Realms of Literacy: Early Japan and the History of Writing by David B. Lurie (review)

Realms of Literacy, by David Lurie, is an extraordinarily important book. It is a major contribution to the understanding of early Japanese language, literature, religion, history, and archaeology and largely supersedes earlier fine work by Roy Andrew Miller and others for the period and topics it investigates. Its treatment of the development of the Japanese language and its interaction with Chinese and Korean is superb.

A fundamental distinction is made in the book between writing that is representative of sound and writing that is representative of meaning: the “glottographic” (phonographic representation of language) and the “semasiographic” (logographic repre sentation of meaning); the vast majority of pre-Heian texts, for example, are logographic. Lurie expands on this distinction to demolish received views of the linguistic opposition between Chinese and Japanese in early Japanese writing.

The adaptation of Chinese characters to write the Japanese language is often described in phonographic terms: characters that represented Chinese words were associated with Japanese syllables that had pronunciations similar to the original words. This kind of adaptation did play a major role, but a different approach to writing was more important. Char acters that represented Chinese words were associated with Japanese words that had similar meanings, and entire texts written in accord with Chinese vocabulary and syntax were vocally rearranged and read off as Japanese texts, in a process traditionally known as kundoku (liter ally, “reading by gloss”). This process, which combines reading and translation into a single integrated act, could be used to produce new texts as well as to comprehend existing ones. Rather than phonographic transcription, it was this method of reading / writing that dominated all modes of literacy in early Japan, from at least the mid-seventh century on. This means that we cannot describe texts arranged in accordance with Chinese vocabulary and syntax as being written ‘in Chinese’ (no matter what their origins), a conclusion that has profound implications for Japanese cultural history, which has been framed by a linguistic op position between Chinese and Japanese.

(p. 5)

Given how widespread the kundoku reading of texts was—and kundoku “can almost certainly be traced back to techniques used in the Korean states as early as the sixth century”—“it is impossible to distinguish between Chinese and Japanese writing in early Japan” (p. 10). Moreover, argues Lurie, “Phonography is certainly important, but it is logography, in both Chinese and non-Chinese styles, and mediated throughout by kundoku, that provides the key to understanding the history of Japanese inscription, even up until the present day” (p. 183).

Lurie makes four major points about kundoku. First, kundoku is “interlingual.” As he explains, “because even texts that originated in China could be read as Japanese, traditional reading practices did not necessarily in volve awareness of texts as written in one language or the other.” Second, kundoku is “reversible.” In other words, kundoku “is a method of writing as well as of reading. It was used to produce Japanese-language logographic texts (or at least, logographic texts that could potentially be read in Japanese) as well as to read / translate texts with non-Japanese origins” (p. 180). Third, it is “productive,” as manifested by the following: the enormous amount of conventional kanbun written well into the modern period; the generation of “a number of styles of logographic or principally logographic inscription that departed in varying degrees from literary Chinese order and usage” (p. 181), that is, hentai kanbun; and the transposition into Japanese of numerous locutions and vocabulary items (“The incor poration of Sino-Japanese readings was a major vehicle for the adoption of the Chinese loanwords that by the eighth century had already begun to reshape the Japanese lexicon” [p. 183]). And fourth, kundoku can be “invisible.” Absent diacritics or hentai kanbun locutions (either of which makes kundoku practice explicit), any logo-graphic text can be read aloud with Japanese-language readings, regardless of where in...