Private Libraries and Library Catalogues: An Overview
The spread of printing (both geographically and across the social spectrum) in late Imperial China (1368â1911) was accompanied by a concomitant rise in the number of private libraries. In the Ming Dynasty (1368â1644), 897 people left written records of their book collections and collecting activities, more than those in the Song and Yuan Dynasties (960â1368) combined. Of these documented collectors, 231 allegedly owned more than 10,000 fascicles, and at least eight more than 70,000 fascicles. In the Qing Dynasty (1644â1911), the number of documented private libraries rose to 2082, more than in all previous dynasties combined; of these, 543 contained more than 10,000 fascicles, and at least 74 contained more than 100,000.1 The proliferation of books led not only to considerable growth in the size and number of private libraries, but also to an expansion of the range of books they contained as well as the social and cultural functions they assumed: as a locus of knowledge production, a site of scholarly collaboration and exchange, an expression of the private self, a gateway to prestige and power.
As Walter Benjamin compellingly observed, âif there is a counterpart to the confusion of a library, it is the order of its catalogueâ.2 To chart the rapidly evolving world of books, collectors compiled catalogues increasingly varied in format, elaborate in organisation, and diverse in purpose. There are 167 known catalogues of private libraries from the Ming Dynasty, of which forty-eight survive in full manuscript or printed copy; and of the 670 known library catalogues from the Qing Dynasty, 350 have survived.3 This chapter focuses on catalogues of private libraries in the long seventeenth and eighteenth century (roughly from the 1550s to the 1810s). The first part of the chapter presents an overview of these extant catalogues, outlining their various formats, their organisational structures, and their different types. Some are shelf-lists that functioned as finding mechanisms, while others focus on rare books. Some catalogue books by subject, or genre, or association with geographical area or historical period. Some were compiled for inheritance purposes, others are legal documents (e.g. inventories of confiscated properties of fallen officials), and yet others were compiled for commercial purposes (e.g. sale catalogues of private libraries, often made by the owners and circulated among friends and colleagues).
Most of these extant library catalogues are simple inventory lists, with entries that provide only the title of the book and the number of its volumes or fascicles. A few eccentric collectors in the late Ming Dynasty, such as Qi Chenghan (1563â1628) and Xu Bo (1563â1639), experimented with the table-format. Though this practice never gained momentum among private collectors, it inspired some imperial librarians in the eighteenth century, who adopted the table-format when drawing up a shelf catalogue for the magisterial âComplete Library of the Four Treasuriesâ, a collection of 3,461 books in 36,381 volumes compiled by 361 scholars from 1773â1782 under the imperial edict of Emperor Qianlong (1711â1799). Perhaps the most peculiar library catalogue from the late imperial period is the one compiled, or more precisely, composed by Gu Guangqi (1766â1835) for the collection of rare Song Dynasty (960â1276) editions assembled by his friend and fellow bibliophile, Huang Pilie (1763â1825). From 1792 to 1802, drawing upon his profound knowledge of the book trade (as a bookseller and publisher himself) as well as his extensive networks with collectors, book dealers, and scholar-officials, Huang managed to acquire more than one hundred fiercely sought-after Song imprints. To commemorate this feat, he invited Gu Guangqi to catalogue these treasures, not in lists or tables but in rhymed parallel prose, a highly formal and ornate genre in the Chinese literary tradition. The result is both a literary curiosity and a delightful variation on traditional cataloguing practices. Restricted by rhyme schemes, metrical patterns, and sentence lengths, Guâs beautifully worded catalogue, however, does not make an easily consultable record, nor does it contain much bibliographical information. To redress this lack, Huang Pilie added an extensive line-by-line interlinear commentary, detailing full titles, date of publication, size and style of script, number of characters in each column, number of columns per page, and the presence of commentary. Sometimes he included anecdotes about a bookâs provenance or publication history, how he came across it, how it compared with other editions, and so forth. This rhymed catalogue, complemented by Huangâs expert commentary, constitutes one of the most singularly formatted, thoroughly annotated bibliographical works in pre-modern China.
