further research.  Invisible Scholars and the Intellectual History of the Qing One of the principle aims of this study is to try to understand the society and scholars of Qing China. To do so, instead of focusing on the most prominent scholars or scholar-officials, I also pay attention to little-known scholars, who are unfamiliar to most modern readers but who contributed much to Qing scholarship and our understanding of ancient texts. At the same time, instead of highlighting scholars’ socio-political activities or philosophical achievements, I mainly examine their practices with marginalia, investigate various texts they read and composed, and explore  225 relationships among different scholars and between scholars and texts. Through investigating scholarly practices of underappreciated scholars laegely ignored by our scholarship, I expect to reveal some hidden corners of intellectual history.  Figure 6-1  First half leaf of the “Su Xueshi wenji jiao yu,” attached at the end of the Su Xueshi wenji, in the Sibu congkan.  226 Since most Qing scholars were involved in the marginalia culture, I seek to understand Qing scholars from their marginalia and the scholarly practices related to the marginalia they read, wrote, disseminated, and used. I argue that these Qing scholars did not passively yield to severe political pressure occasioned by the central government and socio-economic transformations; neither did they mechanically transmit philosophical concepts and doctrines. They sought to understand the world and themselves, and tried to live properly in the world under the guidance of the meaning they gained from the texts they read and collated. They expected to search for the Way, or recover the Way of antiquity—trying to build a new world on the foundation of ancient texts. To read and collate books, to research and transcribe texts, to live properly following the doctrines in the classics, and to do anything related to scholarship was their way of conducting “practical affairs,” which, in their opinion, could guide them to a new world, or a world that had long ago been conceived and formulated by sages. Conventional research on Qing intellectual history has held a general understanding that there are some differences in scholarly method, style, and intellect between scholars of the Qing and the Ming, that is, that there was an Ming-Qing transition or Ming-Qing transformation. Ming learning is considered to have been dominated by the Neo-Confucianism that came into being in the Song dynasty and had two distinct factions: the Cheng-Zhu Learning of Principle and the Lu-Wang Learning of Mind and Heart. Qing learning, however, is believed by many researchers to have been dominated by the evidential research. The more important question is how much do scholars of the Ming and Qing differ? How did evidential research gather momentum, and how did its popularity wax and wane in different regions and different time periods? How did different groups of Qing scholars treat the legacy of Ming learning?  227 As discussed in the introduction, Liang Qichao, and other scholars of his generation, stressed discontinuity between the Qing learning and the Ming learning. By declaring that “The point of departure of Ch’ing (Qing) learning was a violent reaction against the Neo-Confucianism of the Song and Ming,”4 Liang emphasized this difference. Focusing on the prominent scholars of the Qing—mostly the evidential scholars—he argued that it the severe political pressure occasioned by Manchu rule that compelled scholars to conduct apolitical evidential research. Liang pointed out the determinative impact of political factors on scholars and their scholarship, and seemed to have the idea that evidential research was prevalent all over the empire. Recent research, especially that by western scholars, also examines the social-economic factors that contributed to the rise of evidential research, and finds that the evidential research was only popular only in some regions rather than the whole empire. Moreover, contrary to the prevailing view, Qian Mu and other scholars have focused on the development of the Cheng-Zhu and Lu-Wang branches of Neo-Confucian thought in the Qing dynasty and found a continuity of Confucian teachings from Ming to Qing. More recently, Yu Yingshi has employed the concept of “inner logic” of intellectual trends “regardless of political, economic and other external factors” to develop Qian Mu’s argument.5 These scholars have explored the best-known Cheng-Zhu and Lu-Wang scholars and make the claim that both schools were still rather popular and underwent a great development in the Qing. These studies are instructive, but they meanwhile raise a host of new questions: Did evidential scholars conduct research employing entirely new approaches? In which aspects and to what extent did they learn from the Ming scholars? How did evidential research influence Neo-Confucian                                                         4 Liang Qichao, Intellectual Trends in the Ch’ing Period, 27. 5 Yu Yingshi, “Some Preliminary Observations on the Rise of Ch’ing Confucian Intellectualism,” 112, 128; “Qingdai sixiangshi de yige xin jieshi,” 199.  228 scholars, of whatever stripe? Or, more broadly, how did the new scholarship, evidential research, influence scholarly culture and the whole society in the Qing? To answer these questions, in addition to political and social-economic factors and the evolution of philosophical thought itself, we should also take scholars and their scholarly practices into the account. In this study, I have taken a cultural view of history, concentrating on scholars themselves as real human beings and examining their practices with marginalia, so as to explore the characteristics and transformation of their scholarly styles, thoughts, and mental states.  