This book represents the second part of the Descriptive and Comparative Grammar of Western Old Japanese. The first part covered the phonology, writing system, lexicon, and nominal parts of speech. The second part includes chapters on adjectives, verbs, adverbs, conjunctions, particles, and postpositions, as well as comprehensive indexes for both volumes.
As already mentioned in the acknowledgements, several important technical developments happened between the publication of the first part and the completion of the second part. As a result, a number of modifications and expansions were introduced into the second part that need to be mentioned here.
First, with the availability of the Man’yōshū CD-ROM (Kinoshita 2001), a new index to the Man’yōshū (Kinoshita et al. 2003) which is far superior to an older index to the Man’yōshū (Masamune 1974), and Sven Osterkamp’s Man’yōshū Searcher engine, there was no longer a necessity to treat some volumes of the Man’yōshū as major sources, and others as supplementary. In the second part all text of the Man’yōshū is treated as the major source. Still, when citing textual examples, I normally give preference to the phonetically or partially phonetically written volumes of the Man’yōshū: books 5, 15, 17, 18, 19, and 20. The reader will notice that particularly in chapter six, dedicated to the description of the Western Old Japanese verb, the examples from book 15 of the Man’yōshū are especially frequent. This is due to the fact that MYS 15, in spite of numerous phonetic misspellings found there, is probably the most important Western Old Japanese text as far as the grammatical system is concerned: many forms attested therein are not found anywhere else in the Western Old Japanese corpus. Also, I frequently opted to provide more examples from MYS 15 because it is not dominated by one single poet like, for example, books 17–20 are clearly dominated by Opotömö-nö Yakamöti.
Second, before I started to write the second part, I switched over from Macintosh to PC. This gave me access to the Mojikyō map for Mojikyō fonts that was not available for Macintosh four years ago. Consequently, there was no longer the necessity to use substitute characters for the man’yōgana signs that were not easily accessible or altogether lacking in Macintosh. Thus, the second part reproduces Western Old Japanese in its original form, without any substitute characters. If a second edition of this book ever materializes, the appropriate corrections will be made to part one as well.
Third, I have introduced two minor changes into my transcriptional conventions. First, due to the same switch to PC, long vowels in modern Japanese are no longer written with an accent circumflex mark, but with a macron over the vowel (that was extremely difficult to type on the Macintosh). Thus, for example, in the second part the reader will see Man’yōshū and not Man’yôshû as in the first part. Second, the otsu-rui vowel /o₂/ that was spelled as ø in part one is transcribed as ö in part two. This is due to feedback received from some colleagues who suggested that ø looks too alien to Japanologists. Since, while I am using Yale transliteration for Old Japanese, I was and am still reluctant to use its notation o̱ for the otsu-rui vowel /o₂/, I opted for a compromise and borrowed ö from the traditional transliteration of Old Japanese. I hope that my readers will view these changes only as minor inconveniences.
Fourth, over the years that this book was written my understanding and/or analysis of some Western Old Japanese expressions and passages has changed. This resulted, on some (although not numerous) occasions, in discrepancies between chapter four in the first part and the following chapters in the second part. To give an example, I used to understand the expression pyitö kuni ‘land/province of other people’ as consisting of the numeral pyitö ‘one’ and kuni ‘land, province.’ Now I prefer to analyze it as consisting of pyitö ‘person’ and kuni ‘land.’ The overall meaning of the expression does not change, but the analysis does. Rather than preserving the uniformity with the first part and repeating the erroneous analysis, I opted for the discrepancy and the correction of what I believe was a mistake.
Fifth, the reader will undoubtedly notice the paucity of examples from the Norito in the second part as compared to the Senmyō. There are three main reasons for what may seem to be a neglect, partially outlined in 1.1. The Norito is a heterogeneous text, and the sixteen oldest Norito representing Western Old Japanese and not Middle Japanese are comparatively short as compared to the Senmyō. The text of Norito also uses much more semantographic writing as compared to the Senmyō, therefore it is not as valuable as the latter. Finally, there is already a grammar of Norito in English (Bentley 2001), and I was reluctant to duplicate his work. Thus, the examples from Norito were used only if a point in question could not be illustrated by any other Western Old Japanese texts.
Alexander Vovin
Honolulu, October 2007