In another contribution to the present volume, Joseph Black notes that the main sources used for the Private Libraries in Renaissance England (PLRE) project are inventories, catalogues, and wills, with other records occasionally edited from transportation lists, donations, purchase records, and lists of books seized in raids. The variety, though certainly not the quantity, of sources available for Chinese book historians of the late imperial period is more limited by comparison. To begin with, there are few extant recordsâwills or other legal documentsâabout bequests of books. Since primogeniture was not the norm, properties were usually divided up between the male children upon the death of the patriarchâthat is, if the family chose not to stay as a unified unit. But, so far as I know, the only documented case of an inventory of books explicitly and exclusively compiled for inheritance purpose was the one made by Wan Shihe (1517â1587), which, unfortunately, does not survive.4 Was a library customarily divided equally among the sons? Or would the eldest boy receive the largest proportion? Or maybe the entire library would go to the son with the strongest scholarly tendencies, regardless of seniority? Without relevant records, we can only speculate. Of course, some collectors, like many of their western counterparts, abhorred the idea of breaking up their libraries. For instance, Fan Dachong (1540â1602) stipulated in the family rulebookâa ritually binding document blessed by the ancestors and endowed with tremendous moral power in late imperial Chinaâthat the family library would be collectively owned and managed by all family members, and under no circumstances be divided. Against all odds, through five hundred years of social upheavals and political turmoil, his order was followed by every succeeding generation, until the government took the collection over in 1949.5 Other collectors, in fear of their books falling into the hands of âunfilial sonsâ (meaning sons who did not share their fatherâs bibliographic obsession) choose to donate them instead to clan schools. This way, their books could remain in the extended family and have a better chance of survival. But records of donations, too, are scarce, and the only one we know of is the booklist compiled by Sun Xingyan (1753â1818) upon his bequest of part of his private library to the Sun lineage school, a list that will be discussed in greater detail below.6
Late imperial China certainly had its share of disgraced officials, and the properties seized upon their fall constituted a semi-regular source of revenue for the imperial court. The recordsâadmittedly scantyâof these seizures also provide us with valuable information about material culture of the period and the book collecting habits of some of the richest, most powerful individuals China has ever seen. For example, when the estates of Yan Song (1480â1567) were confiscated, the executors compiled an inventory that contains more than 60,000 words.7 As expected, pride of place goes to jewellery, antiques, and works of art, but the inventory also lists eighty-eight rare books, among which are forty Song imprints (now we can better understand the pride Huang Pilie took in his 109 Song editions), thirty Yuan (1279â1368) imprints, two high-prestige historical works containing 277 and 544 fascicles respectively, first Ming editions of two canonical historical works, and a collection of rare manuscripts on occult subjects. While these rare books were listed individually, in the same manner as paintings and calligraphy scrolls, the common editions in Yan Songâs possession were summarily lumped together: we are told merely that he also owned 5,852 sets of Confucian classics, history, philosophy, and literature, and 914 sets of Buddhist and Daoist scriptures. The cursory treatment of these common editions suggests the relatively ready availability and affordability of books in early sixteenth-century China: the executors seemed neither surprised nor impressed by the size of Yanâs library. Interestingly, while the executors noted that the scriptures would be donated to temples âto be chanted in worshipâ, and the 5,852 sets of non-religious books distributed among government schools across the empire, the destination of the rare books was left unspecified. We do not know if they were inventoried separately from the common editions because Yan Song stored them separately, or because, given their exponentially higher monetary value, the executors considered them less books than artefacts, and therefore more on a par with the antiques, works of art, and jewelry. In any case, the inventory provides an early instance of bibliographic distinctions being made between rare imprints, manuscripts, and common editions, a practice that would become more common over the next two centuries and eventually become a cataloguing norm.
Indeed, as Hilde de Weerdt notes, none of the three extant catalogues of private libraries from the Song Dynasty specify if a title listed was published in print or manuscript form.8 But, over the course of the seventeenth and eighteenth centuries, boosted by the emerging interest in philology and the newly felt urgency to restore the textual integrity of foundational Confucian texts, scholar-collectors became increasingly attuned to nuanced differences between editions and between imprints and manuscripts, textual discrepancies having proved to have paradigm-shifting implications for contemporary understandings of Confucian orthodoxy. In Principles of Book Collecting, the first systematic treatise on book collecting in China, Sun Congtian (1692â1767) suggests that collectors record âwhether a book is a Song edition, a Yuan edition, a Ming edition, a contemporary edition, a manuscript copy of Song or Yuan date, an ancient manuscript copy, a manuscript copy of Ming date, or a recent manuscript copyâ.9 To further foreground the exceptionality of manuscripts and rare imprints, he advised those with the financial means to store their rare books in a separate room, or better still, a separate building, and to compile a special catalogue apart from the general inventory:
With rare books, a book collector should clearly indicate whether his books were published in the Northern Song or in the Southern Song Dynasty, or whether they were printed in the Song, Yuan, or Ming Dynasty (from the blocks originally cut in Song times); he should also record the colophons, seals, and names of previous collectors, and whether the books are complete or incomplete, collated or uncollated. The same applies to Yuan editions, which must be kept in a separate cabinet. Their cataloguing should be done in the same manner. The cabinets, containing Song and Yuan books, must be locked and should not be opened without your permission. Finely transcribed manuscript copies, ancient manuscript copies, manuscript copies of Song and Yuan dates, and rare books are also to be catalogued in the same style. He must distinctly record the names of the scribes, and the colophons, seals, and names of previous collectors to whom they once belonged, and indicate also whether a given book is complete or not, whether it is ânot for lendingâ, whether it is a traced manuscript copy of a Song edition, and whether the original printing-blocks still exist or not. If the book is collated, write down the name of the collator; also indicate whether the book is an original copy or a transcription. Enter all these items of information in one volume.10
Containing information about scribes and collators, transcriptions of colophons left by previous owners, and sometimes the textual affiliations between different editions, this model of catalogue offers a trove of copy-specific data, illuminating the features and histories of rare books that were not just texts but material artefacts whose value lay not only in their contents but also in the quality of the paper used, the aesthetic appeal of the script, the colour, texture, and smell of the ink, and the cultural prestige of those who once owned and read it. The increasingly elaborate differentiation between manuscripts and imprints also sheds light on the evolving relationship between manuscript and print, the convoluted web of their interpenetration and interdependence formed during long co-existence. Indeed, the discussion of nuanced distinctions and affiliations between manuscripts and imprints was only made possible, and necessary, by a print culture sufficiently well-established to have assumed its own defining characteristics.11
A few catalogues of rare books appeared in the first half of the seventeenth century, including The Catalogue of Books of Song and Yuan Dates in Chuanshi Tower by Xu Qianxue (1631â1694), The Catalogue of Books of Song Date in the Possession of Ji Cangwei by Ji Zhenyi (1630â?), and The Catalogue of Books of Song Date in Shugu Hall by Qian Zeng (1629â1701). But these catalogues contain only minimal information on the bibliographical features of individual copies. Interestingly, the first annotated catalogue of rare books that includes extensive and systematic notes along the lines suggested by Sun Congtian is a sale catalogue drawn up by Mao Yi (1640â1713) between 1699 and 1708 when he was negotiating an en bloc sale of some books in his family library. To advertise their exceptional merits (and justify his high asking prices), Mao Yi provided each item with a detailed description of its condition, provenance, transmission, and distinctive bibliographical characteristics. One innovation he initiated was the hitherto unprecedentedly refined treatment of different types of manuscript copies. Mao Yi divided manuscripts into ancient manuscripts and contemporary manuscripts, with the latter being further subdivided into six categories, traced manuscripts of Song or Yuan books, regular manuscript copies of Song or Yuan books, fine manuscripts, and manuscripts by famous hands. Initially designed as a pricing strategy, these divisions were later adopted by literati collectors in cataloguing their rare books. Though motivated by commercial concerns, the format and approach of The Catalogue of Treasured Rare Books of Jigu Pavilion exerted a profound influence on textual and bibliographical scholarship in the mid- and late-Qing Dynasty, and contributed to the formation of an âedition-oriented mentalityâ which set collectors and scholars of this period apart from their medieval counterparts.