The scholarly practices entailed with the composition and transcription of marginalia, as undertaken by He Zhuo, Jiang Gao, Yao Shiyu, Weng Fanggang, Zhang Yu, and other lesser-known scholars mentioned in this study, show Qing scholars practicing new ways of doing research, but these scholars also learned from the Ming scholars. In terms of reading ancient texts, Ming scholars liked making emphasis marks and writing literary, historical, and philosophical commentaries, and they published a great many book filled with these devices. These practices were condemned by some of the most prominent evidential scholars in the Qing dynasty. For example, in the Siku zongmu, few books of this kind produced in the Ming were included. Nevertheless, from the marginalia of common Qing scholars, we can see that emphasis marks and commentaries were still rather popular through the whole of the Qing dynasty. Scholars did not do evidential research every time they read a book. They also appreciated the text and commented on it just as Ming scholars had done. The practice of writing and reading commentaries had become a kind of reading habit since the Song, and the Qing inherited this practice from their predecessors. In addition, textual criticism became one of the representative sub-disciplines of evidential research in the Qing. But we should keep in mind that many Qing scholars, especially the ones who lived in the early- and mid-Qing period, were doing textual criticism on Ming editions. These  229 Ming editions, which the Qing scholars read and collated, were a great scholarly legacy that Qing scholars inherited from the Ming. Liang Qichao claimed that Qing learning was a “violent reaction” against Song-Ming learning, but this “reaction” was also built on an inheritance. Qian Mu and Yu Yingshi argued that Qing learning was an internal development of Ming learning, but I want to make a complementary claim, that Qing scholars also learned scholarly approaches and reading habits from Ming scholars. The Ming-Qing transition was not only caused by political, social-economic or philosophical factors, but was also the result and manifestation of developments in scholarly culture. The study of Qing scholars’ practices with marginalia can add more details in the understanding of past scholars, and help us drew a broader pictures of Qing scholarship. In this study, I focused on He Zhuo’s marginalia, discussed their characteristics, investigated their circulation, and explored their influence on later scholars and scholarship. It is just a starting point for marginalia culture. In order to draw a clearer and broader picture, we need to do more works such as cataloguing books that contain marginalia, examining their interactions with other types of texts, and doing more case studies of marginalia on a particular title or by a particular reader.    230 Bibliography  Ban Gu Ćd. Hanshu ñÀ. Beijing: Zhonghua shuju, 1962. Blair, Ann. Too Much to Know: Managing Scholarly Information before the Modern Age. New Haven: Yale University Press, 2010. Boyarin, Daniel. Intertextuality and the Reading of Midrash. Bloomington & Indianapolis: Indiana University Press, 1990. Brokaw, Cynthia J. Commerce in Culture: The Sibao Book Trade in the Qing and Republican Periods. Cambridge: Harvard University Press, 2007. Brokaw, Cynthia J. and Kai-wing Chow, eds. Printing and Book Culture in Late Imperial China. Berkeley, Los Angeles and London: University of California Press, 2005. Cerquiglini, Bernard. In Praise of the Variant: A Critical History of Philology. Translated by Betsy Wing. Baltimore: Johns Hopkins University Press, 1999. Chartier, Roger. The Order of Books: Readers, Authors, and Libraries in Europe between the Fourteenth and Eighteenth Century. Translated by Lydia G Cochrane. Stanford: Stanford University Press, 1994. Chen Donghui ƴÊƎ. “Guanyu Shangtu cang Yili zhushu pijiaoben yu Yili zhushu xiangjiao zhi guanxi” Ʋ¶fŜ-ĦèĒªÌÆŏ-ĦèĒźÌƲ$. Zhongguo dianji yu wenhua e:Ķŏ´G 3 (2012): 98-103. Chen Shou ƴl. Sanguo zhi e¡. Beijing: Zhonghua shuju, 1964. Chen Suzhen ƴŞƪ. Chunqiu yu “Handao”: Liang Han zhengzhi yu zhengzhi wenhua yanjiu ºĨŏ“ñƖ”5ñ°çŏ°ç´GĠĮ. Beijing: Zhonghua shuju, 2011.  231 Chen Xianxing ƴ1ť. Zhongguo guji gao chao jiao ben tulu eRĶĬƤÌÆfƦ. Shanghai: Shanghai guji chubanshe, 2014. Cherniack, Susan. “Book Culture and Textual Transmission in Sung China.” Harvard Journal of Asiatic Studies 54, no. 1 (1994): 5–125. Chia, Lucille. Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th-17th Centuries). Cambridge: Harvard University Asia Center, 2002. Chow, Kai-wing. “Paratext: Commentaries, Ideology, and Politics,” in Publishing, Culture and Power in Early Modern China. Stanford: Stanford University Press, 2004. Darnton, Robert. The Kiss of Lamourette: Reflections in Cultural History. New York: Norton, 1990. Deng Zhicheng Ɲż. Qingshi jishi chubian ìŷĸ@ľ. In Zhou Junfu ed., Qingdai zhuanji congkan, v. 020. Du You É . Tongdian Ɣ:. Beijing: Zhonghu shuju, 1992. Eliot, Simon, and Jonathan Rose, eds. A Companion to the History of the Book. Malden, MA: Blackwell, 2007. Elman, Benjamin. Classicism, Politics and Kinship: The Ch’ang-chou School of New Text Confucianism in Late Imperial China. Berkeley: University of California Press, 1990. ——. A Cultural History of Civil Examination in Late Imperial China. Berkeley: University of California Press, 2000. ——. From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China. Revised Edition. Los Angeles: University of California, 2001. ——. Civil Examinations and Meritocracy in Late Imperial China. Cambridge, Massachusetts; London, England: Harvard University Press, 2013.  232 Elman, Benjamin and Alexander Woodside, eds. Education and Society in Late Imperial China. Berkeley: University of California Press, 1994. Fan Ye Ŕ¿. Hou Hanshu šñÀ. Annotated by Liu Zhao B». Beijing: Zhonghua shuju, 1973. ——. Woodblock edition cut at the Jigu pavilion [Jigu ge åRư] of the MaoÜ family in the late Ming, held at the Peking University Library. Fan Zhixin Ŕ¡µ, ed.  Wenxuan He Zhuo jiao ji zheng ´ƙ!ùÌƸƀ. Zhengzhou: Henan daxue chubanshe, 2016. Feng Ban ǂĆ. Dunyin zalu ƣVƺƦ. Commented by He Zhuo, edited by Yang Haizheng Îë‡ and Wei Yinzong ƼŌ. Nanjing: Fenghuang chubanshe, 2017. Feng Menglong ǂoǍ, compiler. Stories Old and New: A Ming Dynasty Collection. Translated by Shuhui Yang and Yunqing Yang. Seattle and London: University of Washington Press, 2000. Feng Youlan ǂNş (Fung Yu-lan). A Short History of Chinese Philosophy, Derk Bodde, ed. New York: Macmillan, 1960. Foucault, Michel. “What is an Author?” Translated by Josue V. Harari. In James D. Faubion, ed., Aesthetics, Method, and Epistemology. New York: The New Press, 1998. Gardner, Daniel K. “Confucian Commentary and Chinese Intellectual History.” Journal of Asian Studies 57 no. 2 (May 1998): 397-422. Genette, Gérard. “The Proustian Paratext.” SubStance: a Review of Theory and Literary Criticism 17.2 (1988): 63-77. ——. Paratexts: Thresholds of Interpretation. Translated by Jane E. Lewin. Cambridge: Cambridge University Press, 1997.  233 Ge Zhaoguang ř02. Zhongguo sixiangshi, di’er juan: qi shiji zhi shijiu shiji zhongguo de zhishi, sixiang yu xinyang e£¥SIJLĸŎIĸeėğƁ£¥ŏ). Second edition. Shanghai: Fudan daxue chubanshe, 2013. Grafton, Antony. “Is the History of Reading a Marginal Enterprise?” Papers of the Biblographic Society of America 91(1997): 139-157. Guo Shaoyu ƛĹŢ. Zhongguo wenxue pipingshi ´´{ªŵS. Shanghai: Shanghai guji, 1979. Gu Yanwu ƾöÙ. Rizhi lu jishi¸ğƦƸƟ. Compiled by Huang Rucheng《â¨, collated and punctuated by Luan Baoqun Ó'Ń and Lü Zongli XD . Shanghai: Shanghai guji chubanshe, 2006. ——. Gu Tinglin shiwen ji ƾËŷ´Ƹ. Beijing: Zhonghua shuju, 1983. Guy, Kent. The Emperor’s Four Treasures: Scholars and the State in the Late Ch’ien-lung Era. Cambridge: Harvard Council on East Asian Studies, 1987. Gu Yongxin ƾàµ. Jingxue wenxian de yansheng he tongsuhua: Yi jingu shidai de chuanke wei zhongxin Ļ{´ĀėŦčZƔ&G——ƑR½ė,Aþ . Beijing: Beijing daxue chubanshe, 2014. Hamaguchi Fujiō ôQ‚kƶ. Shindai kōkyogaku no shisōshiteki kenkyū ìņ¯{£¥SėĠĮ. Tokyo: Kokusho kankōkai, 1994. Hegel, Robert. Reading Illustrated Fiction in Late Imperial China. Stanford, CA: Stanford University Press, 1998. Henderson, John B. Scripture, Canon and Commentary: A Comparison of Confucian and Western Exegesis. Princeton: Princeton University Press, 1991.  