This presentation is part of a project comparing vernacular glossing in East Asia to glossing traditions in the medieval West (Moran and Whitman under review). The specific comparison here involves the role of glosses in dictionary compilation.
Tsukishima (1959) demonstrated that a major portion of the entries in the Zushoryō-bon Ruiju myōgishō (図書寮本類聚名義抄1081-1100) were copied from specific glosses in kunten 訓点 glossed texts. The Ruiju myōgishō is a Chinese character dictionary, probably compiled by clerics of the Hossō 法相sect. Entries give the Chinese character or word heading, a definition and source, and a Japanese equivalent in man’yōgana or katakana. Tsukishima showed that the source of the entry is predictable from this: entries with man’yōgana glosses were copied from other lexical sources, many Chinese, while entries in katakana are copies of kunten glosses from various texts glossed in Japan.
There are potential parallels here with the practice of compiling glosses into dictionaries in the Latin West. A classical description of this practice is given by Lapidge (1986: 53-4):
… we should bear in mind the various stages in the compilation of a glossary: first, various (perhaps random) interpretations or interpretamenta are copied into a manuscript above or alongside particular difficult words (or: lemmata) secondly, the various lemmata and their accompanying interpretamenta are collected and copied out separately in the order in which they occur in the text (we refer to these glossae collectae); thirdly, the various glossae collectae are sorted roughly into alphabetical order, with all items beginning with the same letter being grouped together (a-order); finally, the entries under each letter are re-sorted into more precise alphabetical order, taking account of the first two letters of each lemma (ab-order). A surviving glossary may (and usually does) include materials or batches of words treated in any of these ways, though it will be obvious that when one is trying to identify the text on which the glosses were based, glossae collectae offer the clearest evidence.
In this three-step process from gloss to glossary, there are formal and functional similarities between glossae collectae and Japanese ongi 音義. The latter are compendia of characters, words, and phrases with an indication of their Chinese or Sino-Japanese pronunciations (usually by means of a hansetsu/fanqie 反切 spelling), one or more definitions, and with increasing density beginning in the late Nara period, a Japanese gloss. Like glossae collectae, ongicollate annotations from specific texts and arrange them in the order that their lemmata appear in the text, and are thus useful for readers without access to glossed versions of the text. Both are the products of monastic scholarship.
However the comparison breaks down on closer examination, at least for early ongi. The Zushoryō-bon Ruiju myōgishō collects some glosses from various ongi, but these are treated on a par with glosses from Chinese and Japanese dictionaries. Nara and early Heian ongi appear to be closely based on original Chinese yīnyì, and their entry selection is highly sporadic. Thus the Ishiyama-dera Daihannya-kyō ongi (大般若経音義; late 8thcentury, attributed to Shingyō) appears to have excerpted lemmata from a non-surviving yīnyìon the Mahāprajñāpāramitā Sūtra complied byXuányīng 玄應. Entries average less than one per scroll of the sutra, and there are a total of only fifteen Japanese glosses in man’yōgana.
Early ongi, then, appear to be motivated by the prestige of cultural transmission rather than the practical purposes fulfilled by glossae collectae in the West. In the reading portion of the presentation we’ll look at excerpts from the Shinyaku Kegon-kyō ongi shiki (新訳華厳経音義私記; late 8thcentury), the Daihannya-kyō ongi and the Zushoryō-bon Ruiju myōgishō and its kunten gloss sources.
Lapidge, Michael. 1986. The school of Theodore and Hadrian. Anglo-Saxon England15, 45-72.
Moran, Pádraic and John Whitman. Under review. Glossing and reading and in western Europe and East Asia: a comparative case study.
Tsukishima, Hiroshi. 1959. Kundoku shijō no Zushoryō-bonRuiju myōgishō[訓読史上の図書寮本類聚名義抄The Zushoryō-bonRuiju myōgishōfrom the standpoint of the history of kundoku]. Kokugogaku37, 35-53.