Scholar-collectorsâ fascination with rare books, as reflected in their cataloguing practice, was also a byproduct of the proliferation of print from the sixteenth century onwards. As books became more readily available to a broader section of the population, the mere act of possessing them ceased to function as an adequate marker of cultural sophistication. The quality of the books in oneâs collection, rather than their sheer number, increasingly came to define the intellectual status of the owner and differentiate him from undiscriminating cultural imposters. Having differentiated âbooks easy to acquireâ from âbooks hard to come byâ in the opening of âAn Essay on Tianyige Libraryâ, Huang Zongxi (1610â1695) proceeded to ridicule those who failed to recognize this distinction, and expelled from the ranks of book lovers those who boasted about owning a sizable library, which in fact contained only âcommercially produced books that could be easily bought in bookstoresâ.12 In a similar vein, Cao Rong (1613â1685) in âCovenant on the Circulation of Ancient Booksâ lamented the downside of the flourishing of print and the consequent overabundance of books, which, like âa blinding haze of coal smokeâ, clouded peopleâs judgement. To search for rare editions among the profusion of commercially available imprints had become as challenging as âcollecting precious jade in the deep mountainsâ.13 As a result, it became imperative for the cultural elite to reassert their privileged position by rewriting the rules of book collecting, shifting the locus of prestige from the size of a library to the quality and rarity of its holdings. To own a complete set of Confucian classics no longer marked one as culturally refined; to own it in a Song or Yuan edition did.
Apart from special catalogues of rare books, scholar-collectors also compiled separate catalogues for books associated with a certain place, published in a certain period, or most common of all, on a certain subject. For instance, Mei Wending (1633â1721) compiled a list of works on astronomy and mathematics; Zhu Yizun (1629â1709) meticulously catalogued Confucian texts; and Lv Tiancheng (1580â1618) and Qi Biaojia (1602â1645) drew up annotated catalogues of vernacular plays, which include such detailed analyses that they were (and still are) read in their own right as critical studies of drama. The practice of cataloguing books by subject dated back to the first century when Liu Xin (c.50 BCEâ23 CE), a curator of the imperial library, adopted a classification system that divided books into six major categories: Confucian classics, philosophy, literature, military, astronomy/geomancy/mathematics, and medicine/alchemy. This classification scheme was eventually replaced by the sibu system in the seventh century, which divides books into four categories, namely Confucian classics, history, philosophy, and literature. Sibu was by far the predominant method of classification throughout pre-modern China.14 Despite various revisions and expansions over succeeding centuries, it remained the basic framework for the organisation of books and knowledge in China until the late nineteenth century, when Western learning began to challenge traditional Chinese taxonomies.15
How were books arranged on the shelves, or, as some collectors preferred, in cabinets? In The Ten Commandments for Book Collecting, Ye Dehui (1864â1927), one of the last classically trained bibliographers of the old order, summarizes traditional advice on how to arrange oneâs library. According to Ye, a library should be divided into six sections: Confucian classics, history, philosophy, literature, collectanea (a category discussed below), and rare books. Within each section, books would be arranged chronologically, with those belonging to the same school of thought next to one another for easy consultation. In the case of anthologies and collections, the date of composition takes precedence over the date of compilation. Anthologies of writings by people hailing from the same city or county should be grouped according to geographic proximity. Besides subject matter, chronology, geography, and edition, size should also be factored in. Pocket books that measure less than five tsâun (about the size of a European octavo) and large-format books that measure more than one chi (folio-sized) should be put in specially designed cabinets. Finally, to highlight their political and cultural significance, the following books should be stored apart from the rest of the collection: imperially commissioned editions, Buddhist and Daoist scriptures, and books on western learning.16
In the late imperial period, scholar-collectors created new categories in their catalogues to accommodate the increasing variety of available books. For example, collectanea was first proposed as an independent subcategory by Qi Chenghan (d. 1628) in The Four Principles of the Organization of My Library in the Year of Gengshen. As Qi observes, âcollectanea did not constitute a subcategory in their own right in ancient times, but they have been rapidly growing in number and popularity these daysâ.17 Unlike traditional anthologies that gathered published works in a given genre or domain, from a given time period, or by authors originating from the same place, collectanea yoked titles of varied natures and origins without necessarily presenting any thematic, chronological, or geographical coherence. They also differ from leishuâoften translated as âclassified encyclopediaâ for lack of a better termâin that they contain texts in their entirety instead of excerpts. In effect, collectanea are designed to facilitate the accumulation, preservation, and circulation of texts on a broad range of subjects rather than to organize and manage information on a given subject. With their heterogeneous accumulations of titles, collectanea defied integration into existing categorisation frameworks and called for innovative cataloguing strategies. Qi Lisun (1625â1675), a grandson of Qi Chenhang, further revised his grandfatherâs system of classification. In his own library catalogue, he elevated the status of collectanea from a secondary category under âphilosophyâ to a primary category on par with classics, history, philosophy, and literature. The emergence of collectanea as an independent secondary category in the early seventeenth century and its quick rise to the level of a primary category attested to the steadily growing popularity, availability, and intellectual prestige of these comprehensive compilations.