234 He, Yuming. Home and the World: Editing the “Glorious Ming” in Woodblock-Printed Books of the Sixteenth and Seventeenth Centuries. Cambridge: Harvard University Asia Center, 2013. He Zhuo !ù. Yimen xiansheng ji ńƯ1čƸ.  Woodblock edition cut in Gusu tŞ in the 13th year of the Daoguang reign [1833], held at the Peking University Library, photo-reproduction in Xuxiu Siku quanshu ŀ*b4À. ——. Yimen dushuji ńƯƂÀű. Beijing: Zhonghua shuju, 1987. Howsam, Leslie, ed. The Cambridge Companion to the History of the Book. Cambridge; New York: Cambridge University Press, 2015. Huang Kan 《#. Wenxuan pingdian ´ƙŵNj. Shanghai: Shanghai guji chubanshe, 1985. Huang Jinxing 《ƕŐ (Huang Chin-shing). Philosophy, Philology, and Politics in Eighteenth- Century China: Li Fu and the Lu-Wang School under the Ch’ing (Cambridge: Cambridge University Press, 1995. ——. Shengxian yu shengtu ʼnƈŏʼnœ. Beijing: Beijing daxue chubanshe, 2005. Huang, Martin W. “Author(ity) and Reader in Traditional Chinese Xiaoshuo Commentary.” Chinese Literature: Essays, Articles, Reviews (CLEAR) 16 (1994): 41-67. Huang Yongnian 》àŒ. Guji banbenxue RĶÿÆ{. Nanjing: Jiangsu jiaoyu chubanshe, 2012. Hua Zhe Ŗ_. Li shi Zheng xue Ħ¼ƞ{. Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2018. Hulvey, Monique. “Not So Marginal: Manuscript Annotations in the Folger Incunabula.” Papers of the Bibliographical Society of America 92 (1998): 159-176.  Hu Pu’anŋÒ}. Zhongguo xuguxue shi eŰŴ{S. Shanghai: Sanlian shudian, 2014. Hunt, Lynn, ed. The New Cultural History. Berkeley: University of California Press, 1989.  235 Jackson, Heather. Marginalia: readers writing in books. New Haven: Yale University Press, 2001. . “Marginal Frivolities: Readers’ Notes as Evidence for the History of Reading.” In Robin Myers et al. eds., Owners, Annotators and the Signs of Reading, 137-151. New Castle, DE: Oak Knoll; London: British Library, 2005. Jiang Liangji śőDž. Donghua lu ÊŖƦ. Beijing: Zhonghua shuju, 1980. Kessler, Lawrence. “Chinese Scholars and the Early Manchu State.” Harvard Journal of Asiatic Studies 31 (1971): 179–200. Legge, James, trans. The She King or The Book of Poetry. Taipei: SMC publishing Inc., 1991. Lewis, Mark Edward. Writing and Authority in Early China. Albany: State University of New York Press, 1999. Liang Qichcao Í]Ɖ (Liang Ch’i-ch’ao). Intellectual Trends in the Ch’ing Period. Immanuel C. Y. Hsü, trans. Cambridge: Harvard University Press, 1959. ——. Zhongguo jin sanbainian xueshu shi,eƑĖŒ{ŧS. Zhu Weizheng ÇļƧ, ed. Shanghai: Fudan daxue chubanshe, 1985. Lin Gangˆ. Ming Qing xiaoshuo pingdian¹ì…ŽŵNj. Beijing: Beijing daxue chubanshe, 2012. Lin Qingzhang ˧˜. Mingdai jingxue yanjiu lunji ¹Ļ{ĠĮžƸ. Taipei: Wenshizhe, 1994. Lin Qingzhang ˧˜ and Zhang Shouan •l}, ed. Qianjia xuezhe de yili xue a{ŇėńĈ{, 2 vols. Taipei: Zhongguo wenzhe yanjiu suo, 2003. Liu Shipei. B‹g. Liu Shipei shixue lunzhu xuanji B‹gS{žŘƙƸ. Shanghai: Shanghai guji chubanshe, 2006.  236 Liu Xie BF. Wenxin diaolong zhu ´ ƹǍè. Annotated by Fan Wenlan Ŕ´õ. Beijing: Renmin wenxue, 1962. Liu Yeqiu BŗĨ. Lidai biji gaishu Úijűσ. Beijing: Zhonghua shuju, 1980. Li Zhengyu È×|. “Dunhuang yishu zhong de biaodian fuhao” ³úƚÀėÑNjıţ. Wenshi zhishi 8 (1988): 98-101. Luo Dajing łpĻ. Helin yulu džËāƻ. Beijing: Zhonghua shuju, 1983. Luo GuanzhongłƇ. Sanguozhi yanyi e¡ðń. Edited and annotated by Mao Zonggang܆. Hong Kong: Shangwu yinshuguan, 1974. Lu Xinyuan Ƶ í. Bisonglou cangshuzhi Ę~ÐŜÀ¡. Woodblock edition carved in 1882. Lu Zonga ƵƗ and Wang Ning Ä. Xungu yu xungu xue ŰŴŏŰŴ{. Taiyuan: Shanxi jiaoyu chubanshe, 1994. Makeham, John. Transmitters and Creators: Chinese Commentators and Commentaries on the Analects. Cambridge: Harvard University Press, 2003. Maoshi zhenya Üŷ¬Ʒ. Woodblock edition carved in the late Ming period, held in the Research Institute for Oriental Cultures at the Gakushuin University. McDermott, Joseph. A Social History of the Chinese Book: Books and Literati Culture in Late Imperial China. Hong Kong: Hong Kong University Press, 2006. McKenzie, D. F. Bibliography and the Sociology of Texts. Cambridge; New York: Cambridge University Press, 1999. McLaren, Anne E. “Ming Audience and Vernacular Hermeneutics: The Uses of The Romance of the Three Kingdoms.” T‘oung Pao 81 no. 103 (1995): 51-80. Mencius. Translated by David Hinton. Washington, D.C.: Counterpoint, 1999.  237 Mingshi ¹S. Beijing: Zhonghua shuju, 1974. Myers, Robin, et al. eds. Owners, Annotators and the Signs of Reading. New Castle, DE: Oak Knoll; London: British Library, 2005. Ni Qixin+9 . Jiaokanxue dagang ÌE{pĽ. Beijing: Beijing daxue chubanshe, 2004. Ōki Yasushi pĐ. “Minmatsu Kōnan ni okeru shuppan bunka no kenkyū” ¹ÅãJ 	=ÿ´GĠĮ. Hiroshima Daigaku bungakubu 50:1 (1991): 1-173. Owen, Stephen. Readings in Chinese Literary Thought. Cambridge, Massachusetts and London: Harvard University Press, 1992. Pi Xirui ęƩċ. Jingxue lishi Ļ{ÚS. Hong Kong: Zhonghua shuju, 1961. Tsien Tsuen-Hsuin (Qian Cunxun) ƨyŰ. Paper and Printing. Vol. 5, part I of Science and Civilisation in China. Edited by Joseph Needham. Cambridge: Cambridge University Press, 1985. ——. Written on Bamboo & Silk: The Beginnings of Chinese Books & Inscriptions. Second edition. Chicago: University of Chicago Press, 2004. Qian Muƨĭ. Zhongguo jin sanbainian xueshu shi eƑĖŒ{ŧS. Shanghai: Shangwu yinshuguan, 1937. Qian Zeng ƨÁ. Dushuminqiuji jiaozheng ƂÀ±áűÌ×. Revised by Guan Tingfen ĴŽŒ and Zhang Yu. Shanghai: Shanghai guji chubanshe, 2007. Qiao Xiuyan `ħˆ. Yishuxue cunwang shi lunńĒ{ySž, Taipei: Wanjuanlou, 2013. Qi Chenghan Ģ«ǎ. Danshengtang cangshumu òčhŜÀĜ and Danshengtang dushuji òčhƂÀű. Shanghai: Shanghai guji chubanshe, 2015.  