Shimotsumaki

The lower scroll begins in full narrative and ends in a list. Nintoku is given at length, including the tale of his climbing a hill, seeing no cooking smoke over the houses, and remitting taxes for three years while his own palace fell into disrepair. Succession quarrels, a jealous empress, exiles, revenges and a great deal of song fill the following reigns. Then, in the last stretch, the text contracts sharply. A sovereign is named, his consorts are named, his children are counted, his tomb is located, and the scroll moves on.

The contraction is often read as failure, as though the compilers ran out of material or interest. It is better read as the book finishing its job. Once the line has been carried unbroken from the sun goddess to the historical court, the only thing that must not be omitted is the succession itself. Anecdote is optional; the chain is not. The genealogical passages that a modern reader skims are, on the book's own terms, the part that could least afford to be lost.

The end point is worth noticing. The scroll stops with Suiko, whose reign ended in 628, and the work was presented to Empress Genmei in 712. Roughly eighty years of recent and politically awkward history, including the reforms and factional violence of the seventh century, sit outside the frame. The Kojiki closes before it reaches anything its readers would have argued about.

For most of its life the lower scroll, and the Kojiki with it, was the second chronicle. The Nihon Shoki, completed in 720 in proper literary Chinese and organised by reign year, was the court's official history, lectured on formally and cited in ceremony. The Kojiki, in its awkward hybrid script and without dates, circulated far less. Its surviving manuscript tradition is thin, and the earliest complete copy is medieval.

Its elevation came late and came from scholarship. The kokugaku movement of the Edo period turned to the Kojiki precisely because its language was less Chinese, arguing that it preserved something native that the Nihon Shoki had dressed up. Motoori Norinaga spent decades on a commentary that had to reconstruct how the text should be read aloud before it could be interpreted at all, and that labour made the book legible to a wide readership for the first time in a thousand years.

What followed was more than literary. From the nineteenth century the mythology of the first scroll, transmitted through this reading, was put to work in the account of imperial descent taught in schools and enforced by the state, and after 1945 that use was dismantled. The book itself did not change. Shinto had no scripture, and the Kojiki was not written as one; it was made to serve as one, twice over, by readers with reasons of their own. Any account of it that starts from the assumption that it is a sacred text starts a thousand years late.