? This field displays the person or organization listed as the creator/ issuer/ manager of the item.
? This field displays the place where the item is created, published or collected.
? This field displays the subject from the title of the item or the subject with which the item is associated.
? This field displays the date when the item is created, published, issued or collected.

Chinese Rare Books Digital Humanities Empirical Research Platform

In the tradition of Chinese bibliography, one’s higher purpose is “to categorize and to explore the origin and development of a work.” Scholars distinguish and analyze the detailed meaning behind the lines of ancient works, first investigating the original work’s cause of creation, development process, and even secondary changes, to illustrate the historical background and intellectual exchange underpinning its various theories. As Qing historian Wang Ming-sheng (1720-1797) once said: “When reading books, the most essential thing is bibliography. If the bibliography is clear, then it is readable. If the bibliography is unclear, the reading will be haphazard” (from Discussion and Consideration for 17 Books of History, juan 7. The work of editing Chinese ancient books was detailed and complicated, roughly encompassing: identifying forgeries, compiling lost elements, collating different editions, finding errors and misinterpretations within the text and related literature, clearly understanding the meaning of a book’s content, and using bibliography clearly delineate the nature of each author’s intellectual work.

However, since Western academia was imported into China, in addition to the rapid development of science and technology, the construction of digital libraries has also become the normal state of book collection today. At the same time, using and exchanging digital documents over the internet has for scholars become an indispensable part of conducting research. For this reason, traditional bibliographic data must now possess a certain degree of standardization, and new catalog specifications are being formulated. Under this trend, when we examine the ancient Chinese texts currently stored in locations across the world, and compare each storage institution’s data cataloguing format, we discover that bibliographic data regulations based on Western academic traditions may be unable to fully meet the academic needs of the “Chinese character/language cultural sphere.”

Looking back on the relationship between research on Taiwanese historical documents and the rise of digital humanities, Fu Ssu-nien (1896-1950), founder of the Academia Sinica Institute of History and Philology, wrote about his institute’s “research objectives” in the preface of the Bulletin of the Institute of History and Philology (BIHP) (1928), proposing:

  1. Where scholarship can directly study materials, it will progress.
  2. Where scholarship can broaden its materials, it will progress. If it cannot, it will regress.
  3. Where scholarship can broaden its tools for research, it will progress. If it cannot, it will regress.
If we take these three guideposts as indices for the vigorous development of digital humanities, then we have indeed witnessed great expansion in scholarship and the broader social environment. Today, the goal of digital technology application is to assist scholars to more effectively collect and integrate research materials, to break through national and procedural barriers, to link research resources around the world, and, through new data insights, to attempt new methods to penetrate existing research questions, discover new phenomena, and open up new directions for research, thereby advancing interdisciplinary research and exchange, and presenting and disseminating the results of research in diverse ways.

In this spirit, the Chinese Rare Books Digital Humanities Empirical Research Platform adopts the Academia Sinica’s Chinese Rare Book catalog data as its object of study, adopting the BIBFRAME 2.0 developed by the U.S. Library of Congress to structure an ontology suited to meet the needs of an academic bibliography in the “Chinese character/language cultural sphere,” and serve as a basis for Linked Open Data conversion. The adopted “Digital Library of Chinese Rare Books” database was originally constructed by Academia Sinica’s Fu Ssu-nien Library from 2007-2009, commissioned by the Chiang Ching-kuo Foundation for International Scholarly Exchange and built in collaboration with the Asian Division of the U.S. Library of Congress, the East Asian Library of Princeton University, and the Harvard-Yenching Library of Harvard University. The database includes 13 series, 1,070 monographs, and 6,352 series volumes, for a total of 7,435 items in the hierarchical catalog data, covering almost 600 years of writing (including block-printed editions and handwritten manuscripts) from the early Ming dynasty to the mid-20th century (1368-1953). The platform has converted the catalog to SPARQL semantic data to explore the possibilities of semantic query use by humanities scholars, and also offers analysis and visualization tools. By enabling empirical data investigation methods and advancing research on Ming-Qing commercial publications and cultural spread, the platform opens up new possibilities in empirical research for digital humanities.

