Paper short abstract:

This paper introduces the Integrated Database of Hanzi Dictionaries in Early Japan, also known as the HDIC project, which is composed of three main dictionaries of the Heian period. We provide a detailed report of the full-text publication of one dictionary: Tenreibanshōmeigi 篆隷万象名義 in the HDIC.

Paper long abstract:

This paper introduces the Integrated Database of Hanzi Dictionaries in Early Japan, also known as the HDIC project, which is composed of three main dictionaries of the Heian period. We provide a detailed report of the full-text publication of one dictionary: Tenreibanshōmeigi 篆隷万象名義 in the HDIC.

HDIC is a Unicode based project, which includes three dictionaries: Tenreibanshōmeigi Shinsenjikyō 新撰字鏡, and Ruijumyōgishō 類聚名義抄. Over 70,000 Chinese characters can be processed after the release of Unicode 3.1. By materializing the full-text database of these dictionaries, head for creating a worldwide platform for the study of the dictionaries in early Japan. These dictionaries are a crucial source on the history of the Japanese language, especially in the fields of variant characters, phonemes, and lexicons.

With the spread of Unicode, chief Chinese dictionaries like Shuowenjiezi 説文解字 and Guangyun 広韻 have been provided open access. Meanwhile, though many old manuscripts are well preserved in Japan, owing to the problems such as variant characters and erratum, the data origination of old manuscripts is falling behind. Authors predominantly study only old manuscripts that are handed down.

On 1 September 2016, the full-text data of Tenreibanshōmeigi was released (http://hdic.jp/), which is the first time that the full-text of a dictionary in early Japan has been made public. One of the authors, Shoju Ikeda who is the leader of research group, began the database of the dictionaries in the 1990s, by engaging in the study of the information processing of Chinese characters, like JIS and Unicode, for instance. Furthermore, we worked on the database of Tenreibanshōmeigi, which is an abridgement of the original Yupian玉篇.

As an introduction of HDIC, this paper presents an explanation of the construction of the Tenreibanshōmeigi database, the unification standards, transliteration principles, the full-text, publication system, and future issues. We hope that the data, which has been made public, would be of practical use to researchers, both in the domains of pre-modern Chinese character dictionary studies, and information processing.

Paper short abstract:

"Web Chamame" is the morphological analysis support software using UniDic. That was made for the purpose of support of a Japanese linguistic study. "Web Chamame" has been developed as a web application, and it can be used easily using web browsers.

Paper long abstract:

UniDic offers the computerized dictionary for Modern Literary Japanese and Classical Japanese language made for a morphological analysis of Japanese text. This dictionary consists of modern Japanese language UniDic and early middle Japanese UniDic.　UniDic is covering the time range of the Japanese language　and the various genres.

We should use a suitable dictionary which appropriate grammar and word were indicated, for morphological analysis using MeCab and UniDic. For example, "UniDic for Modern Japanese" is should use to analyze the sentences written in Meiji period. Therefore, preparation of the complex morphological analysis execution environment and the analytic work to which a dictionary was changed are needed for a Japanese researcher to begin to study using morphological analysis technology. The preparation to make a morphological analysis is so difficult that it's enough to hinder the research of a Japanese linguist. Therefore, in this study, we have developed morphological analysis support software "Web Chamame".

"Web Chamame" is implemented as the web application, and it can be used easily using web browsers. It makes a morphological analysis of the system by MeCab on the server using more than one dictionary installed in the server. The user doesn't have to prepare the morphological analysis execution environment for a local PC to make a morphological analysis. Because all morphological analysis processing is performed by the server. This system is available in spite of OS, if a web browser is an installed computer.

The feature of the "Web Chamame" can mention three points. The first point is that it's possible to make a morphological analysis without installing a software and a dictionary in a local computer. The second point is that the dictionary used for a morphological analysis can be chosen easily. Nine dictionaries　can be chosen as an analysis dictionary. For example, UniDic for Early Middle Japanese UniDic and UniDic for Modern Literary Japanese . The third point is with the analysis help functions for researchers.　Such as comparing the result, which made a morphological analysis with more than one dictionary.

Paper short abstract:

The paper will aim to collect some Yiddish words which seem untranslatable or very difficult to translate, and to investigate the way they are rendered into Japanese. This will be done with the example of the Japanese translation of "The song of the murdered Jewish people" by Y. Katsenelson.

Paper long abstract:

The proposed paper will aim to collect some words which seem untranslatable or very difficult to translate (at least as far as can be judged on the basis of several European-language versions which simply make use of borrowings), and to investigate the way they are rendered into Japanese. The instance chosen for this purpose will be a Yiddish text which vividly shows the creative Japanese approach to words specific to a very distant language, namely such words that reflect the culture, religion and everyday life of the Jewish speakers. The Japanese strategy, in many respects quite a unique one, will be demonstrated with the example of Asukai Masatomo and Hosomi Kazuyuki's 1999 translation of "The song of the murdered Jewish people" (Dos lid fun oysgehargetn yidishn folk) by Yitskhok Katsenelson (1885/86-1944). The solutions chosen by these two translators for the Japanese version contrast quite sharply with the corresponding English, German, French or other European renderings.

Words which do not lend themselves to an easy translation are often "disposed of" in Europe by leaving them as they stand (i.e. they are adopted as loanwords), while in Japanese a vast array of new formations (including calques), descriptive phrases, approximate native equivalents and other devices are used side by side with borrowings, although the former are clearly more frequent than the latter. The following two examples, tendentiously selected, may serve as good illustrations: Yiddish (di) bime די בימה is rendered as German (die) Bima, English Bimah, Italian (la) bimà, Russian бима, Polish bima; and Yiddish (der) shames דער שמשׂ is rendered as German (der) Schammes, French (le) chamach, Italian (lo) shames, Russian шамес, Polish szames (now the reader of the present abstract may take a short break to check on Wikipedia what these words mean). But in the Japanese translation the equivalents are, respectively, sekkyōdan 説教壇 and dōmori 堂守, both of which can be roughly understood even without consulting a dictionary. In the proposed paper, these various devices used in the Japanese translation will be collected, analysed as for their structure and compared with some of their European counterparts.