Research Note
Between Paper and Wood, or the Manchu Invention of the Dang’an

Abstract

The modern Chinese word for archive (dang'an) appears in no sources prior to the Qing. This research note briefly introduces the origins of dang'an by tracing its etymology back to pre-conquest Manchu sources. It briefly describes the paper shortages in Liaodong and suggests that this forced the Qing to use other materials, leading to the evolution of a distinct Manchu archival tradition.

研究札记

介于纸木之间，抑或“档案”的满族渊源

Devin Fitzgerald 冯坦风
哈佛大学

现代汉语中的“档案”一词，清代以前的文献中并无记载。本篇研究札记经由追溯满族征服中国以前的文献，简略介绍“档案”一词的历史渊源。本文将简略概括辽东地区纸张短缺的情形，并提出这一情形迫使清廷改用其他书写材料，由此催生了满族独特的档案传统。

研究札記

介於紙木之間，抑或“檔案”的滿族淵源

Devin Fitzgerald 馮坦風
哈佛大學

現代漢語中的“檔案”一詞，清代以前的文獻中並無記載。本篇研究札記經由追溯滿族征服中國以前的文獻，簡略介紹“檔案”一詞的歷史淵源。本文將簡略概括遼東地區紙張短缺的情形，並提出這一情形迫使清廷改用其他書寫材料，由此催生了滿族獨特的檔案傳統。

In the late sixteenth and early seventeenth centuries, paper was rare in the northeastern frontiers of the Ming empire. Hoošan, as it was known in Manchu, first appears in Manchu records as a multifunctional material. Paper money was used as funerary offerings; windows were covered with paper; paper partially paid official salaries; and paper was used to wrap goods.[1] Despite these uses and the local production of cotton, hemp, wood, and other high cellulose materials that could have been used to make paper, paper appears to have never been made in any quantity in the region.[2] Instead, paper arrived in Liaodong from China and Korea and was sought after openly enough to appear in a list of gifts from Korean ambassadors.[3]

Under Nurhaci’s rule, references to paper were infrequent. It was not until early in 1621, several years after the formal announcement of the Latter Jin and shortly after the gift of paper from Korea, that the state began to manage paper looted in military campaigns. In response to a letter from Aita, Nurhaci issued a proclamation about the management of looted goods, noting that fodder in the northeast had been depleted and hence should be “supplemented from those places where there are no Jurchen.” Continuing this train of thought, he commanded that there needed to be an end to the “taking of paper, brushes, salt,” and other goods by private individuals. Henceforth, they would be exchanged for silver from the Khan’s treasury.[4]

Under Hong Taiji, obtaining paper became even more important for the regime. Hong Taiji needed paper to expand government record-keeping capacity. Since trade relations with the Ming were nonexistent, he relied on paper obtained from raiding and conquests, and on paper traded from Korea. Before the Manchu invasion of Korea in 1626 paper, as noted above, occasionally arrived from Korea. After the invasion, Korean paper became a major part of Chosŏn tribute. Before 1626 approximately one half (roughly nine files worth) of the Old Manchu records were written on Korean paper, with the rest being written on Ming paper (sometimes on discarded Ming memorials!). After 1626 the old Manchu records were written exclusively on Korean paper.[5] Fragmentary data from tribute give us an idea of the scale of paper trade. In 1639 1,000 bundles of “good large paper” and 1,500 bundles of “good small paper” were delivered to the Qing.[6] Qing hunger for Korean paper was so great that, following the death of Hong Taiji in 1643, the Manchu state burned enough paper offerings to create a domestic shortage within Korea.[7]

Owing to the scarcity of paper, the Manchu state had to be creative about documentary practices. It was easy to justify the expense of diplomatic letters or major official documents written on paper or silk, but for everyday records, more was needed. To solve this problem, the Manchus turned to an abundant local material: wood. The earliest dang’an in the Qing went by the Manchu term dangse. The word dangse comes from the Chinese word for “a plank of wood” (dang 檔), and became a calque meaning “wood on which a document is written.” The relationship between dang’an and the word was noted by a Chinese exile to the northeast in the early Qing. In describing communication in the paper-starved northeast, he noted that “outside of the borders they often write on wood. When it is being transmitted back and forth, they call it a pai’zi . . . . When they’ve been saved for years, they called them dang’an, or dang’zi.”[8] The Manchu concept of the dang’an – only ever referred to as dangse in Manchu – grew out of the practice of organizing wooden slips.

Examination of the word dangse in pre-conquest materials gives a brief sense of the Manchu roots of the archives. Dangse were:

1. Official communications sent between banner officials (single documents or multiple).
2. The same documents compiled into books or registers.
3. The edited versions of the documents compiled and extracted into chronicles.[9]

The wooden dangse from the Qing are mostly gone. But the legacy of the wooden slips remains. Dangse encapsulated the complete evolution of a document from record in the field to edited court chronicle. They were kept to build the state on a solid documentary basis, and the production of these documents involved nearly every member of the nascent Manchu state.