Highlights of the collected Chinese rare books

Fu Ssu-nien Library, Academia Sinica

In 1928, Fu Ssu-nien (1896-1950), dean of the college of liberal arts at Sun Yat-sen University, founded the Academia Sinica Institute of History and Philology (IHP) in Guangzhou, China. In 1949, IHP moved to Taiwan. In 1960, IHP began building a library to house ancient books and documents. To commemorate its completion in 1961, Hu Shih (1891-1962), then President of Academia Sinica, named it the Fu Ssu-nien Library (FSL). Concerning the transport of documents during exile, IHP records state: “Two warships were used to carry researchers, their families, ancient texts, etc. from China to Taiwan. Throughout the entire Academia Sinica’s relocation to Taiwan, only the Institute of History and Philology underwent this.”

FSL has mainly collected works on history, classics, literature, linguistics, archaeology, ancient scripts, and anthropology. It is a treasure trove of rare books in Taiwan, with an excellent selection of thread-bound books – an important humanities research library suiting specialists’ documentary needs.

Of the FSL ancient book resources linked to by our platform, half are in the genre of local gazetteers and half are in the genre of “collections of individual writers” (別集) under “belles-lettres” (集部). From these, it can be seen that, after the late Ming dynasty, publications of “short popular recreational works” gradually increased. Such “short sketch essays” (小品文) focusing on the sophisticated accomplishments and artistry of the literati leisure class represented a large shift in the concept of literature during the late Ming dynasty, such as in: Cuiyuge pingxuan tangreshi xiansheng xiaopin, juan 2.

The image on the left shows a color scan of the thread-bound, block-printed, Ming edition of “Cuiyuge pingxuan tangreshi xiansheng xiaopin, juan 2, compiled by Tang Xian-zu (Ming dynasty)” from the FSL collection. It is printed with a border on all four sides, with 9 lines of 19 characters each per half-page, and no yuwei (魚尾) decorative pattern in the banxin (版心) page center. Printed upon it are the collection name, juan number, and page number. The character engraving style is “Song typeface (宋體)” also known as “Ming typeface” (明朝體) or “craftsman typeface” (匠體), popular in the Ming dynasty. The first line of the first juan reads, “Cuiyuge pingxuan tangreshi xiansheng xiaopin, juan 1.” The heading above the second and third lines reads, “written by Tang Xian-zu, pen name Ruoshi, of Linchuan.” The second line reads, “collected by Jiang Zhi-huai, courtesy name Daoxing, of Renhe.” The third line reads, “commentary by Lu Yun-long, courtesy name Yuhou, of Qiantang.”

Asian Division, Library of Congress

The U.S. Library of Congress is the oldest federal cultural institution in the United States. The Chinese collection of its subsidiary Asian Division began in 1869 during the reign of the Qing Emperor Tongzhi (1861-1875), when the Qing government gifted 10 titles (in 905 volumes). The Library subsequently continued to purchase Chinese books in the Manchu, Han Chinese, Mongolian, and Hui languages. After the 1950s, it continued with Chinese language books from Taiwan and Hong Kong, and established document exchange agreements with many repository institutions. Today, the Library’s Chinese collection has inventoried over 960,000 volumes and is still in the process of organizing and cataloguing more Chinese titles.

One of the most notable features of the Asian Division books linked on this platform are the numerous Ming-Qing dynasty “books annotated and selected by renowned scholars.” These are titled in bibliographies as “…pingzhu…shixuan” 「○○○評註○○詩選」 or “…pingzhu…wenxuan” 「○○○評註○○文選」, for example: Xinke qiantaishi pingzhu liyulin tang shixuan yu, Tanghuiyuan jingxuan pidian tangsong mingxian celun wencui, etc. It can be seen that Ming-Qing publishers often used the names of “renowned scholars” as a marker of excellence, a quality guarantee, and a marketing technique.