238 Qing shi liezhuan ìS?,. In Qingdai zhuanji congkan ì,űP>, compiled by Zhou Junfu Y、‚. Taibei: Yinwen, 1986. Qi Yongxiang ïàĥ. Qian-Jia kaojuxue yanjiu aņ¯{ĠĮ. Beijing: Zhongguo shehui kexue, 1998. Quan Zuwang 4ģÂ. Quan Zuwang ji huijiao jizhu 4ģÂƸ–ÌƸè. Shanghai: Shanghai guji chubanshe, 2000. Qu Yong Ğƫ. Tieqintongjianlou cangshu mulu ƬĉƥCÐŜÀĜƦ. Woodblock edition carved by the Qu Ğ family in the Xianfeng reign [1850-1861]. Ralston, David. Traditional Chinese Fiction and Fiction Commentary: Reading and Writing between the Lines. Stanford: Stanford University Press, 1997. Ralston, David, ed. How to read the Chinese Novel. Princeton, New Jersey: Princeton University Press, 1990. Ross, Jonathan. “Arriving at a History of Reading.” Historically Speaking 5 (2004): 39. Rusk, Bruce. Critics and Commentators: The Book of Poems as Classic and Literature. Cambridge and London: Harvard University Asia Center, 2012. Saussy, Haun. “The Age of Attribution: Or, How the ‘Honglou meng’ Finally Acquired an Author.” Chinese Literature: Essays, Articles, Reviews (CLEAR), Vol. 25 (2003): 119-132. Sela, Ori. China’s Philological Turn: Scholars, Textualism, and the Dao in the Eighteenth Century. New York: Columbia University Press, 2018. Sherman, William Howard. Used Books: Marking Readers in Renaissance England. Philadelphia: University of Pennsylvania Press, 2008. Shi Nai’an·ň. Guanhuatang diwu caizi shu ShuihuzhuanƇŖhIJ©wÀßî,. Edited by Jin Shengtan ƢʼnÖ. Nanjing: Jiangsu guji chubanshe, 1985.  239 Shisanjing zhushu IĻèĒ. Taipei: Yiwen, 2007. Siku quanshu zongmu tiyao b4ÀĿĜ­Ū. Beijing: Zhonghua shuju, 1965. Sima qian TǁƘ. Shiji Sű. Beijing: Zhonghua shuju, 1963. Stillinger, Jack. Coleridge and Textual Instability: The Multiple Versions of the Major Poems. New York: Oxford University Press, Inc., 1994. Sun Qinshan zÕ^. Zhongguo guwenxianxue shi eR´Ā{S. Beijing: Zhonghua shuju, 1994. Su Shi Şƍ. Su Shi ci biannian jiaozhu ŞƍŶľŒÌų. Edited by Zou Tongqing ƜU§ and Wang Zongtang Ăh. Beijing: Zhonghua shuju, 2002. Van Zoeren, Steven Jay. Poetry and Personality: Reading, Exegesis, and Hermeneutics in Traditional China. Stanford: Stanford University Press, 1991. Wakeman, Frederic. “The Price of Autonomy: Intellectuals in Ming and Ch’ing Politics.” Daedalus 101, no. 2 (1972): 35–70. Waley, Arthur, trans. The Book of Songs. New York: Grove Press, 1996. Wang Dingbao À'. Tangzhiyan \®ŭ. Shanghai: Zhonghua shuju, 1960. Wang, Edward. “Beyond East and West: Antiquarianism, Evidential Learning, and Global Trends in Historical Study.” Journal of World History 19.4 (2008): 489-519. Wang Fuzhi Ăq. Du Tongjian lun ƂƔƭž. Beijing: Zhonghua shuju, 2012. Wang Jilu ĂűƦ. “Lun shichcao” žSƤ. In Shixueshi yanjiu S{SĠĮ, 3 (2016): 1-12. Wang LiĂD. Zhongguo yuyanxue shieŻŭ{S. Zhongguo yuweneŻ´ 3(1993): 233.  Wang Shaoying äĹĚ. “Ruanshi chongke Shisanjing zhushu kao” ƳÝƠAIĻèĒņ. Wenshi ´S 3 (1963): 25-60.  240 Wang ShifuÃď. Louwailou dingzheng tuozhu diliu caizi shuÐmÐŮ×sųIJ7©wÀũ‘ű. Commentated by Jin Shengtan and annotated by Zou Shenmai Ɯʼnō. Early Qing edition cut by Zou Shengmai, held at the University of Auckland Library Special Collections. Wang Xianqian Ă1ſ. Hou Hanshu jijie šñÀƸŬ. Woodblock edition carved in the Xushou tang šOh of the Wang family in 1915. ——. Donghua lu ÊŖƦ. Photo-reproduction in the Xuxiu Siku quanshu. Wang Xinfu ĂÔq . Yishuxuan qie cun shanben shulu Ťŧƌĵy^ÆÀƦ . Shanghai: Shanghai guji chubanshe, 2002. Wang Zhongmin ĂƠÞ. Lenglu wensou ;“´ŝ. Shanghai: Shanghai guji chubanshe, 1992. Wei Li ƼD. Pijiao ben ªÌÆ. Nanjing: Fenghuang chubanshe, 2003. Wen Yiduo Ŋn. Shijing tongyi ŷĻƔń. Wuhan: Hubei renmin chubanshe, 1994. Wang Shifu Ãď. The Story of the Western Wing. Translated by Stephen H. West and Wilt L. Idema. Berkeley, Los Angeles and London: University of California Press, 1995. Wilkinson, Endymion, ed. Chinese History: A New Manual. Cambridge: Harvard University Asia Center, 2013. Wilson, Thomas. Genealogy of the Way: The Construction and Uses of the Confucian Tradition in Late Imperial China. Stanford: Stanford University Press, 1995. ——. “The Ritual Formation of Confucian Orthodoxy and the Descendants of the Sage.” Journal of Asian Studies 55.3 (August 1996): 559-84. Wilson, Thomas, ed. On Sacred Grounds: Culture, Society, Politics, and the Formation of the Cult of Confucius in Imperial China. Cambridge: Institute for East Asian Studies, Harvard University, 2002.  241 Wu Chengxue W«{. “Pingdian zhi xing: Wenxue pingdian de xingcheng yu Nan Song de shiwen pingdian” ŵNjŐ´{ŵNjė—¨ŏJ~ėŷ´ŵNj. Wenxue pinglun 1 (1995): 24-33. ——. “Xiancun pingidan diyi shu: lun Guweng guanjian de bianxuan, pingdian jiqi yingxiang” ăyƆøIJƅR´8ƮėŁƓƆøM9™[. Wenxue Yichan ´{ƚĎ 4 (2003): 72-84. Xu Ke ›ą. Qing bai leichao ìĪƽƤ. Beijing: Zhonghua shuju, 1984. Xu Shen Ų¦. Shuowen jiezi zhu Ž´Ŭxè. Annotated by Duan Yucai ÛāŨ. Shanghai: Shanghai guji chubanshe, 1981. Yan Buke ƱØ3. Shidafu zhengzhi yanshengshi gao kpq°çŦčSĬ. Beijing: Beijing daxue chubanshe, 1996. Yang Nianqun ΢Ń. Hechu shi Jiangnan?: Qingchao zhengtongguan de queli yu shilin jingshen shijie de bianyi !Š¼“ãJ”ìÃ×ĺūėġįŏkËķĤĐėƃđ. Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2010. Yao Silianu£’. LiangshuÍÀ. Beijing: Zhonghau shuju, 1973. Ye Dehui ŗŸƎ, Shulin qinghua ÀËìŹ. Shanghai: Shanghai guji chubanshe, 2008. Ye Jingkui ŗ¾Ś. Juan’an shuba LěÀƋ. Shanghai: Shanghai guji chubanshe, 2006. Yu Jiaxi"aƩ. Muluxue faweiĜƦ{ĕž. Beijing: Shangwu yinshuguan, 2011. Yu Yingshi "œ½ (Ying-shih Yü). “Some Preliminary Observations on the Rise of Ch’ing Confucian Intellectualism.” Ch’ing-hua hsueh-pao 11 (1975): 105-144. ——. Zhongguo sixiang chuantong de xiandai quanshi e£¥,ĺėćŸƟ . Nanjing: Jiangsu renmin chubanshe, 1992.  242 Yu Zhengxie (×—. Guisi cungao ĔŠyĬ. In Congshu jicheng chubian PÀƸ¨@ľ, vol. 430. Zhang Binglin İ÷Lj. Qiu shu xiang zhu ůÀźè. Annotated by Xu Fuguan ›ū . Shanghai: Shanghai guji chubanshe, 2000. Zhang Lijuan •LJv. Songdai jingshu zhushu kanke yanjiu ~ĻÀųĒ>AĠĮ. Beijing: Beijing daxue chubanshe, 2013. Zhang Li •ŕ. “Wenxuan Hailuxuan zhumo taoyinben cunyi” ´ƙëƦƌÇjrKÆyē. Henan tushuguan xuekan æJfÀǀ{> 31 no. 6 (2011): 126-128. ——. “Zhouli zhushu jiaokan ji Hui jiaoben ji qita” YĦèĒÌEű¤ÌÆM9. Wenxian ´Ā 4 (2016): 78-87. Zhang Shou’an •l}. Yi li dai li — Ling Tingkan yu Qing zhongye ruxue sixiang zhi zhuanbian ĦĈ——<”iŏìŗ.{£¥Əƃ. Shijiazhuang: Hebei jiaoyu chubanshe, 2001. ——. Shiba shiji lixue kaozheng de sixiang huoli — Lijiao lunzheng yu li zhi chong xing I6ĸĦ{ņƀė£¥éD——Ħ²žýŏĦĩƠĝ . Tiapei: Zhongyang yanjiuyuan jindaishi yanjiusuo, 2001. Zhang Xiaofei •DŽƿ. “Kunxue jiwen banben yuanliu kaoshu” c{ĸŊÿÆíêņƒ. Zhongguo dianji yu wenhua e:Ķŏ´G 2 (2009): 73-79. Zhang Xuecheng İ{ż, Wenshi tongyi jiaozhu ´SƔńÌè. Annotated and collated by Ye Ying ŗĊ. Beijing: Zhonghua shuju, 1985.  243 ——. Jiaochou tongyi tongjieÌƄƔńƔŬ. Annotated by Wang ZhongminĂƠÞ (Shanghai: Shanghai guji, 1987. Zhang Yuanji •/ó. Hou Hanshu jiankan ji šñÀÌEű. Beijing: Zhonghua shuju, 1999. Zhao JunlingƊ%Ą. “Jin chuan sanzhong He Zhuo Wenxuan pingdianben bian” ,ī!ù´ƙŵNjÆƐ. Lanzhou xuekan ş‰{> 2 (2008): 181-183. Zhao Yi ƊŅ. Oubei ji ČHƸ. Shanghai: Shanghai guji chubanshe, 1997. Zhongguo guji zongmu eRĶĿĜ . Beijing: Zhonghua shuju; Shanghai: Shanghai guji chubanshe, 2009-2012. Zhou Mi Y. Qidong yeyu njÊơŻ. Beijing: Zhonghua shuju, 1983. Zhou YiliangYő. Zhou Yiliang quanji Yő4Ƹ. Beijing: Gaodeng jiaoyu chubanshe, 2015. Zhu Xi Çû. Shijizhuan ŷƸ,. In Zhuzi quanshu Çw4À. Shanghai: Shanghai guji; Hefei: Anhui jiaoyu chubanshe, 2002.   