Dai Nihon komonjo 大日本古文書
Shōsōin monjo mokuroku 正倉院文書目録
Shōsōin Document Search 正倉院宝物検索
Shōsōin komonjo eiin shūsei 正倉院古文書影印集成
Sonoda mokuroku 薗田目録
Shōsōin monjo shūi 正倉院文書拾遺
Dai Nihon komonjo (hennen) 大日本古文書（編年), 25 vols. Tōkyō Daigaku Shiryō Hensanjo 東京大學史料編纂所, eds. Tokyo: Tōkyō Teikoku Daigaku 東京帝國大學, 1901-1940.
This published collection of Nara-period documents is essential to any and all Shōsōin research. It includes movable type editions of almost all of the documents from the Shōsōin collection, with the exception of a few missed pieces known as “uncollected works [mishū (未収)].” The twenty-five volumes of Dai Nihon komonjo are primarily filled with Shōsōin documents, but it also includes other less well-known sources such as manuscripts from the temple Tōshōdaiji 唐招提寺. It is perhaps the most important reference work for Shōsōin studies, as it contains relatively easy to read print editions of the majority of the collection and is readily available at many university libraries (most commonly in a 1968-1970 reprint edition). It is also accessible online in a searchable edition on the webpage of the Tokyo Historiographical Institute as part of “The Nara Period Komonjo full text database.” Dai Nihon komonjo generally provides the location of the original sources in the six-section classification system, so it is possible to quickly cross reference with photographs and other reference works discussed below.
While it would be practically impossible to research Shōsōin documents without this collection, there are several problems with this publication that scholars need to be aware of before engaging in research. For one, the general organization of the collection is difficult to grasp, a problem that arose out of the complicated process of publication, which took almost forty years, beginning in 1901 and finally finishing in 1940. Publication can be divided into three stages: volumes one through six appeared in the first phase, followed by volumes seven through twenty-three, and finally volumes twenty-four and twenty-five. The first six volumes include the documents that the Meiji compilers thought to be the most important, particularly tax and census registers, as well as a range of other documents, including many connected to the Tōdai-ji scriptorium. These are arranged in roughly chronological order. Unfortunately, in the years when these print editions were being published (1901-1904), the original documents were occasionally unavailable and the editors relied on copies made by Meiji historians (see Kosugi and Ōhashi editions). The first six volumes also contain only a fraction of the Shōsōin collection. Soon after the publication of these six volumes, scholars at the Historiographical Office (Shiryō Hensan Gakari 史料編纂掛) began publishing more documents with the aim of including all those not yet published. As in the first six volumes, the compilers aimed to arrange the remaining texts in chronological order. This work was published in volumes seven through twenty-three, which appeared between 1907 and 1939. Since some documents from the first six volumes had mistakes, the compilers also republished new versions of some of these previously published documents. After reaching volume twenty-three, scholars realized that they still had forgotten many documents and decided to publish all of these remaining documents in volumes twenty-four and twenty-five. In this way, the collection is actually composed of three sets of chronologically arranged documents, with each set covering the entire temporal range of the Shōsōin collection. This complicated arrangement represents the first hurdle that Shōsōin scholars must overcome.
But there are other problems as well. What was once a single document often appears scattered throughout the collection in multiple volumes. This is due to the fact that the documents had been cut up and reassembled by the time this project started; the problem was further exacerbated by the intended chronological order of the collection. For example, what was once a single document may now be split up into several parts appearing in volumes 8, 9, and 24; scholars need to consult each of these volumes to read the document in its original order. Another problem is that the editors often gave titles to the individual fragments that do not accurately reflect the original names of the documents. In some cases, fragments that were once a single document now have several names in the published version, none of which correspond to the original title. References to the six-section classification system are occasionally misnumbered as well. Finally, the published versions ignore many of the codicological features of the documents such as where there are breaks in individual sheets of paper. This data is invaluable for trying to piece together the original order of the documents.
Despite these problems, Dai Nihon komonjo is still a foundational source for Shōsōin studies. Many of the above problems can be overcome when Dai Nihon komonjo is used alongside the next source listed, the Shōsōin monjo mokuroku, as well as the photographic reproductions discussed on this page. (Bryan Lowe)
Shōsōin monjo mokuroku 正倉院文書目録, 6 vols [to date]. Tōkyō Daigaku Shiryō Hensanjo 東京大学史料編纂所, eds. Tokyo: Tōkyō Daigaku Shuppankai 東 京大学出版会, 1987-.
These volumes (sevem published to date) serve as the definitive catalogue for Shōsōin documents. Each volume is arranged by scroll and bundle numbers for each of the six collections. Most importantly, the catalog contains data about the original configuration of the text by listing what was previously connected to the right and left sides of any given section (dankan 断簡). A two page entry is provided for each section of a document. The recto appears on the right hand page and the verso on the left. Each heading is arranged as follows: collection, bundle number (only for Zokuzokushū 続々修), scroll number, and section (dankan) number. There are two types of section numbers listed in each entry: a circle signifies that white paper has been inserted between sections; parentheses marks sections that the editors believe to have been originally distinct but are currently configured as a single unit without white paper between them. This information is followed by the page number in the modern print edition of Dai Nihon komonjo. For example, 続々修三ノ四①⑵（十五ノ一８−三）means that the document is in bundle three, scroll four of the Zokuzokushū and that within that scroll it is in section one (marked by white paper) subsection two (no white paper). The parenthetical entry shows that a published edition appears in Dai Nihon komonjo volume fifteen page one (line eight) through page three. This information is followed by the title of the document, its date, a symbol indicating the presence or absence of preceding and succeeding sections, the first and last characters for the entry, and information about any markings on the manuscript, the paper type, and size. A circle ○ appears above the title of the side of the document used first (ichiji monjo 一次文書) and a double circle ◎ appears above the title of the reused document (niji monjo 二次文書). The final part of each entry provides information about the connection (setsuzoku 接続) with other sections by listing what documents originally appeared to the left and the right of the present section (dankan 断簡). The primary weakness of this collection is that only seven volumes have been published to date. As a result, most of the Zokuzokushū remains uncatalogued. The entire project, which is being supervised by the Tokyo Historiographical Institute, is decades away from completion. Some data for uncataloged entries can be found in the Catalogue Cross-listed with the Dai Nihon komonjo Series (Dai Nihon komonjo taishō mokuroku 大日本古文書対照目録), which appears at the end of each volume. This catalogue lists correspondences between Dai Nihon komonjo and Shōsōin documents including some documents not catalogued individually in the main text. Volume four of the Dai Nihon komonjo taishō mokuroku contains data from the previous three volumes as well and is, therefore, the most exhaustive correspondence table published to date. Much of the data from this project is now being made available online via SHOMUS. (Bryan Lowe)
[top]Shōsōin Document Search 正倉院宝物検索.