Other collectors adopted idiosyncratic subject classification systems that reflected their intellectual interests, moral convictions, and political stances. For instance, the library catalogue of Mao Kun (1512â1601), a military strategist and Confucian scholar, would have confounded many of his contemporaries. After establishing his reputation as a major literary scholar with his immensely popular Anthology of Essays by the Eight Masters of the Tang and Song, Mao embarked on a brilliant military career characterised by strategic daring. In the final years of the reign of the Emperor Jiajing (r. 1522â1566), he served as counselor to Hu Zongxian (1512â1565), the powerful general-minister in charge of maritime campaigns against Japanese pirates. His dual accomplishments in literature and military strategy were mirrored in both the contents of his library and the way it was catalogued. Apart from titles in the categories of Confucian classics, history, and literature, he owned a substantial number of books on military art, mathematics, diplomacy, and the so-called âconcrete studiesââa rising intellectual movement in the late Ming that emphasised statecraft, administrative practicality, and, as its name suggests, concrete political action as opposed to moral-metaphysical speculations; the category of âphilosophyâ is a conspicuous absence in his library catalogue.18 To further indicate his endorsement of this emerging school of thought, Mao assigned books on âconcrete studiesâ to a separate category on an equal footing with the more established categoriesâa defiant bibliographic gesture in keeping with his character and his successful political and military career.
Mao Kun was not alone in expressing intellectual sympathies through defiance of bibliographic conventions. Instead of following the traditional sibu system, Lu Shen (1477â1544) categorised his books under thirteen headings: Confucian classics and their related commentaries; works of neo-Confucian scholars; official histories; ancient texts that do not fall into the categories of classics and histories; philosophy; prose collections; poetry collections; encyclopedias; miscellaneous histories; geography; rhyme books; philology and medicine; divination, geomancy, mathematics, astrology, alchemy, and Daoist mysticism.19 So far as I know, he was the only book collector in late imperial China to combine philology and medicine into one category. His justification was that âgreat philological achievements can only be attained by those who have studied since an early age, just as fast recovery from illness can only be achieved by those who seek medical help promptly. The same rule applies to both disciplinesâ.20 Luâs unorthodox classification scheme was in keeping with his nonconformist approach to learning. On one occasion he observed that âancient books should be read and used to serve oneâs own interests and purposesâ, and a scholar should not blindly follow popular opinions as âa short person who cheers with others in the audience without even knowing what is happening on the stageâ.21 Evidently he did not see the need to follow the popular sibu method either.
Even those who adhered to the sibu classification framework sometimes introduced refinements that reflected their own intellectual agenda. For instance, Zhu Mujie (1518â1587), a member of the Ming imperial family, followed the sibu system by organizing his books into four categories: Confucian classics, history, philosophy, and literature. But his conformity to bibliographic conventions ends here. Thirteen Confucian texts had been canonised as âThe Thirteen Classicsâ since the end of the Song Dynasty. By the time Zhu Mujie built his library, it had become a norm to divide the category of âConfucian classicsâ into thirteen subcategories, one for each of these texts and their extensive commentary traditions. But Zhu Mujie divided his Confucian classics into only eleven subcategories, incorporating books on âThe Gongyang Traditionâ and âThe Guliang Traditionâ into the subcategory of âThe Spring and Autumn Annalsâ instead. The downgrading of these two culturally significant texts marked a contentious intellectual position that would not go unnoticed by Zhuâs fellow scholars. Among other unconventional alterations, the most intriguing is Zhuâs treatment of The Book of Music, an early Confucian text lost during the massive book-burning launched by the first emperor of China (259â210 BCE). Zhu accords this work its own subheading. Of course, the lost Book of Music itself did not feature in the category. In fact, most of the titles listed under its namesake subheading bear no relation to The Book of Music, or even to Confucianism in general. Instead, they comprise ballads, anthologies of the musically oriented yuefu poetry, books on the melodic models and formal tunes of traditional Chinese drama, and collections of sanqu, a form of poetry whose tonal patterns derive from folk songs. To place vernacular genres rooted in popular culture under âConfucian classicsâ was deeply provocative in the sixteenth century. In late imperial China, it was a commonplace that music (at least the right kind of music) was endowed with a morally transformative power. Many Confucian scholars devoted themselves to the elaboration of the relationship between music, morality, and governance. Putting musically oriented vernacular literature side by side with Confucian classics, Zhu Mujie argued for the potential affinity between the two, and by implication, pointed to the civilizing and moralizing influence of vernacular works, works customarily considered as cheap entertainment that corrupted the young and misled the ignorant.