244 APPENDIX I Books Containing He Zhuo’s Marginalia and Their Transcriptions  The table below includes books that contain He Zhuo’s marginalia and their transcriptions. In this table, books are displayed in the four-branch (sibu bɧ) classification scheme, following the order in the Siku quanshu zongmu tiyao b°2Ąǐƚæț. For each copy, I give the title, number of juan, contributor, edition, the location it is held, and the transcriptionist(s). The contributor column lists the author, compiler (abbr. “comp.”), annotator ( abbr. “ann.”), and collator. Abbreviations of libraries are listed at the end of the table. If a copy contains—or is supposed to contain—He Zhuo’s holographic marginalia, the transcriptionist(s) column is left blank; if the transcriptionist is unknown, the transcriptionist(s) column is marked “unknown.” The copies that I have examined have an asterisk (*) following their locations.    Title No. juan Contributors Edition Location Transcriptionist(s) 1 Zhouyi benyi tongshi Yûčǖɖɯ 12 Hu Bingwen ǤŨñ, comp. Manuscript copy produced in the Qing NLC* Wang Zhensheng Źàǣ 2 Xubu juye bidu shijing ǒȘǫħÉȾȩnj 4 Chen Feimu ʊʝČ, comp. Woodblock edition carved at the Yunjian tang ʆn in the Kangxi reign ZUL Chen Benli ʊčƩ 3 Zhouli zhushu YƩȦƓ 42 Zheng Xuan ɫŷ, et al., ann. Woodblock edition carved by Li Yuanyang ď-ʎ in the Jiajing reign SL* Shen Tong ł¾  245  Title No. juan Contributors Edition Location Transcriptionist(s) 4 Zhouli zhushu YƩȦƓ 42 Zheng Xuan ɫŷ, et al., ann. Woodblock edition carved at the Jigu ge ŁKʇ of the Mao ĺ family in 1628 SL* Wu Xin Uü 5 Xia xiao zheng yinyi s—ķʡǖ  Unknown Manuscript FSNL  6 Chunqiu jing zhuan jijie ýƮnj(ʖȠ 30 Du Yu đʤ, ann., Mu Wenxi Ƹñū, ed., Ge Nai Ǽˀ, re-ed. Ming woodblock edition carved at the Yonghuai tang ľÙn SL* Unknown 7 Chunqiu Zuo zhuan ýƮ¤( 30 Du Yu đʤ, ann. In Shisanjing guzhu CnjKŇ, woodblock edition carved by Jin Pan ɳȎ in the Chongzhen reign HB  8 Chunqiu Gongyang zhuan ýƮ4ǔ( 28 He Xiu , ann. In Shisanjing guzhu CnjKŇ, woodblock edition carved by Jin Pan in the Chongzhen reign HB  9 Chunqiu Gongyang zhuan ýƮ4ǔ( 20 He Xiu , ann. Woodblock edition carved in the Longqing reign NCL Ying gu shi šƷĻ 10 Chunqiu Gongyang zhushu ýƮ4ǔŇƓ 28 He Xiu , et al., anns. Woodblock edition carved at the Jigu ge of the Mao family in the Chongzhen reign NLC Yao Shiyu}ɷ 11 CZ Wang Zhensheng Źàǣ 12 Chunqiu Guliang zhuan ýƮƷĞ( 20 Fan Ning Ǵ’, et al., anns. Woodblock edition carved at the Jigu ge of the Mao family in the Chongzhen reign HB  13 Woodblock edition carved at the Yonghuai tang in the Qianlong reign HB  14 Chunqiu Guliang zhushu ýƮƷĞŇƓ 20 Fan Ning Ǵ’, et al., anns. Woodblock edition carved at the Jigu ge of the Mao family in the Chongzhen reign NLC Yao Shiyu    246  Title No. juan Contributors Edition Location Transcriptionist(s) 15 Chunqiu Dan Zhao er xiansheng jizhuan zuanli ýƮ_Ɋ/Ǝʖ(Ǒ 10 Lu Chun ʍŗ Woodblock edition carved in the Kangxi reign, in the Yulinglong ge congke ŸŻƍʇJ:č compiled by Gong Xianglin )ǚʼ SL*  16 Chunqiu Dan Zhao er xiansheng jizhuan bianyiýƮ_Ɋ/Ǝʖ(ɑƔ 10 17 Chunqiu jizhuan weizhi ýƮʖ(Æ÷ 3 18 Chunqiu si zhuanýƮb( 43 Hu Anguo Ǥ†f, comp. Ming woodblock edition carved by Ji ChengPŤ and re-edited by Fan Xianke īŶƯ TJ Unknown 19 Xiaojing zhushu ‚njȦƓ 9 Li Longji ďʐm, ann. Woodblock edition carved at the Jigu ge of the Mao family in 1629 NCL Unknown 20 Kanmiu zhengsu 9ȼķ  8 Yan Shigu ʦ§K Qing manuscript NLC* Unknown 21 Shuowen jiezi ȲñȠ€ 15 Xu Shen ȥÕ Woodblock edition carved at the Jiaohua yin fangġǹTǭ in 1773 SL* Unknown 22 Guangyun ·ʢ 5 Chen Pengnian ʊÁ«, et al., comp. Woodblock edition carved in 1705 NLC  23 Shiji NȤ 130 Sima Qian Oʭɞ Woodblock edition carved at the Jigu ge of the Mao family in 1641 NLC* Shao EnduoɣÏu 24 Hanshu şĄ 100 Ban Gu ƀe Woodblock edition carved at the Northern Directorate of Education in 1597 PKU*  25 BNU     247  Title No. juan Contributors Edition Location Transcriptionist(s) 26 Hanshu 100 Ban Gu  Woodblock edition carved at the Jigu ge of the Mao family in 1642 SL* Yao Hengtang }ȓn 27 SL* Shen Jiong łŧ 28 NLC Weng Tonghe ǘQˇ 29 Woodblock edition carved at the Wuying Palace in the Qianlong reign FD Jiao Xun ŪÅ 30 Woodblock edition carved by Zhong Renjie ʁ& in 1619, with Zhong Renjie’s collected commentaries SL* Unknown 31 Hou Hanshu ÂşĄ (including Xu Hanzhi ǒşÊ) 120 Fan Ye ǴĂ, Sima Biao OʭÀ Woodblock edition carved by Wang Wensheng ŀñƘ in the Jiajing reign SL*  32 Woodblock edition carved at the Jigu ge of the Mao family in 1643 PKU* Unknown 33 NLC* Wang Bing’en ŹƭÏ, Qian Taiji ɼňP 34 Woodblock edition carved at the Yunlin jixiu tang ʙĕƹƬn in 1627, with Chen Renxi’s  ʊɾ commentaries SL* Xie Putai Ȼŏň 35 SL* Unknown 36 Sanguo zhi fÊ   65 Chen Shou ʊr Song woodblock edition carved at the Quzhou  Ȕ£ Prefectural School and amended at the Southern Directorate of Education in the Wanli reign NCL  37 Woodblock edition carved at the Southern Directorate of Education in 1596 SL* Chen Shuhua ʊĮǹ 38 NLC* Unknown 39 NLC* Unknown 40 NLC* Unknown    248  Title No. juan Contributors Edition Location Transcriptionist(s) 41 Sanguo zhi  65 Chen Shou Woodblock edition carved at the Southern Directorate of Education in 1596, juan 3-14 are Qing edition SL* Bao Anbao A†! 42 Woodblock edition carved at the Jigu ge of the Mao family in 1644 SL* Yueshan Ljœ 43 NLC* Zhang Yu Ƽɷ 44 NLC* Unknown 45 Woodblock edition carved in the Ming with Chen Renxi’s commentaries NLC* Unknown 46 Wood movable type edition, Jinling shuju ɳʋĄ˜ in 1867 SL* Bi Yuanshen ɃśŖ 47 Woodblock edition carved at the Jinling shuju ɳʋĄ˜ in 1869 NCL Unknown 48 Woodblock edition carved at the Jinling shuju ɳʋĄ˜ in 1870 FD Liu Lüfen =›ǯ, Zhang Ruo’ai ¼Dzʜ 49 Nanshi EN 80 Li Yanshou ď¸r Woodblock edition carved at the Baoshu tang •Ąn in the Chongzhen reign SL* Unknown 50 NLC* Liu Lüfen , Wang Mingsheng ŹʷƘ 51 Beishi BN 100 Li Yanshou ď¸r Woodblock edition carved at the Jigu ge of the Mao family in 1639 SL*  52 Wudai shiji NȤ 74 Ouyang Xiu Ķʎ# Woodblock edition carved in the Jiajing reign NLC Chen kui ʊå 53 Woodblock edition carved at the Jigu ge of the Mao family in 1630 SL* Yao Shiyu 54 NLC* Zhang Yu 55 NLC*  56 Zhuozhong zhi ɭÊ 24 Liu Ruoyu =DzÓ Ming manuscript NLC     249  Title No. juan Contributors Edition Location Transcriptionist(s) 57 Huayang guo zhi ǹʎfÊ 12 Chang Qu ©Ƌ Woodblock edition carved at Liao Yin’s ¶ Tijin guan ʥȚʫ in 1814 PKU Unknown 58 Woodblock edition carved by the Tao ʌ family of Kuaiji ćƵ in 1890 NLC* Unknown 59 Gujin shi yi KN edition NLC*  60 Diaoji li tan ɴƢƻȴ 1 Shi Xubai NȌƕ In the Lianting shier zhong ĤCƳ, woodblock edition carved in Yangzhou in the Qing dynasty NCL  61 Wujun tujing xu ji UɦhnjǒȤ 3 Zhu Changwen