Word-Class Labeling in the New Chinese–Vietnamese Dictionary: A Data-Based Approach

Abstract

Word-class labeling, which significantly affects the quality and practical value of dictionaries, has remained a persistent challenge in Chinese–foreign language lexicography. Grounded in quantum thinking and transdisciplinary methodology, this study applies the Two-Level Lexical Categorization Theory to systematically examine the word-class labeling in the New Chinese–Vietnamese Dictionary. Drawing on a self-built database and corpus-based usage survey, the study evaluates the current labeling practices, explores the underlying reasons for the problems, and proposes strategies for improvement. The study reveals that while the typical class membership of the entries is accurately labeled, there are still notable defects: (1) mistakenly recognizing non-word morphemes as lexical entries; (2) artificially minimizing the number of heterosemous entries and failing to capture the true usage of Modern Chinese entries contained in the dictionary. These problems may partially stem from certain biased conceptions of Modern Chinese word classes, and the misapplication of the "Principle of Parsimony/Simplicity" under the constraints of scientism/Newtonian thinking. To address these issues, this paper argues that the New Chinese–Vietnamese Dictionary and other Chinese–foreign language dictionaries should abandon the "Principle of Parsimony/Simplicity" in word-class labeling, adopt corpus-based evidence to reflect actual Modern Chinese usage, and thereby enhance dictionary reliability and usability for learners and translators. Keywords: Chinese–foreign language dictionary, New Chinese–Vietnamese Dictionary, word-class labeling, heterosemous entries, corpus-based, quantum thinking