Whereas Zhu Mujie promoted works of vernacular literature by classifying them under the heading of Confucian classics, Gao Ru (c.1514â1563) exalted them to the category of âhistoryâ, the second most prestigious in the sibu scheme. In the annotated library catalogue he compiled from 1534 to 1553, Gao created three new subcategories under the heading of history, namely âunofficial historyâ, âprivate historyâ, and âshort historyâ.22 Fifty-nine plays appear under âunofficial historyâ, thirteen novellas, collections of short stories, and librettos under âshort historyâ, and two historical romancesâRomance of the Three Kingdoms and The Water Marginâunder âprivate historyâ. In the early sixteenth century, simply to own seventy-four vernacular works would be considered eccentric, and to categorize them as âhistoryâ was a bold cultural statement. Gao justified his categorisation by foregrounding thematic connections between traditional histories and historical romances. In the entry on Romance of the Three Kingdoms, he notes that:
(The book) is compiled by Luo Guanzhong (c.1330â1400) on the basis of Sanguo zhi (The Records of the Three Kingdoms) written by Chen Shou (233â297). Deriving from official histories and incorporating fictive and anecdotal sources ⦠it is therefore neither vulgar nor fake, and is easy to read and understand. It chronicles tens of thousands of historical events that happened in the span of a hundred years in a style devoid of both the ancient solemnity of official historians and the flippancy of oral storytellers.23
Interestingly, The Records of the Three Kingdoms, the most authoritative historical work on the three-kingdom period (184/220â280 CE) and a staple in private libraries in late imperial China, was not listed in Gaoâs catalogue. Its absenceâall the more striking given his family background in the Imperial Guard and his personal interest in military strategiesâleaves Romance of the Three Kingdoms as the only work he owned that systematically treats the subject. One cannot but speculate that Gao Ru considered Romance of the Three Kingdoms, its literary and fictive embellishments notwithstanding, to be a legitimate historical substitute for The Records of the Three Kingdoms. Putting Romance of the Three Kingdoms and The Water Margin in the same category as official histories, Gao Ru not only recognised the historicity of historical romances, but also implicitly called attention to the constructed literariness of traditional historical narratives. His classification scheme, anticipating the rise of vernacular novels, pointed up the permeability of generic distinctions between history and fiction.
Library Catalogues: Texts in Their Own Right
One revelation of these catalogues is that educated elites in late imperial China engaged with a range of books much broader than we tend to expect. Rather than poring over the prescribed philosophical works, histories, and classical literature central to the Confucian curriculum and dominant in contemporary discourses, they owned books in areas as diverse as medicine, law, geography, cartography, astrology, divination, geomancy, mathematics, agriculture, architecture, and military studies. These books, existing outside of the state-patronised Confucian orthodoxy and certainly beyond the expertise of a typical Confucian scholar, nevertheless appeared with great consistency in private libraries. Their presence reminds us that many scholars (at least those lucky enough to land official posts) were also governors, judges, diplomats, and, in times of war, generals and strategists. They were the administers of the empire, and they turned to their books for practical knowledge of the legal system, the construction of hydrological projects, and the logistics of grain transportation as much as for classical allusions or metaphysical insights. Catalogues of private libraries thus open up an unsuspected vista into the world inhabited by late imperial elites, who read more broadly than they let on in their own writings and who struggled to balance a contemplative life with a life of action.
Take for example the private library of Qian Qianyi (1582â1664), one of the foremost scholar-officials in his time. According to his library catalogue, Qian owned 2,894 books categorised under seventy-three headings, with books under each heading stored in a separate cabinet.24 Besides the usual Confucian classics, histories, literary works, and philosophical treatises, his library contained thirty-five works on mathematics, thirty on astrology, seventeen on divination, six on physiognomy, forty-eight on geomancy (an alarming number of which were dedicated to choosing an auspicious burial site), another forty-eight on an arcane branch of divination associated with the Book of Changes, 144 medical works ranging in focus from typhoid and surgery to pediatrics, midwifery, and obstetrics, five on translation (one on Korean, one on Mongolian, the rest on languages of the ethnic groups on the empireâs western frontier), nine on the geography, history and culture of Korea, nine on Vietnam, and five on Japan. Qian was also the first scholar-collector who classified books written or translated by Jesuit missionaries under a separate category: the previous norm was to incorporate such books into existing categories alongside their Chinese counterparts. Under âCatholicismâ, a heading that reflected the religious status of the authors more than the subject matter of their works, Qian listed seven titles, including two books written in classical Chinese by Matteo Ricci (1552â1610); an introduction to world geography by Giulio Aleni (1582â1649); a book by Diego de Pantoja (1571â1618) on how to overcome the seven sins; a phonetic Latin-Chinese dictionary compiled by Nicolas Trigault (1577â1628); a treatise on earthquakes by Nicolò Longobardo (1559â1654); and a large anthology comprising nineteen works by Jesuit missionaries, including ten scientific treatises and nine on religious and philosophical subjects. Apparently, Qian kept up to speed about the kind of work Jesuit missionaries were doing in China and the kind of ideas they were advocating. We do not know his position on Catholicism, or rather the version of it preached by the Jesuitsâcharacterised by considerable accommodations to Confucianismâbut the inclusion of their books in his library nevertheless debunks the stereotype of Confucian scholars as intensely inward-looking, deeply suspicious of the new and scared of the unfamiliar. Whether he endorsed or challenged the Jesuitsâ messages, he was prepared to engage them in conversation. His library catalogue also gives us some idea about the reach and influence of Jesuit missionaries among Chinese elites: if they had the attention of someone like Qian, they had found their way into the very centre of late Ming intellectual life. Qian Qianyi had the misfortune to live in interesting times, serving in important military roles through the collapse of the Ming dynasty and the ascension of the Manchu Qing: other books in his library reflect his active official career. He owned altogether 238 titles on military studies, covering subjects from routine drilling operations, the art of archery, and fire weaponry to river defense, maritime logistics and defense, coastal cartography, and hydrology. His is the library of a minister who performed administrative duties and conducted diplomacy, a military leader who organised armies and drew battle plans, as well as a scholar who wrote poetry and contemplated the moral teachings of ancient sages.