Ďʂñ Woodblock edition carved at the Jiangsu shuju ĿȋĄ˜ in 1873 FD Wang Xinfu Źijz, Bao Qiyuan Ȅ^ś 62 Shuijing zhu jian ĽnjŇǀ 40 Li Daoyuan ɬɛ- and Zhu Mou Ďȷ, anns. Woodblock edition carved by Li Changgeng ďʂ® in 1615 NCL  63 NCL Unknown 64 WUL  65 NLC Unknown 66 Woodblock edition carved in 1629 NLC* Unknown 67 Shuijing zhu ĽnjŇ 40 Li Daoyuan ɬɛ-, anns. Woodblock edition carved at the Xiang ʣ family’s Qunyu shutang ǕŸĄn in 1715 SL* Unknown 68 FD Unknown 69 Woodblock edition carved at Huang Sheng’s ʽÿ Huaiyin caotang ĩȂǶn in 1753 SL* Shen Dacheng łwÛ 70 Ming manuscript NLC  71 Zhongwu ji wen ULJǡ 6 Gong Mingzhi )ú	 Woodblock edition carved at the Jigu ge of the Mao family in the late Ming NLC*  72 NLC*  73 NLC* Lao Quan @ı 74 NLC* Zhang Shaoren¼lj  250   Title No. juan Contributors Edition Location Transcriptionist(s) 75 Zhongwu ji wen 1 Gong Mingzhi Woodblock edition carved at the Jigu ge of the Mao family in the late Ming NLC* Yuan Tingtao Ȗ¹İ, Wu Zhizhong UÊË 76 Wudai huiyao ćț 30 Wang Pu ŹŜ Manuscript by Sun Qian „Ţ, 1667 NLC*  77 Jinshi lu ɳƞɻ 30 Zhao Mingcheng ɊúȰ Woodblock edition carved by Xie Shiji Ȼǁ in 1647 SL* Dong Chun ǽɮ 78 Woodblock edition carved at the Luƙ family’s Yayu tang ʕʘn in 1762 SL* Ye Zhishen ǻÊȬ 79 NLC*  80 NLC Gu Guangqi 81 Qing manuscript NLC Wu Zhizhong 82 Li shi ʓɯ 27 Hong Kuo Ōɝ Ming manuscript SL*  83 Li xu ʓǒ 21 Hong Kuo Ōɝ Ming manuscript NLC* Liu Yi’an =Ě² 84 Shike puxu ƞ:ɺí 2 Zeng Hongfu ƈŲ Manuscript by Dong Zhaoyuanǽ.-,  1760 NLC* Dong Cong ǽŮ 85 Jinji linlang ɳȅƅƂCG 20 Du Mu ɩƸ  NLC* Xu Han ȥŦ 86 Shitong Nɖ 20 Liu Zhiji =Ɲ¬ Ming woodblock edition, with Li Weizhen and Guo Yannian’s commentaries SL* Unknown 87 Woodblock edition carved by Lu Shen ʍŖ in 1536 SL* Weng Tonghe 88 Woodblock edition carved by Zhang Zhixiang ¼	ɂ in 1577 SL* Jiang Gao ȁē 89 NLC Wu Cipei UÔl 90 Woodblock edition carved by Zhang Dingsi ¼ʿÍ in 1602 SL* Ye Jingkui ǻĀǾ    251  Title No. juan Contributors Edition Location Transcriptionist(s) 91 Shitong tongshi Nɖɖɯ 20 Pu Qilong ŏɉ( Woodblock edition carved at the Qiu fangxin zhai ëÈ˂ of the Pu family in 1752 PKU Unknown 92 Xin Xu ó­ 10 Liu Xiang =S Ming woodblock edition NLC* Unknown 93 NLC* Gu Guangqiʨ·j 94 NJ  95 Shuo Yuan ȱDZ 10 Liu Xiang =S Woodblock edition carved by Cheng Rong ƲĨ in the Ming ML Jiang Deliang ĿÇɲ 96 Xinzuan menmu wuchen yinzhu Yangzi fayan óǑʃƚǦʡȦçņȡ 10 Yang Xiong çʔ; Li Gu ďɌi, et al., ann. Woodblock edition carved  at the Shide tang Çn in 1533 SL* Ye Changchi ǻùŰ 97 NLC* Shen Yanł¡ 98 Woodblock edition carved at the Tongyin shuwu ĜȂĄš in the Ming dynasty SL* Yuan Tingtao 99 Yangzi fayan Ģņȡ 13 Yang Xiong çʔ Woodblock edition carved at the Qin Ʊ family’s Yanzhai Ɵ˂ in 1818 NLC* Weng Tonghe 100 Zhong shuo ȱ 10 Wang Tong Źɖ; Ruan Yi ʈə, ann. Ming woodblock edition NLC Fu Zengxiang%oř 101 Tai xuan yŷ 10 Yang Xiong çʔ; Fan Wang ǃĊ, ann. Woodblock edition carved at Hao Liang’s ɥĞ Wanyu tang ǺŸn in 1524 NLC*  102 NCL  103 Fa shu yao lu ņĄțɻ 10 Zhang Yanyuan ¼¿ɜ In the Jindai mishu  ŋɗưĄ, woodblock edition carved at the Jigu ge of the Mao family in the Chongzhen reign NLC*  104 NLC Fu Zengxiang    252  Title No. juan Contributors Edition Location Transcriptionist(s) 105 Gengzi xiaoxia  ji ®ɹsȤ (with Xianzhe xuan tie kao ʅǟɍ¦Ǟ) 8 Sun Chengze „ßť Woodblock edition carved at Bao Tongbo’s  ʴ¹F Zhibuzu zhai Ɲɋ˂ in 1761 NLC* Unknown 106 NLC* Ye Shang ǻ\ 107 SL Gong Cheng )į 108 SL Unknown 109 Lead moveable type edition, 1911 NLC Fu Zengxiang 110 Yan jian Ơǀ 4 Gao Sisun ʱ„ Woodblock edition carved in 1706 NLC Fu Zengxiang 111 Huainan honglie jie ŕEʸũȠ 21 Xu Shen, Gao You ʱȮ, ann. Woodblock edition carved in the Ming dynasty NLC Unknown 112 Yanshi jiaxun ʦĻȣ 7 Yan Zhitui ʦ	ä Yuan woodblock edition SL*  113  Woodblock edition carved in the Wanli reign NLC  114 Dongguan yulun Ēȟʪȵ 2 Huang Bosi ʽÍ Woodblock edition carved in the Chongzhen reign NLC Fu Zengxiang 115 Yijue liao zaji ŵȞ“ʗȤ 2 Zhu Yi ĎǙ Qing manuscript NSL Unknown 116 NJ Unknown 117 Nenggai zhai manlu ǥê˂Šɻ 18 Wu Zeng UĆ Manuscript NCL  118 Rongzhai suibi Ž˂ʑƾ  Hong Mai Ōɡ Qing manuscript NLC Wu Qian Uʯ 119 Bin tui lu ɇɕɻ 10 Zhao Yushi Ɋǩø Qing manuscript SL* Hu Ting ǤƁ 120 NLC Zhou ShutaoYIº 121 Kunxue jiwen d…LJǡ 3 Wang Yinglin ŹØʼ Woodblock edition carved by UŶM in 1603 SL* Unknown 122 20 Ming woodblock edition NLC* Jiang Gao 123 Zhu shi ʻN 3 Wang Dechen ŹÄǦ Ming manuscript NCL   253  Title No. juan Contributors Edition Location Transcriptionist(s) 124 Lengzhai yehua 8˂vȫ 10 Huihong ÑŌ Woodblock edition carved in the Wanli reign NLC Fu Zengxiang  125 Laoxue an biji ǝ…²ƾȤ 10 Lu You ʍɚ Woodblock edition carved in the Wanli reign NLC Fu Zengxiang 126 Woodblock edition carved at the Chongwen shuju ŸñĄ˜ in Hubei in 1877 PKU Zhao Zhen 127 Xianju lu ʅ™ɻ 1 Wu Yan Wȑ Manuscript produced at Cao Rong’s ąŞ Xueru shuwu …ǠĄš in the Shunzhi reign NCL  128 Manuscript NCL  129 Dunyin (laoren) zalu ɵTǝʗɻ 10 Feng Ban ʮƀ Woodblock edition carved by Feng Wu ʮĸ in 1679 SL* Unknown 130 NCL Unknown 131 Qing manuscript NCL Unknown 132 Men shi xinhua ãȍóȫ 8 Chen Shan ʊ` Manuscript NCL  133 Shishuo xinyu ȱóȯ  3 Liu Yiqing =ǖÖ; Liu Xiaobiao ann. Ming woodblock edition with Liu Chenweng’s =ɒǘ commentary SL Zhang Sijiao¼bî 134 3 Woodblock edition carved in 1609 NLC Fu Zengxiang 135 6 Woodblock edition carved at the Chongwen shuju ŸñĄ˜ in Hubei in 1877 PKU Zhao Zhen Ɋĥ 136 Nanbu xinshu EɧóĄ 10 Qian Yi ɼû Ming woodblock edition NLC  137 Gui qian zhi ĹŢÊ 8 Liu Qi =ƣ Manuscript NCL  138 Shan hai jing zhuan œŒnj( 18 Guo Pu ɨƊ Woodblock edition carved in the Wanli reign NLC Zhou shutao 139 Woodblock edition carved at the Hexia tang ńƧĭ in 1714 NLC Unknown  254  Title No. juan Contributors Edition Location Transcriptionist(s) 140  Lu yi ji ɻƒȤ  Du Guangting đ0± Ming manuscript NLC  141 Tao shi jizhu ʌȩʖȦ 4 Tao Qian ʌŢ; Zhan Kuixi ȭtɾ, comp. Woodblock edition carved at the Mobao tang –pn in 1694 SL* Guan Tingfen ǂ±ǯ 142 Tao Yuanming ji ʌŘúʖ 10 Tao Qian Woodblock edition carved by Li Wenhan ďñʠ in the early Qing SL* Shen Tingfen ł¹ǰ, Wu Suchang Uʂ 143 Ming woodblock edition PKU  144 Woodblock multi-color-printed edition carved in the late Qing PKU Huang Enchang ʽÏʂ 145 Xie Xuancheng shiji ȻŒkȩʖ 5 Xie Tiao ȻƜ Manuscript by Jiang Gao, 1710 NLC Han YingbiʠØʉ, Jiang Gao ȁē 146 Xie Xuancheng shi ȻŒkȩ 4 Woodblock edition carved in 1735 NLC Weng Tonghe ǘQˇ 147 Yu Kaifu shiji ´ʄ¯ȩʖ 6 Yu Xin ´" Woodblock edition carved by Zhu Yuefan ĎăȈ in the Jiajing reign NCL  148 Li Hanlin ji ďǛĕʖ 10 Li Bai ďƕ Woodblock edition carved by Lu Yuanda ʍ-w in 1519 NLC  149 NLC Fu Zengxiang 150 Wang Mojie ji ŹéȪʖ 10 Wang Wei ŹǍ Qing manuscript NLC* Shao Enduo ɣÏu 151 Qian Kaogong shiji ɼǞ>ȩʖ 10 Qian Qi ɼɉ Ming manuscript NLC  152 Liu Suizhou shiji =ʑ£ȩʖ 11 Liu Zhangqing =ʂH Woodblock edition carved in 1550 NLC Fu Zengxiang  255  Title No. juan