The involvement of the banners in building the state archive appears early in Manchu sources. The early Manchu chronicles show that dangse were used by the early Manchu state to tie the military (banner) structure of the Latter Jin to the central regime.[10] The earliest two references to dangse in pre-conquest chronicles link them explicitly to the military. In the first, dating from 1619, Nurhaci announced the need to maintain a public record of his accomplishments as Heaven-favored Khan and of the achievements of his people. He commanded banner leaders and other nobles to submit oaths of loyalty if they found him worthy, and these would be recorded in dangse for him to check in the future to see if they lived up to their obligations.[11] The second record of dangse is more quotidian. In 1621 the dangse are mentioned in reference to the place of record for the promotion of Kittangūr to the position of assistant vice-general, as well as a place to record the names of dead Chinese.[12]

Despite the fact that the word dangse only rarely appears in the text, the Manwen yuandang is in fact a dangse compiled from other dangse. The structure of the text gives us a complete picture of how archives built the state during the early Qing. The first point to note is that the Manchu archive was a collaborative venture created by a population that was registered and reorganized entirely by the state. The banner system uprooted local society, and for the first decades banner affiliation existed only on paper (or wood). A volume from the Laodang detailing events from 1626 begins by noting its source: the dangse belonging to the Plain Red Banner.[13]

Another source to illustrate the long reach of early Manchu archives is a surviving volume recording the arrival and flight of refugees to and from the early Qing state. The volume gathers different notices from each of the eight banners. The brevity of these entries leaves little doubt that these were field reports submitted on wooden slips.[14] These field reports would then probably have been submitted to each banner’s record management offices. From there, they would have been submitted to the Wenguan 文館 or to the Neiguo shiyuan 內國史院, where the archive was preserved until its discovery in the early twentieth century.[15]

Once these slips of wood reached the central archives, they were copied down into a register according to their unifying theme.[16] In this case, the archive preserves the arrival of refugees to the Latter Jin in support of the success of Nurhaci’s refugee inducement laws first promulgated in 1619.[17] The dangse, which covers the years 1626-1630, was thus a record of how this campaign proceeded and detailed the reassignment of arriving groups of refugees to service under one of the banners. Consider briefly the examples of document seven and document eight.

Both documents were submitted to the central archive by the red banners. In the corner of the section of the Ming Huidian into which they are transcribed, a small seal notes their provenance. Document seven relates that refugees arrived from the Khalka Mongols on February 24, 1627. The Khalka refugees reported that the Chahar khan Lingdan had begun a campaign against the Khalkha Mongols. Lingdan claimed, “Those who surrendered will be protected, and those who fight shall die.” Rather than surrendering, they fled to arrive eventually at the Jin state.[18] Document eight, dating to March 7, 1627, give a less rosy picture. The document describes the flight of several Mongols from the Manchu Manggūltai Taiji. A few of them were apprehended before getting away, with only one male escaping. They were flogged as a result.[19]

The Ubašame dangse was not the last stop for all of these accounts. At some point after the register was compiled, a scribe went through the records looking for useful evidence to place in the Manwen yuandang. Document seven, which relates some detail about conflicts with the Chahars and the arrival of refugees, has a note in bold letters saying, “Write it.” Document eight has the same bold handwriting. But since accounts of Mongols fleeing the state could damage its reputation, he wrote, “Do not.” Unsurprisingly, when we consult the Manwen laodang, we find the complete text of document seven.[20] Document eight is nowhere to be found.

This type of information processing is not entirely surprising and, in fact, has many parallels with the processing of Ming information into historical narrative. The major difference, however, is that throughout this whole process, all of the documents from the letters in the field, the registers in the banners, the register at the center, to historical compilations were referred to as dangse. Dangse were more than guanwen, because they were documents that soon came to rest in a place of the same name. The introduction of this concept into China after 1644 would influence Chinese ways of thinking about and describing different sorts of official documents.

Concluding Thoughts

After the Manchu invasion, the regime began to translate. In their translation of Chinese materials and texts, the dangse looms as a large point of difference. Between 1644 and the early nineteenth century, dangse transformed into dang’an, a fusion of the Manchu notion of the dang with the Chinese an. Althought the Ming paid significant attention to the role of the archive in local government, the roots of modern Chinese archival thinking are equally indebeted to the Manchu creation of the dangse, which eventually led to the emergence of the term dang’an, a word quite different from all previous terms for “archive.” This diveregence is something that will be explored in greater detail in my dissertation.