People do not always read the books they own; nor do they always admit to owning the books they read. So far as I know, not a single documented book-collector listed The Plum in the Golden Vase (the most infamous erotic novel in pre-modern China) in their catalogues, even though we know from their correspondence and diaries that they not only read the book but sometimes hand-copied it and wrote comments in the margins.25 Salacious literature was not the only category excluded. Contemporary works, especially single-authored literary collections, also rarely found their way into catalogues of private libraries. Once again, the problem was not lack of availability. Rather the opposite: literati customarily sent copies of their collected writings (often in manuscript or privately printed at their own expense) to friends, colleagues, and mentors. The profusion of laudatory colophons and prefaces they wrote for each other reveals the networks through which such titles were circulated, exchanged, and disseminated. From time to time, established scholars would even complain about the contemporary works that inundated their studios, most of which were presented by people seeking endorsements, comments, or patronage.
Clearly, catalogues of private libraries in late imperial China were not straightforward inventories. They documented not what a library actually contained but what its owner considered worthy, what he would like to be known for reading and collecting among peers and by posterity. In other words, these records were carefully curated to project an image, an image of a serious scholar, a capable official, a man of impeccable taste and flawless character. The self-fashioning efforts of book-owners turned out to be warranted. Catalogues of private libraries were much sought after and widely circulated (in both manuscript and print). By the sixteenth century, they had become a distinctive bibliographic category in their own right, one regularly featured in library catalogues. We know that Zhu Mujie, the advocate of vernacular literature discussed above, owned twenty catalogues of private libraries; the inventory of the book collection of Huang Yuji (1619â1691) contained forty-two. Then, as now, people wanted to know what others read, how they organised their books, and what titles or editions were once available but no longer extant. Far from being produced for private use, library catalogues were, from the start, intended for the public eye. For instance, Qian Qianyiâs library catalogue was widely circulated in manuscript before its publication in print in 1850. Due to his stature in the intellectual world, it soon became a fashion for scholars and collectors to copy it out (preferably in oneâs best handwriting) and use it as a buying guide. Over time, copies transcribed, collated, or edited by famous scholars became themselves hot commodities: in 2004, a copy of the catalogue transcribed and bearing a long colophon by Wu Yifeng (1742â1819) was sold for $62,400 at an auction in China. The circulation of private library catalogues also served a special function in late imperial China. In The Covenant on the Circulation of Ancient Books, Cao Rong (1613â1685) suggested that collectors exchange their library catalogues with one another to identify titles missing from their collections but in the possession of others. The owners of these books could then hire scribes to make finely transcribed manuscript copies and trade them for copies of books they lacked. His proposal was put into practice by a number of collectors in the seventeenth century, who formed âbook clubsâ and regularly met to exchange catalogues and transcriptions.
Book collectors in late imperial China compiled library catalogues to impose order on the unprecedented abundance of texts engendered by the advent of print. Cataloguing therefore constitutes a discursive means to conceptualize and organize the world of books. Through the arrangement of titles in catalogues (and on shelves), collectors set books in dialogue with one another, evoking significant comparisons and generating new meanings. Instead of merely functioning as inventories that document book ownership, catalogues of libraries in late imperial China served a broad range of didactic and epistemological purposes. In the first catalogue of the imperial library, Liu Xin mapped out the genealogies of scholarly disciplines, identified the origin and development of various schools of thought, and offered detailed analyses of the strengths and limitations of specific editions of the classics. The general principles of cataloguing which he laid down continued to influence scholars and bibliographers for the next two thousand years. Featuring elaborate classification schemes and sometimes extensive annotations, library catalogues in late imperial China were often read as standalone scholarly treatises. For early readers, they were texts in themselves, not just lists of texts.
Library catalogues were also used as study guides. For instance, young students often resorted to those with detailed entries for individual titles as a survey of the textual and intellectual field and the connections between different branches of knowledge. One scholar in the eighteenth century went so far as to argue that the study of library catalogues constituted âthe cornerstone of scholarshipâ, without which âyoung scholars would be at a lossâ and âall readings would be done to no availâ.26 Some catalogues were even designed with pedagogical purposes in mind. In 1800, Sun Xingyan compiled the Inventory of the Sun Family Library to categorize the books he donated to his familyâs ancestral templeâa place that doubled as the clan school. In the preface, he noted that âthe titles selected here are representative works of different schools of thought to be used for the education of young boys in the (Sun) clanâ.27 He then delineated the twelve categories into which he divided the books, the origin and evolutionary trajectory of each category, and their fluid relationship with one another. He elevated philology, epigraphy, astronomy, geography, medicine, and lawâall considered secondary disciplines in the sibu systemâto the status of primary categories on par with Confucian classics, a classification system that corresponded with his own intellectual interests and neatly mapped the curriculum he developed when teaching at the prestigious Academy of Gujing. Interestingly, to facilitate learning for beginners, Sun listed the titles within each category in order of increasing difficulty, while presenting the categories themselves in order of their decreasing relevance to Confucian classics. Catalogues like this not only impose order on the world of books, but also dictate the order in which they should be read.