Contributors Edition Location Transcriptionist(s) 153 Liu Suizhou shi =ʑ£ȩ 10 Liu Zhangqing Woodblock edition carved by the Xi ¨ family of Dongting Ŋ± in 1702 SL* Unknown 154 Liu Suizhou wenji =ʑ£ñʖ 10 Woodblock edition carved in 1776 NLC Lu Wenchaoƙñ» 155 Changli xiansheng ji ùʾ/Ǝʖ 52 Han Yu ʠÒ; Liao Yingzhong ¶Ɖ, et al., anns. Woodblock edition carved at the Xu family’s Dongya tang Ēʕn in the Ming SL* Zhu Luo ȶʼn 156 HB Unknown 157 JLU Zhang Wei ¼Ƈ 158 RY Fang Chengui ôÛſ 159 Woodblock edition carved at the Xu family’s Dongya tang in the Ming and amended at the Guanshan tang 7œn in the Qing SL* Shao Qi ɣź 160 JLU Unknown 161 Changli xiansheng shi jizhu ùʾ/ƎȩʖŇ 11 Han Yu ʠÒ Woodblock edition carved at the Gu’s ʨ Xiuye caotang ƬɱǶn in 1699 NKU Hou Xingnong Ǫɓ 162 SC Wei Qianheng ʟȹÎ 163 Han Changli shiji ʠùʾȩʖ  Han Yu ʠÒ Qing woodblock edition SL* Unknown 164 Han Changli shi biannian jianzhu ʠùʾȩʖǎ«ǀȦ 12 Han Yu ʠÒ Woodblock edition carved at Lu Jianzeng’s ƙȜĆ Yayu tang ʕʘn in 1758 NKU Unknown 165 Xinkan wubaijia zhu yinbian Changli xiansheng wenji ó9ƖȦʡɑùʾ/Ǝñʖ 40 Han Yu ʠÒ; Wei Zhongju ʲǫ, comp. Woodblock edition carved in 1784 CWNU Unknown    256  Title No. juan Contributors Edition Location Transcriptionist(s) 166 Zhu Wengong jiao Changli xiansheng wenji Ďñ4ěùʾ/Ǝñʖ 40 Han Yu ʠÒ; Zhu Xi Ďů, collator Woodblock edition carved in the early Ming ECNU Shen Ci łÔ 167 Woodblock edition carved by Zhu Chongmu ĎŸŃ in the Wanli reign ZJ Fan Men Ġʃ 168 BNU Wang Dun ŀï 169 Ming woodblock edition CQ Unknown 170 XA Unknown 171 Han Changli wenxuan ʠùʾñɟ 2 Han Yu ʠÒ Manuscript scribed in the Kangxi reign GS Unknown 172 Han wen chao ʠñɶ  Han Yu ʠÒ Manuscript scribed at the Yishuang louáųĬ in the Qing NLC* Weng Tonghe 173 Zengguang zhushi yinbian Tang Liu xiansheng ji o·Ňɯʡɑ[ę/Ǝʖ 43 Liu Zongyuan ę‰- Woodblock edition carved in the early Ming NLC  174 Woodblock edition carved at the Shanjing tang `ðn in 1448, amended edition NLC Weng Tonghe 175 Liu Zonguan wenji ę‰-ñʖ 48 Liu Zongyuan ę‰-; Liu Yuxi =ƪɾ, comp. Woodblock edition carved by Mo Rushi Ǹ|q of Xinhui in óć 1556 NCL Jin Fengxiang ɳʶǚ 176 Wang Jingshi xiansheng piping Han Liu wen ŹǷƞ/ƎÞȧęñ 12 Wang Xijue Źɾű Ming woodblock edition NLC  177 Liu Binke ji =ɇ‹ʖ 2 Liu Yuxi Woodblock edition carved at the Yexiang tang ɱʬn in the Kangxi reign SL* Yao Shiyu 178 Liu Binke shiji =ɇ‹ȩʖ 9 Liu Yuxi; Zhao Honglie Ɋʸũ, comp. Woodblock edition carved at the Hanbi zhai œơ˂ in 1723 PKU Unknown  257  Title No. juan Contributors Edition Location Transcriptionist(s) 179 Tang Ouyang xiansheng wenji [Ķʎ/Ǝñʖ 8 Ouyang Zhan Ķʎȭ Woodblock edition carved by the Xu à family of Jinling ɳʋ in 1606 NJ Han Zong ʠ‰ 180 Ouyang Xingzhou wenji ĶʎȐYñʖ 10 Ouyang Zhan Ķʎȭ Ming woodblock edition NLC Wu Zhuoxin UD" 181 Meng Dongye shiji ƒĒɱȩʖ 11 Meng Jiao ƒɤ Woodblock edition carved in Jiahe aƫ in the late Ming PKU Yao Shiyu }ɷ 182 Zhou He shiji YɅȩʖ 1 Zhou He YɅ Ming manuscript NLC  183 Jia Langxian Changjiang ji ɆőʂĿʖ 10 Jia Dao Ɇž Woodblock edition carved at the Qinchuan shuwu Ɔ¢Ąš in 1702 FSNL Jiang Gao ȁē 184 Woodblock edition carved in the Kangxi reign NLC Unknown 185 Manuscript by Lu Wenchao in the Qianlong reign NCL Lu Wenchao 186 Jia Changjiang shiji ɆʂĿȩʖ  Jia Dao Ɇž Ming manuscript NLC  187 Changgu ji ùȿʖ 4 Li He ďɅ; Zeng Yi ĆƗ, ann. Ming woodblock edition SL* Unknown 188 Li Changji Changgu ji ju jie dingben ďʂPùȿʖLȠŠč 4 Li He ďɅ; Yao Quan }, et al., anns. Woodblock edition carved in the early Qing SL* Chen Benli ʊčƩ 189 Woodblock edition carved at the Meicun shuwu ğɢĄš in the early Qing NLC* Shao Rui ɣɸ 190 Geshi bian ĵȩǎ 4 Li He ďɅ Woodblock edition carved at the Yinshu lou cĮĬ in the Qing NLC* Zhang Yu  258  Title No. juan Contributors Edition Location Transcriptionist(s) 191 Yuanshi Changqing ji -ĻʂÖʖ 60 Yuan Zhen -ƴ Woodblock edition carved in Ma Yuandiao’sʭ-ȳ Yule xuan ʳĪɍ 1604 NLC*  192 Bai Xiangshan shi ƕʬœȩ 41 Bai Juyi ƕ™û Woodblock edition carved at Wang Liming’s ŀƻR Yiyu caotang ʏǶn in the Kangxi reign NLC* Unknown 193 Li Weigong wenji ďȒ4ñʖ 34 Li Deyu ďÇȗ Ming manuscript NLC  194 Fanchuan wenji ī¢ñʖ 22 Du Mu đĖ Ming woodblock edition SL*  195 Li Yishan shi lu ďǖœȩɿ 3 Li Shangyin ď\ʒ; Zhu Heling Ďʹ˃, ann. Woodblock edition carved by Ye Yongru ǻľǵ of Jinling in 1659  SL* Shen Rixun łöž 196 SL* Unknown 197 Li Yishan shiji ďǖœȩʖ 3 Li Shangyin ď\ʒ Woodblock edition carved by Ye Yongru of Jinling in 1659 SL* Unknown 198 Woodblock edition carved at the Guangzhou Cuishu ·£$Ǔ in 1870 PKU Unknown 199 Li Yishan shiji ďǖœȩʖ 8 Li Shangyin ď\ʒ Woodblock edition carved by the HuǤ family of Haiyan Œʺ in the Kangxi reign NCL Unknown 200 Chongding Li Yishan shiji jianzhu ɰȢďǖœȩʖǀȦ 6 Li Shangyin; Zhu Heling Ďʹ˃, et al., ann. Woodblock edition carved in Wang Zengning’s ŀo’ Dongke caocang ĒĘǶn in 1744 NCL Unknown 201 Yuxisheng shi jianzhu ŸɀƎȩǀȦ 5 Li Shangyin; Feng Hao ʮŐ, ann. Woodblock edition carved at the Deju tang in 1767 SL* Mao Chen ĺƃ    259  Title No. juan Contributors Edition Location Transcriptionist(s) 202 Wen Feiqing shiji ŝʩHȩʖ 9 Wen Tingyun ŝ±ƿ; Zeng Yi ĆƗ, ann. Woodblock edition carved at the Gu’sʨ Xiuye caotang ƬɱǶn in 1967 SL Tang Yuanqi Ś-Ǯ, Yu Yuanfu ʳ-% 203 HB Unknown 204 Chongkan jiaozheng Lize congshu ɰ9ěķƽťȇĄ 6 Lu Guimeng ʍˆǿ Woodblock edition carved at Gu Jian’s ʨģ Biyun caotang ơƿǶ in the Qing SL Unknown 205 Yuntai bian ʙǨǎ  Zheng Gu ɫƷ Woodblock edition carved in the Jiajing reign NLC Zhou Shutao 206 Shiji ȩʖ 3 Li Dong ďŊ Ming manuscript NLC  207 Zhang Pin shiji ¼ȏȩʖ 1 Zhang Pin ¼ȏ Ming manuscript NLC  208 Su Xueshi wenji ȋ…qñʖ 16 Su Shunqin ȋǬĴ Woodblock edition carved at the Baihua shuwu ƕǹĄš in 1698 SL* Gu Guangqi 209 SL* Qian Taiji 210 SL* Yao Shiyu 211 SL* Unknown 212 SL* Unknown 213 NLC Fu Zengxiang 214 Zhijiang Li xiansheng wenji ƛȺď/Ǝñʖ 41 Li Gou ďȝ Woodblock edition carved by Sun Du „Ə in 1518 and amended by Meng Shaoqing ƒljÖ in 1589 NCL  215 Yuanfeng lei gao -Ɂʧƶ 50 Zeng Gong Ćʞ Woodblock edition carved in 1447 and revised in the Chenghua reign SBL  216 Woodblock edition carved by Wang Shu ŹÌ  in 1544 NLC* Wu Cipei UÔl 217 Woodblock edition carved by Huang Xixian ʽ¥× in 1562 NLC Gu Zhikui ʨ	ɘ 218 NLC* Shi Quanlin ƞŅĕ  260  Title No. juan Contributors Edition Location Transcriptionist(s) 219 Yuanfeng lei gao 50 Zeng Gong Woodblock edition carved by Shao Lian ɣµ in 1571 NLC* Fu Zengxiang %oř 220 SA  221 HC Fu Zengxiang 222 Woodblock edition carved by Zeng Mincai ĆìÝ in 1597 NJ Gui Zhaojian Ĺ.DŽ 223 SD Gui Zhaojian 224 Woodblock edition carved by Zeng Mincai in 1597 and amended by Zeng XianĆ/ in 1568 TJ  225 Woodblock edition carved at the Yupu shuyuan of Cili in 1890 NJ Wu Cipei 226 Nanfeng xiansheng Yuanfeng leigao EɁ/Ǝ-Ɂʧƶ 53 Woodblock edition carved by Gu SonglingʨĔ˃ of Changzhou ʂō in 1717 SL*  227 Jiayou ji aƤʖ 15 Su Xun ȋŎ Qing manuscript NLC Weng Tonghe 228 Shi zhu Su