A distinctive collexeme analysis of near-synonym constructions “ying-dang/ying-gai + verb”

A distinctive collexeme analysis of near-synonym constructions “ying- dang/ying-gai + verb” Zhuo Zhang Meichun Liu Dingxuan Zhou Department of Linguistics and Translation City University of Hong Kong 83 Tat Chee Avenue, Kowloon, Hong Kong, China Department of Linguistics and Translation, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong, China School of Mathematics and Statistics, University of Sydney, Sydney NSW 2006, Australia jessyzh@pku.edu.cn meichliu@cityu.edu.hk dingxuan.zhou@ sydney.edu.au Abstract This paper aims to differentiate the two modal auxiliary near-synonyms ying-dang ‘should; ought to; shall’ and ying-gai ‘should; ought to’ through investigating their distinctive collexemes of the near-synonym constructions of “ying-dang/ ying-gai + verb”. We employed the distinctive collexeme analysis (Gries and Stefanowitsch, 2004) to find the distinctive verbal collexemes and then categorized them into different semantic types based on the notion of frames (Fillmore, 1982). The sample was extracted from a news corpus of 2.95 billion Chinese characters. The results found that ying- dang and ying-gai display critical differences in attracted collexemes regarding modality types, usage patterns, and semantic types of verbs. Specifically, ying-dang prefers to exhibit a robust deontic meaning, whereas ying-gai can display both deontic and epistemic senses. Ying-dang also tends to take two-character verbs and appears on formal occasions, whereas ying- gai likes single-character verbs better and shows up in various informal contexts. In terms of semantic frames, ying-dang is inclined to take verbs with purposive efforts, whereas ying-gai attracts verbs of self-motion and emotion. 1 Introduction Modal auxiliary verbs and adverbs are crucial in understanding the intention of a speaker. The Chinese modals pair ying-dang ‘shall; ought to; should’ and ying-gai ‘should; shall’ is considered as should-modals (similar to should) with overlaps in semantics and functions. Despite native speakers themselves may not be able to explain why they choose one word over another, their choices between near-synonym modals are found to display a distinctive trend in distribution (Hilpert and Flach, 2021). The distribution hypothesis that the semantics of a word is reflected by the company it keeps (Firth, 1957; Harris, 1970: 785; Turney and Pantel, 2010) may offer a potential theoretical foundation for this phenomenon. The collocational preferences can be distinguished between near- synonym constructions (Gries and Stefanowitsch, 2004; Hilpert, 2008) as illustrated by usage-based Construction Grammar (Goldberg, 1995; Diessel, 2019). Also, the divergence of collocational profiles associated with different words could offer vital clues to their differentiated semantics (Hilpert and Flach, 2021). The subtle usage and semantic differences of modals can also be found through the pair-wise comparison of near-synonym constructions. Hilpert and Flach (2021) found that English modals could be differentiated by their collocational profiles, and near-synonymous pairs like may and might or must and have to differ in distribution. Urunbaevna (2022) employed Distinctive Collexeme Analysis (DCA) to analyze the strong lexemes of construction “should + verb” and “have to + verb” in 3000 sentences of each structure from the British National Corpus. The study found that have to is inclined to verbs with dynamic senses, whereas should prefers stative verbs that tend …

Content truncated to 500 words