This impressive and thought-provoking book is extremely well written and researched (with a few gaps that will be addressed below). It certainly needs to be read by everyone doing research on any aspect of Old Japanese (OJ) philology, literature, writing systems, and language. One of its nontheoretical merits is that it conveniently presents in one place for the first time in the history of Anglophone scholarship many reproductions of important early inscriptions together with translations.
Lurie proposes a number of innovative theories and insights. Some of them undoubtedly will be accepted in the field, although others are likely to trigger lively debate. Since the limited space of this review does not allow me to address all of them, let me discuss just four.
Lurie's proposal for the centrality of logographic writing in the Man'yōshū (chapter 6) is wonderfully argued and is essentially correct. Ditto for his idea that the major development of literacy in Japan took place only in the late seventh century (part 1). However, some more radical aspects of it, such as the proposal that early Yamato kings and nobility could not read, may be called in doubt. While the evidence to the contrary for the fifth and sixth centuries is only circumstantial (more below), the evidence for the early seventh century is quite straightforward. Although Lurie expresses doubts about the seventh-century provenance of the complete text of Jōgū shōtoku hōō teisetsu (p. 385), the lack of any mistakes in kō-otsu distinctions, and even more importantly, the archaic usage of certain ongana phonograms, point to the first half of the seventh century as non post quem. For example, 里 used for ro 2 instead of ri is atypical even for eighth-century texts.
I find it much more difficult to accept Lurie's theory that the difference between the prose texts of the Kojiki and the Nihonshoki is stylistic rather than linguistic, and that there was no dichotomy between writing in Japanese and in Chinese (parts 2 and 3). Any Chinese or Korean native speaker or a sinologist today with a good knowledge of Classical Chinese can read and understand the prose from the Nihonshoki or prose from the Man'yōshū without any apparent difficulty. However, the same is not true for the prose of the Kojiki: it can be understood with difficulty, at best, and sometimes not at all. In addition, the special kundoku characters in the Kojiki, such as 坐 standing for the OJ honorific imas, leave educated Chinese and Korean speakers or sinologists completely flabbergasted. Lurie's argument for the kundoku nature of the Nihonshoki, based on the Japanese glosses in this text (pp. 238–42), is doubtful for two reasons. First, the overwhelming majority of these glosses are not in the man'yōgana, but in katakana, which makes their provenance much later than the text itself. Consequently, they cannot serve as proof that the original text was intended to be read in OJ. Further, looking at glossing traditions from a comparative perspective, there is no reason to conclude that glossing suggests the language in which the text was to be read. The extremely extensive Chinese glosses on Mongɣol-un Niuča Tobča'an (Secret history of Mongols), for example, do not imply that this Middle Mongolian text was supposed to be read in Chinese. In the same vein, the Man'yōshū poetry is written in OJ, irrespective of whether it is in phonographic, mixed logographic-phonographic, or almost completely logographic script—it cannot be read as Chinese poetry written in Chinese. On the other hand, Kaifūsō is an anthology of Chinese poetry written in Chinese, and trying to read it in Japanese would not be feasible.
I also find unconvincing Lurie's argument against the inherently phonographic nature of all writing. This claim is implicit throughout the book, and explicit on page 363. Although it is true that many indigenous scripts start as predominantly logographic, they tend to change into the general direction of a greater phonography as they develop. The pyramid texts in Old Egyptian or the oracle bone texts of the Shang period are predominantly logographic. But neither was a completely communicatively functional script. As the need for the greater communicational functionality developed, both scripts steadily progressed towards greater phonography. As a matter of fact, I do not know of any fully communicatively functional script that does not include considerable phonography, with a single possible exception of the Tangut writing system, which most Tangutologists consider to be predominantly logographic.
Further, I doubt that a comprehensive study of Early Japanese writing can be done without taking into consideration Chinese historical phonology, Old Korean (OK) philology, and OK historical phonology as well. For example, Lurie claims that the OJ phonogram 止 to 2 is a “vernacular phonogram” (i.e., kungana in the usual terminology) “based on the association” with OJ to 2[mu] “to stop” (p. 266). I am unaware of any other kungana formed by the truncation of the second syllable of verbs. But this is not the greatest problem here. Omodaka et al. correctly identify this phonogram as ongana (1967: 897). Although 止 has the phonetic value of tśɨB in Early Middle Chinese, it is *tśəB in Late Han Chinese (LHC) and *təʔ in Old Chinese (OC). Both LHC and OC correspond nicely with the OJ value to 2 [tə] for the phonogram 止. Remarkably the phonogram 止, unlike the phonogram 支, is not found in the OK writing.Footnote 17 This undermines Lurie's ideas that initial literacy was limited to the influence of Korean scribes and that the early kings and nobility of Yamato did not read.
There are a number of minor errors. For example, on page 174 Lurie writes, “excepting Sinitic loanwords, Japanese morphemes are polysyllabic.” This is incorrect: what about nouns like the OJ me 1 “wife,” me 2 “eye,” te “hand,” na “name,” and ti “blood”; focus particles like pa, mo 2, and so 2; interrogative particles like ya and ka; and ichidan verbal roots like mi 1- “to see” and mi 2- “to go around”? On page 201 Lurie mentions twenty-five Silla hyangka poems in two Koryō period works. As a matter of fact, there are only fourteen Silla-period hyangka recorded in the Samguk yusa, eleven Koryō period hyangka in the eleventh-century Kyunye-cen (both being Koryō period sources), plus one more hyangka found in the Koryō sa compiled during the Yi dynasty. When Lurie discusses Inner Asian scripts on pages 344–45, most of his statements do not reflect recent scholarship. The Jurchen small script, for example, is no longer thought to be based on Kitan small script.
As mentioned above, Lurie is extremely well read, and his bibliography is almost exemplary. But there are two minor problems here as well. Certainly “bibliography” in his or any other book is not an equivalent to “references,” but a reader has a right to expect that the titles listed in the bibliography have been consulted by the author. For example, Lurie focuses too much of his attention on the graphic complexity of the Nihonshoki phonographic script and neglects the question of why the phonography in the Nihonshoki is so different from the rest of the OJ corpus. This issue has been addressed, for example, in Mori (1991), Vovin (2005), and in the most detailed way in Miyake (2003), all of them listed in Lurie's bibliography.Footnote 18 A second problem is that there are some gaps in the bibliography. First, the idea of the standardization in the Kojiki phonographic writing is not original to Lurie (p. 264) but was proposed by Mabuchi Kazuo.Footnote 19 Second, it is not clear why Lurie refers (p. 268) only to partial translations of the Man'yōshū into English, while ignoring three complete English translations: Pierson (1929–63), Honda (1967), and Suga (1991).Footnote 20
These problems aside, I cannot overemphasize that this book is an important contribution to the study of ancient Japan. Although I disagree with the author on a number of points, many of these fall into a grey area of scholarly disagreements. I highly recommend this book to both the established scholars as well as graduate students involved in the study of ancient Japan.