shi õŇȋȩ 42 Su Shi ȋɎ; Shi Yuanzhi õ-	, ann. Woodblock edition carved by Song Luo‡Ŵ in 1699 NLC* Chen Jingyun ʊĀʙ 229 Luancheng ji IJkʖ 50 Su Zhe ȋɏ Ming woodblock edition NLC  230 Houshan xiansheng ji œ/Ǝʖ 30 Chen Shidao ʊ§ɛ Woodblock edition carved by Ma Dun ʭā in 1499 SL* Tang Renshou [r 231 Luchuan ci Ȋ¢Ȩ 2 Zhang Yuangan ¼-ª Ming manuscript NLC  232 Yishan xiansheng wenji ɠœ/Ǝñʖ 10 Yuan Haowen -{] Woodblock edition carved in the Ming NCL   261  Title No. juan Contributors Edition Location Transcriptionist(s) 233 Yanyuan daixiansheng wenji <śÜ/Ǝñʖ 30 Dai Biaoyuan Üȕ- Manuscript NCL  234 Woodblock edition carved by Zhou YiY+ in 1573 NCL Huang Pilie ʽũ 235 Qiu Shancun yigao œĐɠƶ  Qiu Yuan ɜ Qing manuscript PKU  236 Sa Tiaoxi shiji Ȇxʀȩʖ 4 Sa Dula Ȇɩ; Woodblock edition carved at the Jigu ge of the Mao family in the late Ming SL  237 NLC Shen Yan  238 Jie Wenangong wencui èñ†4ñdž 1 Jie Xisi è'ò Woodblock edition carved in 1461 NLC  239 6 Woodblock edition carved in 1872 NLC Fu Zengxiang 240 Lü Jingfu shi liuzhong Xðzȩ5Ƴ 6 Lü Cheng XȰ Qing manuscript NCL  241 Zhang Laiyi wenji ¼+ñʖ  Zhang Yu ¼Ǘ Manuscript scribed at the Xiao langhuan fudi —~ƌƨi in the Qing dynasty NCL Huang Pilie 242 Qing manuscript NLC  243 Lei  bogao ʧFƶ 10 Yue Zheng ķ Ming woodblock edition NCL  244 Si xuan ji Íŷʖ 16 Sang Yue ĝÐ Woodblock edition carved in 1505 NCL  245 Zhushi wenji ƦĻñʖ 10 Zhu Zhishan Ʀėœ Manuscript by Xie Yong Ȼ³ in 1544 NPML  246 Gui Taipu wenchao Ĺy)ñɶ 1 Gui Youguang ĹĈ0 Qing manuscript SL Unknown    262  Title No. juan Contributors Edition Location Transcriptionist(s) 247 Wenxuan liuchen zhu ñɟ5ǦŇ 60 Xiao Tong ȃNj, comp., Li Shan ď`, et al., anns. Woodblock edition carved by Hong GengŌĦ in 1549 SL* Unknown 248 Wenxuan zuanzhu pinglin ñɟǑȦȧĕ 12 Xiao Tong, comp., Zhang Fengyi ¼ʵǜ, ann. Woodblock edition carved in the Wanli reign SL* Unknown 249 Wenxuan ñɟ 60 Xiao Tong, comp., Li Shan, ann. Woodblock edition carved at the Jigu ge of the Mao family in the Chongzhen reign SL* Wang Youdun ŀƐï 250 SZ  251 CQ  252 NLC* Unknown 253 NLC* Deng Chuanmi ɪ(‘ 254 LN Qi Junzao ƣ”ȉ 255 ZJU Zhu Yu ȶŬ 256 Woodblock edition carved at the Jigu ge of the Mao family and amended by Qian Shimiɼqȸ in the Kangxi reign HY Liu Shi’an=§†; Jiang Zonghaiȁ‰Œ 257 NJ Zhu Delin ƦÇʼ 258 UBC Unknown 259 ZJ Unknown 260 Woodblock edition carved at the Songshan shuwu  œĄš in 1760 TY Xiashan jushi ʛœ™q 261 Woodblock edition carved at the Huaide tang ÙÇn in the Qianlong reign FD Qian Yu ɼɷ 262 SX Jiang Jinhe ȁɽZ  263  Title No. juan Contributors Edition Location Transcriptionist(s) 263 Wenxuan 60 Xiao Tong, comp., Li Shan, ann. Woodblock edition carved in the Qing ZJ Dong Zengru ǽo, 264 CS Zhu Keshao ȶ1lj 265 Woodblock edition carved by Hu Kejia Ǥ1 in 1809 NKU Wang Tingzhen ŀ¹ż 266 Wenxuan ñɟ 24 Xiao Tong, comp. Woodblock edition carved by Wu Jinren Uɔ in 1595 ZJ Sun Qi „Ŕɻ 267 Manuscript by Liang Tongshu ĞQĄ in 1747 JLU Chen Shuhua ʊĮǹ 268 Zhongxing jian qi ji Ǫʆļʖ 2 Gao Zhongwu ʱĸ, comp. Woodblock edition carved in the Chongzhen reign NLC Unknown 269 Doushi lianzhu ji ƺĻǢŽʖ  Dou Chang ƺ© Woodblock edition carved at the Jigu ge of the Mao family in the late Ming PKU Unknown 270 Cai diao ji Ýȳʖ 10 Wei Hu ʟǏ Woodblock edition carved by Shen Yuruo łʘDz in the Jiajing reign SL Shen Baolian ł•Ȁ 271 Wenyuan yinghua bianzheng ñDZdzǹɐȽ 10 Peng Shuxia ÁIs Woodblock edition carved by Xiong Qi ŭș in 1614 SL*  272 Jianzhu Tang xian jueju santi shi fa ǀŇ[ɈNJLʰȩņ 20 Zhou Bi Y½, comp.; Yuanzhi gǧ, et al., ann. Ming woodblock edition NLC Unknown 273 Tang santi shi [ʰȩ5G 6 Zhou Bi Y½, comp.; Yuanzhi gǧ, et al., ann. Woodblock edition carved at the Langrun tang ĉţn in the Kangxi reign PKU  274 NCL  275 Zhongzhou ji £ʖ 10 Yuan Haowen -{] Woodblock edition carved at the Jigu ge of the Mao family in the late Ming NLC* Unknown  264  Title No. juan Contributors Edition Location Transcriptionist(s) 276 Tangshi guchui [ȩˁV 10 Yuan Haowen -{], comp.; Hao Tianting ɥxâ, et al., anns. Woodblock edition carved by Lu Yidian ʍɄ6 and Qian Chaoding ɼċʿ in 1659 CQ  277 NLC  278 Gu yin ȿʡ 2 Du Ben đč, comp. Woodblock edition carved in the Chongzhen reign PKU  279 Woodblock edition carved in Ming dynasty NLC Fu Zengxiang 280 Yushan mingsheng ji ŸœR?ʖ 2 Gu Ying ʨƈ, comp. Ming manuscript NLC  281 Tangshi ji cun [ȩLJ 15 Wu GuanUƄ, Fang Yiyuan ô-, et al., comps. Ming woodblock edition reprinted from Wu Guan’s UƄ edition NCL  282 Tangshi ji [ȩLJ 170 Wu Guan, comp. Ming woodblock edition NLC Fu Zengxiang 283 Tang yin wu qian [ʡÚDž 265 Hu Zhenheng Ǥʚ, comp. Woodblock edition carved at Hu Shenzhi’s ǤƑ	 Nanyi tang EƗn in 1686 SL*  284 SL* Shen Yan ł¡ 285 NCL Unknown 286 Tang sijia shi [bȩ 8 Wang Liming ŀƻR, comp. Woodblock edition carved in 1695 SZ  287 San Tang ren wenji [ñʖ 34 Mao Jin ĺþ, comp. Woodblock edition carved at the Jigu ge of the Mao family in the late Ming and amended in the Jiajing dan Daoguang reigns NLC* Zhang Yu 288 San Tang ren ji [ʖ 37  Woodblock edition carved in 1875 NLC Fu Zengxiang  265  Title No. juan Contributors Edition Location Transcriptionist(s) 289 Tang ren bajia shi [3ȩ 42 Mao Jin , comp. Woodblock edition carved at the Jigu ge of the Mao family in  1639 SL* Wu Peibo U 290 Tang san gaoseng shiji sanzhong [ʱ*ȩʖƳ 47 Mao Jin , comp. Woodblock edition carved at the Jigu ge of the Mao family in the late Ming SL Unknown 291 Tangren xuan Tangshi bazhong [ɟ[ȩ3Ƴ 23 Mao Jin , comp. Woodblock edition carved at the Jigu ge of the Mao family in  1628 PKU Unknown 292 NLC Wu Jingen UĀÏ 293 NLC Fu Zengxiang 294 SL* Unknown 295 SL*  296 Jin dai mishu ŋɗƥĄ 342 Mao Jin , comp. Woodblock edition carved at the Jigu ge of the Mao family in 1630 NCL   Abbreviations BNU Beijing Normal University Library CQ Chongqing Library CS Cishu chubanshe CWNU China West Normal University Library CZ Changzhou Library ECNU East China Normal University Library FD Fudan University Library FSNL Fu Ssu Nien Library, Taipei GS Gansu Library HB Hubei Library HC Hanchengxian Library  266 HY Harvard-Yenching Library JLU Jilin University Library LN Liaoning Provincial Library ML University of Macau Library NCL National Central Library (Taipei) NJ Nanjing Library NKU Nankai University Library NLA National Library of China NPML National Palace Museum Library, Taipei NSL National Science Library, Chinese Academy of Sciences PKU Peking University Library RY Rui’an Yuhai lou SA Suzhou Antique Management Office SBL Seikadō Bunko Library SC Sichuan Library SD Shandong Library SL Shanghai Library SX Shanxi Library SZ Suzhou Library TJ Tianjin Library TY Tianyi Pavilion UBC University of British Columbia Library WUL Wuhan University Library XA Xi’an Antique Management Office ZUL Zhengzhou University Library ZJU Zhejiang University Library ZJ Zhejiang Library  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0379407/manifest