The Shōsōin Office published images of the entire Shōsōin collection in 2019 online. The database is available in English and Japanese. To use it, you first click “Shōsōin Document Search.” From there, you can select any of the six collections and simply click search to display the entire collection. Or you could search for a specific volume; for example, one could search “第２巻” and click 続修 to find the second volume in the zokushū. In addition to the Shōsōin documents proper, there are also documents from the northern section of the storehouse including Dedicatory Records 献物帳 and documents about airings 曝凉 and other matters about upkeep, as well as the collection of the Tōnan-in monjo 東南院文書 from the middle section. (Bryan Lowe)
Shōsōin komonjo eiin shūsei 正倉院古文書影印集成, 17 vols [to date]. Kunaichō Shōsōin Jimusho 宮内庁正倉院事務所, eds. Tokyo: Yagi Shoten 八木書店, 1988-.
These volumes contain high quality images of Shōsōin documents. They are organized in accord with the six collections. At present, seventeen volumes have been published, including all of the collections other than the Zokuzokushū. Verso and recto appear in separate volumes. Each image also lists the corresponding page number for Dai Nihon komonjo below the photograph. Documents that do not appear in the Dai Nihon komonjo seriers are labeled as uncollected works (mishū 未収). In addition to being more exhaustive than the print edition, these photographs are invaluable because the published versions in Dai Nihon komonjo contain numerous mistakes and do not fully account for the codicological and paleographical data that provide important clues in Shōsōin research. Photographic reproductions should be consulted whenever possible. In addition to the photographs, the explanations (kaisetsu 解説) at the end of each volume are invaluable resources for Shōsōin scholars. Entries are divided into sections that provide measurements, information about the paper, and the lining of the document. Moreover, the explanations often contain careful analysis based on data related to scribal corrections made in the original document and other markings that might be difficult to view in the photograph. The explanations also provide cross references to the catalog numbers used in Shōsōin monjo mokuroku. The biggest weakness of this collection is that it does not include the Zokuzokushū at present. It has also recently been surpassed by the online publication of images through Shōsōin Document Search in terms of the ability to zoom in on images, though the notes remain a vital resource. (Bryan Lowe)
Zokuzokushū, yonjū-nana chitsu 続々修, 47帙. Kunaichō Shōsōin Jimusho宮内庁正倉院事務所, eds. Tokyo : Tōkyō Karā Rabo 東京カラー·ラボ, 1957-63.
Until 2019, images of the Zokuzokushū were previously only available in microfilm or printed copies of photographs (yakitsuke shashin 焼き付け写真). The original microfilms were produced based on photographs taken by the Imperial Household Agency (Kunaichō 宮内庁) between 1954 and 1965. The collection lacks any explanations and the sheets are occasionally numbered improperly. These have recently been surpassed with the publication of Shōsōin Document Search. (Bryan Lowe)
“Sonoda mokuroku” 薗田目録. In Kyūsai to sono ronri 救済とその論理, vol. 4, ed. Nihon Shūkyōshi Kenkyūkai 日本宗教史研究会, 40-48. Kyoto: Hōzōkan, 1974.
This catalogue, compiled by Sonoda Kōyū 薗田香融, lists individual sutra copying projects from the Tenpyō period. It contains most of the non-canon copying sutra transcription projects from this era. Entries are arranged chronologically. Each is given a number, which is occasionally referred to as the Sonoda number in Shōsōin research. The chart also lists the title of the transcribed sutras, the number of scrolls, the starting and ending date for the project, general comments, and citations to the volume and page number from Dai Nihon komonjo. It is an invaluable source for learning what texts were copied in the Nara period. By tracing the documents related to the sutra copying projects, scholars can precisely date the transcription, identify the patron, and study the process of the manuscripts reproductions in detail. This type of research promises to shed light on the cultic, doctrinal, political, and economic aspects of Buddhism in early Japan.
Sonoda’s catalogue is somewhat dated and has some problems. It does not include every sutra copying project and sometimes treats the same project as two separate efforts. But it remains the best starting point for research on sutra copying projects. (Bryan Lowe)
Shōsōin monjo shūi 正倉院文書拾遺. Kokuritsu rekishi minzoku hakubutsukan 国立歴史民俗博物館, eds. Kyoto: Benridō, 1992.
This volume is a collection of documents that were once held in the Shōsōin but left its walls at some point in time for a variety of reasons. The content was gathered in conjunction with an exhibition held at the National Museum of Japanese History. The Shōsōin monjo shūi includes seventy-five color plates, additional enlarged black and white photographs for some sources, commentaries, and transcriptions of the documents. (Bryan Lowe and Christopher Mayo)