The images on the right display Xuanshi, juan 3; Waibian, juan 3; Shiyi, juan 2 by Yang Shen (Ming dynasty) from the U.S. Library of Congress collection. Printed during Ming Emperor Wanli’s reign, it is currently bound in 8 volumes with scanned color images publicly available. Among the collected 8 volumes, aside from the forewords written by Yang Shen and Bu Da-you, the main text is uniformly formatted, with a printed border on all four sides, 9 lines of 20 characters each per half page, and two small lines for annotations. The banxin (版心) is decorated with a single yuwei (魚尾), below which, at the height of the 8th and 9th character, is written the juan number. The page number is written at the height of the 16th character. The typeface is in the style of Ouyang Xun (歐體字). The image is stamped with a blue “Library of Congress” watermark. There are a total of four stamps on the pages: “yangshan/qingtang/cangshu,” “fu/dashu/mu” (read as: dafushumu), “xiao/ting,” and “yangyin/shansi” (read as: yangshansiyin). Taking Xuanshi, juan 3 as an example, the first line of the first juan reads, “Xuanshi juan zhi yi.” The second line reads, “Ming (space) cheng (space) dou (space) sheng (space) an (space) yang (space) shen (space) yin (space) zhu.” The third line reads, “(indent space) Fuzhou shaomei tanqi (space) menren jia jiangqingchengmao qi jiaoding.” The fourth line reads, “(indent space) Baoying shebei zhuyuefan zengzhu (space) xiushuiyiquan bu dayou jiaozi.”
Reference:Wen-jiun Lin, “The Six Dynasties Theory of Yang Shen--as well as the interpretation and reception of Ogyu Sorai” (楊慎「六朝學」研究──兼論江戶時代荻生徂徠的容受). (PhD diss., National Taiwan Normal University, 2020), 161-163.

East Asian Library, Princeton University

The East Asian Library of Princeton University collects East Asian materials, mainly in the Chinese, Japanese, and Korean languages, with the majority of their Chinese rare book collection being from the Ming dynasty. The works linked to this platform can be summarized in the two categories below.

(I) Imperial examination preparatory literature: These books can be organized into 9 types: the Four Books, Five Classics, eight-legged essays, selected books on classical Chinese literature, books on how to write argumentative essays, Hanlin school books, books on general history, leishu, and Hundred Schools of Thoughts books.

(II) Applied practice books: These include books on divination, medicine, local geography, and a few on religion. Such books occupied a significant proportion of sales in the Ming-Qing publishing market, but as everyday reading material for the lower classes, they were not stored in famous collections. Times have changed, and Ming-Qing scholars and historians today must seek this commonplace literature from overseas collections in order to uncover commercial, practical, or everyday aspects of the Ming-Qing printing industry. Thus, our platform’s Linked Open Data meets scholars’ bibliographic needs regarding historical texts and editions, enabling humanities researchers to more easily acquire resources on rare and unique copies of books stored in overseas collections.

The images on the left display Xinkan zhaihuiyuan quanbuzi shilun cebiaoti huihai jieyao, juan 29 edited by Zhai Jin-chun (Ming dynasty) from Princeton University’s East Asian Library collection. This Ming block-printed edition is currently thread-bound in 29 juan, with scanned color images publicly available. In this current edition, aside from the hand-copied foreword and table of contents, the main text is block-printed and uniformly formatted, featuring a printed border on all four sides, division of text into an upper and a lower row, and 14 lines of 28 characters each per half page. The upper row summarizes and discusses each section’s content; the lower row sets down a selection of historical events and essays. The banxin (版心) features two yuwei (魚尾) designs. The yuwei face the same direction and are located at the level of the bottom row. Between the yuwei is printed the juan number. The typeface’s style falls between Yan Zhen-qing and Liu Zong-yuan.

The first line of the first juan reads, “Xinkan zhaihuiyuan quanbuzi shilun cebiaoti huihai jieyao juanzhiyi” (boxes indicate damaged text). The second line reads, “(indent 15 spaces) Changshu (space) kunhu (space) zhaijingchun (space) jiaobu.” The third line reads, “(indent 15 spaces) Qiantang (space) ruoyu (space) chensuyun (space) yuanbian.” The fourth line reads, “(indent 15 spaces) Qiantang (space) lincang (space) moyunlong (space) yuanding.” The fifth line reads: “(indent 15 spaces) Puyang (space) qiwu (space) liaohongzhen (space) xuding.” The sixth line reads, “(indent 15 spaces) Jianyi (space) yunting (space) zhengziming (space) chongzi.”

