Pieces in a Puzzle: Changing Approaches to the Shōsōin Documents

For two weeks every autumn the Nara National Museum is a madhouse swarming with uniformed high school students, antiquarians, elderly couples in awe of the imperial family, and a few curious foreigners. The event is the annual exhibition of imperial possessions stored for over a thousand years in one of the world's oldest wooden repositories—the Treasure House (Shōsōin ) on the grounds of Tōdaiji in Nara. Consisting of the personal effects of Emperor Shōmu (r. 724-749) and other courtly accoutrements, the renowned and exotic objects were donated at the time of the emperor's death in 756 to the temple, the construction of which had been the major project of his reign. The collection includes bowls and vessels, musical instruments, furniture, medicine, clothes, weapons, mirrors, beads, and innumerable other items from an otherwise remote era. An annual rite since 1948, this exhibition has helped stimulate popular interest in Japan's distant past while at the same time enhancing the aura surrounding a once-sacrosanct reigning family.

Among the priceless rarities are about ten to twelve thousand fragmentary documents, perhaps the most illuminating and, until the recent excavation of wooden tablets with writing on them (mokkan), copious historical materials for eighth-century Japan. Covering an era from 702 through 780, the documents tell historians more about such topics as family, diet, labor, trade, government, religion, and finance than nearly any other written source, and they do so with a vitality and realism that few other records can command. In this article, I shall examine four interrelated questions: 1) what kinds of documents does the Shōsōin cache comprise? 2) how did they come to be preserved? 3) by what process did they come to be published? and 4) how have they changed historians' interpretations of the Nara era and in turn been reviewed and reread in conjunction with information from other sources, including the law codes (ritsuryō), literature such as the Man'yōshū, and archaeological data?

The Contents of the Shōsōin Collection

To comprehend the contents of the Shōsōin documents, the reader should be aware of three interrelated characteristics. First, officials of the eighth century considered many of these papers to be trash (hogo): they were once-used documents recycled when they were no longer necessary or pertinent to government administration. During the eighth century, paper was an invaluable commodity, and officials commonly saved expenses by making use of both sides of a record that today would have gone into a wastebasket or shredder. Second, reflecting this situation, most of these records contain writing on both front and back (they are thus known in Japanese as shihai monjo, "documents written on the reverse side [of something else]"). Third, in the process of recycling, functionaries often needed only part of the paper once used for a long document and thus cut the original piece into strips. The result is that long documents that once had the force of law have come down to us today as fragments. Because they are scraps, historians frequently find a record lacking a beginning, middle, or end. The result is the jigsaw puzzle that we know today as the Shōsōin collection.

Despite these challenges and the enormous quantity of fragments, historians in Japan have laboriously reassembled the puzzle.¹ As a whole, the Shōsōin cache consists of two overarching categories of documents (see figure 1): I) those related to items put in the storehouse during the eighth century, especially immediately after Shōmu's death; and II) those originally preserved in the large government Office of Sutra Transcription (Shakyōsho ), a bureau charged with the copying of sacred Buddhist texts and established as a subsection of the ad hoc administrative entity known as the Office for the Construction of Tōdaiji (Zō Tōdaiji Shi ).² Each of these two categories includes in turn two subgroups of records, most of which have writing on both front (recto) and back (verso). Recto here indicates the document for which the paper was ultimately used...