Who Were They? The Demographics of Private Book Collectors in Late Imperial China
Scholar-officials dominated the landscape of book collecting in late imperial China. Of the 231 collectors who owned more than 10,000 fascicles in the Ming Dynasty, all were scholar-officials with the exception of five doctors, four merchants, three non-office-holding scholars, two publishers, one bookseller, one academy teacher, and twenty-six reclusesâan occupation popular among the culturally sophisticated and politically disillusioned. Among the 543 collectors who owned more than 10,000 fascicles in the Qing Dynasty, alongside the majority of scholar-officials were seventeen non-office-holding scholars, fourteen academy teachers, one bookseller, three professional book collectors, one bibliographer, eleven merchants, nine doctors, two painters, four diplomats, one translator, and interestingly, only two recluses.28 The demographics of book collectors in the Qing period reveals some of the broader social and cultural developments of the time, such as the increasingly significant role played by private academies in Qing intellectual life; the falling out of fashion of the recluse way of life; the expansion of contacts with foreign nations, which led to the rise in power of people in charge of dealing with them (diplomats, translators, export merchants etc.); the growing proportion of classically educated men who either opted out or were pushed out of a career in office; and the professionalisation of scholarship in general and book-collecting in particular. This last trend allowed for the emergence of professional book collectors like Bao Tingbo (1728â1814), who built a whole career on his library, publishing high-quality facsimile editions of the rare titles in his possession, cultivating networks with prominent scholars and officials, and even receiving an honorary degree for donating rare editions to the imperial court and helping out with the compilation of The Complete Library of the Four Treasuries, a pet project of Emperor Qianlong.
Records of womenâs book ownership from this period are considerably scantier than those of their male counterparts. However, there are 3,750 documented women writers in late imperial China, and if this number is any indication of female literacy, it would not be uncommon for women, especially those of genteel background, to be able to read fluently.29 Though no records survive of womenâs libraries or library catalogues, the plays, novels, paintings, and woodblock illustrations of the period give the impression that the presence of books was an expected feature in a gentry womanâs boudoir, suggesting that women were not only readers but also book-owners. The books they read, contrary to our assumption, were not confined to the usual conduct manuals that preach female virtues and dictate ladylike behaviour. Instead, they read broadly, educating themselves on subjects ranging from Confucian classics, literature and art to history, religion, and philosophy. We know from their memoirs and epitaphs that they were often tutored from an early age, and the curricula designed for girls were not much different from those for their brothersâat least until the boys reached the age when they started training for the imperial examination. The learning of these women, acquired over years of formal schooling and private reading, was most eloquently testified by the range and complexity of allusions in the tens of thousands of poems they have left us. It is unclear if unmarried women customarily had access to their fatherâs library or the family library. But in an episode in The Dream of the Red Chamber (written in the mid-eighteenth century) one young lady recounts that when she was seven or eight years old, she used to sneak into her grandfatherâs library to read plays and romancesâbooks deemed inappropriate for the innocent eyes of young girls. And Wang Zhenyi (1768â1797), one of the first female mathematicians and astronomers in China, inherited her grandfatherâs entire library (seventy-two cabinets of books in total). Some married women in this period enjoyed full access to their husbandsâ libraries. For instance, Shen Cai, the concubine of Lu Meigu (c.1761), served as his librarian, and as such was in charge of organizing and cataloguing his private collection. Many of Luâs books that still survive today bear astonishingly erudite and insightful colophons written by Shen in her exquisite calligraphy. Similarly, Yao Wanzhen, married to Zhang Rongjing (c.1802), shared her husbandâs bibliophile passion. Not only did Zhang name his library after his wife, but several of the rare Song editions in his possession were inscribed with Yaoâs handwritten poems, and stamped with her collectorâs seal. After marrying Liu Rushi (1618â1664) (the Chinese Veronica Franco), Qian Qianyi built her a two-storied mansion, with the first floor serving as their living quarters and the second as a library where his 2,894 books were stored in 73 cabinets. According to a memoir of Liu penned by one of Qianâs students, Liu was of great assistance to Qian in his scholarly work, often looking up references for him and fact-checking his manuscriptsâtasks that would have involved frequent use of his library. Their story deeply resonated with the late Ming fantasy of companionate marriage between intellectual equals, a fantasy in which the library was often figured as the ultimate locus of domestic bliss, where mutual love was made all the sweeter by a shared love of books.
On an individual level, catalogues of private libraries reflect a book-ownerâs intellectual affinities, political affiliations, interests, tastes, and ambitions; collectively, they shed light on patterns of book ownership, the emergence of new genres and new subjects (e.g. the titles introduced into China by Jesuit missionaries), and the rise and fall of intellectual movements, such as the ascendancy of philology over neo-Confucian moral philosophy over the course of the eighteenth century. How can we use these catalogues as sources that do not merely corroborate what we already know but reveal histories we did not suspect and challenge what we assumed? How do we read them not only as a trove of bibliographic information but also as stories of human drama, of love and friendship, loss and perseverance, obsession and pain? And how do we make the silences, erasures, and absences that lurk under deceptive comprehensiveness speak to us across the distance of time and space? This chapter ventures some thoughts along these lines and hopes to show that critically informed âreadingâ of library catalogues offers us valuable insights into a world of books inhabited by people who read many titles that we no longer read, and often read in different ways and for different purposes from our own.