Works Cited

• Bi Gong 畢恭. Liaodong zhi 遼東志. Liaohai congshu ed. Dalian: Youwenge Liaohai shushe, 1934.
• Choson Sillok 朝鮮實錄, 2015. http://sillok.history.go.kr/.
• Kanda Nobuo 神田信夫 et al., eds. Manbun rōtō 滿文老檔, Tongki Fuka Sindaha Hergen-I Dangse. 7 vols. Tokyo: Tōyō bunko, 1955.
• Li Deqi 李德啓. Aqige lüe Ming shijian zhi Manwen mupai 阿濟格略明事件之滿文木牌. Beijing: Guoli Beiping gugong bowuyuan wenxianguan, 1935.
• Matsumura Jun 松村潤. Shin Taiso jitsuroku no kenkyū 清太祖實錄の研究. Tokyo: Tōhōku Ajia bunken kenkyūkai, 2001.
• Naoto Kato 加藤直人. Tō jin tō 逃人檔, Ubašame ukame jihe, museingge ukame genehe dangse. Tokyo: Tōhōku Ajia bunken kenkyūkai, 2007.
• Yang Bin 楊賓. Liubian jilüe 柳邊紀略. Beijing: Beijing airushen shuzihua jishu yanjiu zhongxin, 2009. http://server.wenzibase.com/.
• Zhang Cunwu 張存武. Qing Han zongfan maoyi, 1637-1894 清韓宗藩貿易. Taipei: Zhongyang yanjiuyuan jindaishi yanjiusuo, 1978.

Notes

1. Kanda Nobuo 神田信夫et al., eds., Manbun rōtō 滿文老檔, Tongki Fuka Sindaha Hergen-I Dangse) (Tokyo: Tōyō bunko, 1955), 1.250, 297, 357, 472. Subsequently cited as MBRT.

2. Bi Gong 畢恭, Liaodong zhi 遼東志, Liaohai congshu ed. (Dalian: Youwenge Liaohai shushe, 1934), 1.21a–25a.

3. MBRT 1.394

4. MBRT 1.449.

5. Matsumura Jun 松村潤, Shin Taiso jitsuroku no kenkyū 清太祖實錄の研究 (Tokyo: Tōhōku Ajia bunken kenkyūkai, 2001), 28–29.

6. Zhang Cunwu 張存武, Qing Han zongfan maoyi, 1637-1894 清韓宗藩貿易 (Taipei: Zhongyang yanjiuyuan jindaishi yanjiusuo, 1978), 27.

7. Chosŏn sillok 朝鮮實錄 (sillok.history.go.kr ed., 2015), Injo 21.9.2:戶曹啓曰: “瀋陽之喪, 賻儀諸物中, 最可悶者紙地也。 前後所求之數, 將至七八萬束, 我國紙地所産, 只在三南, 不得已分定於下三道, 使之措備上送, 而此後未准之數, 宜搜括京市, 給價貿送。” 從之。 Yang Bin 楊賓, Liubian jilüe 柳邊紀略 (server.wenzibase.com ed., Beijing: Beijing Airushen shuzihua jishu yanjiu zhongxin, 2009), 3.16b: 邊外文字多書於木往來傳遞者曰牌子以削木片若牌故也存貯年久者曰檔案曰檔子以積累多貫皮條挂壁若檔故也然今文字之書於紙者亦呼為牌子檔子猶之中土文字漢以前載在竹簡故曰簡以韋編貫故曰編今之人旣書於紙為卷為部而猶呼之為編為簡也.

8. Yang Bin, Liubian jilüe, 3.16b: 邊外文字多書於木往來傳遞者曰牌子以削木片若牌故也存貯年久者曰檔案曰檔子以積累多貫皮條挂壁若檔故也然今文字之書於紙者亦呼為牌子檔子猶之中土文字漢以前載在竹簡故曰簡以韋編貫故曰編今之人旣書於紙為卷為部而猶呼之為編為簡也.

9. These divisions will be discussed below.

10. Naoto Kato 加藤直人, Tō jin tō 逃人檔, Ubašame ukame jihe, museingge ukame genehe dangse (Tokyo: Tōhōku Ajia bunken kenkyūkai, 2007).

11. MBRT 1.161-62. (This passage contains reference to sajin and fafun.)

12. MBRT 1.333.

13. MBRT 3.1017.

14. For examples of these slips, see Li Deqi 李德啓, Aqige lüe Ming shijian zhi Manwen mupai 阿濟格略明事件之滿文木牌 (Beijing: Guoli Beiping gugong bowuyuan wenxianguan, 1935).

15. Naoto, Tō jin tō, ii-iii.

16. The dangse in this case is written in the margins of a copy of the Da Ming huidian.

17. MBRT 1.164-65.

18. Naoto, Tō jin tō, 3-4. The transcription in Naoto differs from the photograph of the document on 2.

19. Naoto, Tō jin tō, 4.

20. MBRT 5.6: ice uyun de, kalka i monggo ci ukanju jifi alame, cahar han cooha tucifi, meni kalka be gemu gamaha, jarut, korcin i baru burulaha, afahakū dahaha niyalma be ujihe, afaha niyalma be waha seme alaha.