Ethnonyms in “On a Chinese screen” by W.S. Maugham: a lexical and corpus-based analysis

2024

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.31489/2024ph2/128-137
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

In the article the topic about the recreation of ethno-cultural information based on linguistic and corpus analysis of the work of the famous classical British writer S. Maugham, consisting of a collection of short stories “on a Chinese screen” was considered. Subject of analysis is in science, is established that ethno-cultural information is directly related to the knowledge, experience, beliefs, traditions of a particular country. Classical British literature, including the book of S. Maugham, is distinguished by a unique, authentic, artistic, realistic description at the interconnection of different cultures. It was found out that S. Maugham's work, which was written during his travel to China, reflected on the cultural, spiritual values, traditions of the Chinese country, for which he felt with his heart, covered issues that have not lost their relevance to this day, such as inequality in society, cultural diversity, modernization at the cross-cultural relation and the influence between East and West, etc. The result of the article is related to S. Maugham's work, ethnonyms were collected using the esearching way, the frequency of their activity in literary books was determined through the Google Books NGram Viewer, search engine from 1880 to 2019, the frequency and lexical meaning were distinguished by a lexical and corpus-based analysis.