These statistics are drawn from the invaluable lists of private library catalogues compiled by Fan Fengshu in Zhongguo sijia cangshu shi (Zhengzhou: Daxiang chubanshe, 2001).
Walter Benjamin, trans. Henry Zohn, Illuminations (New York: Schocken Books, 2007), p. 60.
Fan, Zhongguo, pp. 251â263, 428â475.
Fan, Zhongguo, p. 259.
Fan, Zhongguo, p. 310.
Sun Xingyan, Sunshi citang shumu (Beijing: Shangwu yinshu guan, 1936), p. 3.
Tianshui bingshan lu (Beijing: Shangwu yinshu guan, 1937).
Hilde de Weerdt, âByways in the Imperial Chinese Information Orderâ, in Peter Kornicki and Cynthia Joanne Brokaw (eds.), The History of the Book in East Asia (New York: Routledge, 2016), pp. 257â259.
Sun Congtian, Cangshu jiyao (Taipei: Guangwen shuju, 1968), p. 26. All translations in this article are my own.
Sun, Cangshu jiyao, pp. 27â28.
Hilde de Weerdt argues that the lack of distinction between manuscript and printed editions in library catalogues of the Song Dynasty indicates the ubiquity of print. That is, print âwas in and of itself not noteworthy in late twelfth- and thirteenth-century catalogsâ, and âwas mentioned only when a particular printed edition had unusual features in comparison to other manuscript or printed editionsâ. My study of private catalogues of the Ming-Qing period leads me to propose the opposite: instead of indicating the ubiquity of print, the lack of differentiation between manuscripts and printed editions in fact suggests that print was still in its infancy in the Song Dynasty and too underdeveloped for book collectors of that time to consider systematically the complex dynamics between print and manuscript culture. After all, print could not possibly be more ubiquitous in the Song Dynasty than in the Ming and Qing Dynasties, and the Ming and Qing period witnessed the development of a keen awareness of the distinctions between manuscript and printed editions. See De Weerdt, âBywaysâ, pp. 257â259.
Huang Zongxi, âAn Essay on Tianyi Pavilionâ, reprinted in Ye Changchi (ed.), Poems and Prose Accounts of Book Collecting (Shanghai: Shanghai guji chubanshe, 1999), p. 315.
Cao Rong, âA Covenant on the Circulation of Ancient Booksâ, reprinted in The Collection of Jueju Poems on Book Collecting, A Covenant on the Circulation of Ancient Books, The Covenant of Ancient Pleasure Club, and Ten Covenants of Book Collecting (Shanghai: Gudian wenxue chubanshe, 1957), p. 24.
The sibu system was first adopted in the bibliographical section of the official history of the Sui Dynasty (581â618) compiled by scholar-officials in the early Tang Dynasty (618â907).
See Lianbin Dai, âChinaâs Bibliographic Tradition and the History of the Bookâ, Book History, 17:1 (2014), pp. 1â50. According to Dai, among the earliest attempts to systemize both European and Chinese knowledge into a unified catalogue was Guyue Cangshulou shumu by Xu Shulan (1837â1902), who employed both traditional Chinese categories and newly introduced European ones.
Ye Dehui, The Ten Covenants for Book Collecting, reprinted in Collection of Jueju Poems, pp. 39â44.
Qi Chenghan, Four Principles of Organizing My Books, online: <ctext.org/wiki.pl?if=gb&res=757272&remap=gb>.
Zheng Yuanqing, Hulu jingji kao (Taipei: Guangwen chubanshe, 1969), pp. 317â319.
Bian Yongyu (ed.), Shigutang shuhua kao (Hangzhou: Zhejiang renming meishu chubanshe, 2012), vol. 2, pp. 115â116.
Shigutang shuhua kao, pp. 115â116.
Shigutang shuhua kao, pp. 115â116.
Gao Ru, Baichuan shuzhi (Shanghai: Shanghai guji chubanshe, 2005), pp. 58â63.
Gao, Baichuan shuzhi, p. 63.
Qian Qianyi, Jiangyunlou shumu (Beijing: Shangwu yinshu guan, 1936).
The earliest surviving edition of The Plum in the Golden Vase contains a preface dated to 1617. But the novel had been circulated in manuscript among a small coterie of literati elite since the late sixteenth century. In 1596, Yuan Hongdao (1568â1610) wrote to Dong Qichang (1555â1636), making inquiries about where to find a copy of the second half of the book so that he could make a copy himself. In his diary, Yuan Zhongdao (1570â1624), Yuan Hongdaoâs brother, commented on the artistic achievement of the novel with unreserved enthusiasm. Though Dong and both of the Yuan brothers were high-profile book collectors of their time, none of them mentioned The Plum in the Golden Vase in the catalogues of their private libraries.
Wang Mingsheng, Shiqi shi shangque (Beijing: Shangwu yinshu guan, 1937), p. 6.
Sun Xingyan, Sunshi citang shumu (Beijing: Shangwu yinshu guan, 1936), p. 3.
Fan, Zhongguo, pp. 168â187, 271â320.
Hu Wenkai, Lidai funv zhuzuo kao (Shanghai: Shanghai guji chubanshe, 2008), p. 12.