日本の著者名典拠データにFRAD適用する際の課題

摘要

Purpose: The purpose of this study was to analyze structures and contents of authority data on Japanese personal and corporate names, which were made in Japan, and to clarify the degree to which such data are consistent with Functional Requirements for Authority Data (FRAD). Methods: Consistency was measured based on three conditions: (1) the data should have mandatory and highly recommended elements prescribed in Mandatory Data Elements for Internationally Shared Resource Authority Records (MLAR), which provides the basic data elements to be included for international sharing of authority data; (2) the data should contain elements adopted by UNIMARC Manual Authorities Format (UNIMARC/A) or MARC21 Format for Authority Data (MARC21/A), which have already been modified in accordance with FRAD; (3) the data should have fields or subfields allowing the incorporation of elements by UNIMARC/A and MARC21/A. If these conditions are satisfied, then the Japanese data may be sufficiently consistent with FRAD. This paper clarifies empirically whether the Japanese authority data fulfill the three conditions or not, by examining carefully authority formats and manuals obtained from six institutions producing such data in Japan. Results: The results show that Japanese name authority data satisfy condition (1) but not conditions (2) and (3). In order to match the Japanese name authority data with FRAD, the data should be augmented by adding some attributes of entries "Person" and "Corporate body", relationships between different entities and types of relationships. Also, fields for keeping these elements should be made available in the data. Furthermore, the examination revealed unique elements in Japan which are not included in UNIMARC/A and MARC21/A.