A Contrastive Study On Semantic Primitives "Si" And "Die" From The Perspective Of The Natural Semantic Metalanguage

The defining of the basic unit of semantics has always been a headache for linguists. In terms of "define", an object of defining, and words or tools used to define the object are needed. The Natural Semantic Metalanguage, established by Anna Wierzbicka, holds that there are universal semantic primitives, an irreducible semantic core, which can be used to define almost all the complex meanings yet the semantic primitives themselves can not be further defined. Up to now, about 60 semantic primitives have been found by Anna Wierzbicka together with Cliff Goddard through a large number of experiments. These semantic primitives are all expressed in English and languages used in the experiments are mostly from Australia, Africa, Pacific area, Southeast Asia, Europe and North America. However, the so-called experimentation is traditional text analysis, natural observation, and field research. Corpus-based approach is barely adopted in the study of the Natural Semantic Metalanguage. In order to find out whether the Natural Semantic Metalanguage is testable in Chinese, Lu Zhi first adopted corpus-based approach to the study of the Natural Semantic Metalanguage. He studied the collocations and semantic prosody of the 60 or so semantic primitives by comparing their distribution in the corpus through contrastive analysis, yet he did not talk about the specific lexical collocations and colligations of those semantic primitives.This thesis is a corpus-based contrastive study on semantic primitives "si" and "die" from the perspective of the Natural Semantic Metalanguage. It borrowed the research methodology developed by Lu Zhi, and adopted computer software AntConc and SPSS 17.0 to analyze the corpus data which were collected from People's Daily and New York Times from January 1st to June 31st,2015. With a combination of quantitative study and qualitative study, this thesis discussed the overall distribution of "si" and "die" in the corpus, as well as the distribution of valid cases, colligations, and semantic prosody.At first, the study selected out the valid cases of "si" and "die" in the corpus by consulting some English and Chinese dictionaries which are of high credibility. The valid cases were saved and their frequency data were calculated and discussed. Then, the study further classified the valid cases and analyzed the lexical collocations and colligations of "si" and "die". Meanwhile, the semantic prosody of "si" and "die" was discussed. The result shows that there is no significant difference between the distribution of overall distribution of "si" and "die", as well as the distributions of the valid cases and semantic prosody, which is identical with the results of Lu Zhi. Perhaps this is because human beings have been through the same evolution process and formed similar cognition of the world including that of death. Yet when it comes to the distribution of colligations, there exists significant difference, which may be owing to the fact that English and Chinese belong to different language families, and have different cultural backgrounds, for example, English is stative and complex while Chinese is dynamic and simplex.From the perspective of the Natural Semantic Metalanguage, this thesis proves that "si" is identical with "die" in their distribution in the corpus and semantic prosody, which provides supporting material for the Natural Semantic Metalanguage.