Harvard-Yenching Library, Harvard University

The Harvard-Yenching Library is the primary repository for East Asian-related materials at the Harvard Library. It houses works in Chinese, Japanese, Korean, Vietnamese, Tibetan, Manchu, Mongolian, as well as Western languages, totaling over 1 million volumes. The resources linked to in our platform offer, in particular, a glimpse into the Ming dynasty’s “rise of the contemporary text (時文)” (vis-à-vis “new texts” (今文) and “old texts” (古文)), as well as a vast collection of “books on general history” (通史類) compilations.

Lang Yin (1487-1566) wrote in Qixiu leigao – shiwen shike tushuqi: “Before Chenghua, there was no publication of contemporary texts. The local magistrate in Hangzhou, Shen Chen, published a volume of Jinghua richao, and obtained huge profits. After publications rose around Fujian province, they were gradually popularized to many provinces’ official curriculums.” This passage explains that publication of non-classic contemporary texts in the Ming dynasty was not considered significant before Emperor Chenghua’s reign (1465-1487). These types of books were also historically less sought out by collectors. However, much of the Harvard-Yenching Library’s collection consists of such privately printed “non-classics” (俗書), and if their date of publication is examined, most fall after Emperor Chenguha’s reign, serving as evidence of the late Ming to early Qing phenomenon of “contemporary text” publication.

Meanwhile, compilations of “books on general history” constituted a major development in historiography. Before Emperor Jiajing’s reign (1521-1567), these Ming dynasty general history compilations focused on “general outlines” (綱目體). After Emperor Jiajing’s reign, “comprehensive general histories” (綜合體通史) began to be compiled. The appearance of the latter originated from the renewed attention paid during the mid-Ming dynasty to the “twenty-one dynastic histories.” A “full history” was too long for one person to manage, thus editors began to compile abridged general histories. At the time, a few different categories of general history books existed: (1) works continuing or emulating Zheng Qiao’s Tongzhi, such as Hongjian lu or Hanshi; (2) chronological histories abbreviating the twenty-one dynasties, such as Shijiushi jielue or Lidai shishu daquan; (3) categorical history books, such as Zuobian or Yuanshi tongshi. Many of these general histories were produced to meet the demands of imperial exam candidates, whose exam essays had to display to examiners their erudite knowledge of the past and present. Among these general history exam prep books, gangjian (綱鑑) was a popular genre that became widespread.

The images on the right display Qin zixi suxiansheng huizuan lichao jiyao zhinan gangjian, juan 20 by Su Jun, Li Ting-ji, and Xiong Chong-yu (Ming dynasty) from the Harvard-Yenching Library collection. This 1612 block-printed edition is currently thread-bound in 20 juan, with scanned color images publicly available. In this current edition, the paiji publication information on the cover page, printed from right to left, reads: Zhongdetang; in large characters Qin zixi suxiansheng/biaoti gangjian jiyao; then, printed between the two large-character lines are a bagua trigrams image and a publication statement, “Gangjian yi shufangjian hunke duoyi qijian gangmu bubei zhiyi buxiang shinai □□□□□ (□ indicates damaged text)/Xialing/Zixi suxiansheng liuyi shanbu gangjian quanbei biaoti zhiyi jingxiang yiwei juye□□yunming/Bentang kaishu jingzi yizi wue sifang junzi maizhe yuzhi bianyan qingren zhongdetang weiji/Wanli renzisui qiuyue (space) gudan (seven spaces) xiongchongyu jinbai.”

The main text features a printed border on all four sides, with 13 lines of 26 characters each per page. The banxin (版心) is decorated with a single yuwei (魚尾), below which is printed the juan number. The outer blank border of each page (版框) measures 22 x 12.9 cm. The upper row of text summarizes and discusses each section’s content; the lower row sets down a selection of historical events, essays, and reference excerpts. The typeface is in the style of Ouyang Xun.

The first line of the first juan reads on the upper row, “Lunce tiyi” and on the lower row, “Juan zixi suxiansheng huizuan lichao jiyao zhinan gangjian shoujuan.” On the lower row, the second line reads, “(indent four spaces) Zixi (two spaces) su (space) jun (two spaces) bian.” The third line reads, “(indent fourteen spaces) Jiuwo (two spaces) litingji (two spaces) zuan.” The fourth line reads, “(indent fourteen spaces) Taishan (two spaces) yexianggao (two spaces) xiao.” The fifth line reads, “(indent fourteen spaces) Chongyu (two spaces) xiongchengzhi (two spaces) zi.”