1. Introduction

Discussed the impact of translation as a form of language contact on language change, pointed out the differences between translation and general language contact, and emphasized the lack of research on the impact of translation on the development of the target language. The article mentions different perspectives on the impact of translation research and points out the current research status of translation and language contact in China. This study will explore the impact of translated Chinese on original Chinese by analyzing four types of Chinese texts before and after 1930, in order to distinguish the different effects of translation and general language contact on original Chinese.

2. New progress in translation and language contact research

Translation, as an indirect form of language contact, has an impact on the target language. The House team proposed the concept of "implicit translation", and Kranich et al. proposed 10 research hypotheses to expand the framework of language contact theory. Malamatidou introduced the code duplication framework into translation research, and Pang U0026 Wang constructed a diachronic composite corpus, discovering the transfer of translated text features to the target language. Van Rooy u0026 Kotze compared South African and English translations and found that translation is influenced by social factors. Research often regards translation as one of the language contact domains and explores its similarities. Lansty á k u0026 Heltai distinguishes between translation and language contact, believing that translation has explicit features. The transparency of the English source language affects the original text of the target language. Translation is different from general language contact and has a neutral characteristic. Kolehmainen et al. investigated interlingual erosion, while Redelinghuys used a diachronic corpus to study the characteristics of translated texts. Due to insufficient attention in domestic research, this article adopts a multidimensional analysis method to analyze the stylistic features and interrelationships of translated Chinese, language contact, and original Chinese, focusing on changes in stylistic features.

3. Corpus and research methods

Introduced the corpus and research methods used in the study. The corpus includes translated works, works of bilingual writers, works of monolingual writers, and reference corpora, covering novels and essays from the decade before and after 1930. The bilingual author works corpus includes original works by 10 writers such as Ba Jin and Bing Xin, while the monolingual author works corpus includes works by 10 writers such as Shen Congwen and Zhang Henshui. The corpus sample should be within 20000 words, and the same sample should not be sampled more than three times. Use HanLP part of speech tagging set 1 for Chinese language corpus for part of speech tagging. The study used factor analysis to extract factors from multiple language indicator data and explained these dimensions using Promax oblique rotation method. Use the psych package in R language for data processing, and use Kruskal Wallis rank sum test and Dunn test to compare the dimensional features of four types of texts in two or more ways. Using Biber's 67 language feature indicators as a framework, 53 types of Chinese language features were extracted, of which 40 correspond to English and the remaining 13 are newly added based on Chinese characteristics. The stylistic features of English can be divided into 7 main dimensions. This study extracted 6 main dimensions and used feature values with factor loadings greater than 0.30. The first four dimensions can be summarized as information elaboration and simplification, interactivity and informativeness, static narration and instant information processing, and dynamic narration. This study supports Biber's viewpoint that any language contains interactive and informative outputs as well as narrative features.

4. Dimensional feature analysis of four types of texts

This chapter explores the impact of translated Chinese and language exposure on original Chinese through multidimensional analysis of the corpus. Research has found that there is no significant difference in language features between monolingual and bilingual author texts across six dimensions, but there is a significant difference between them and other text types. Dimension 1 analysis shows that the translated text is closer to the monolingual writer's text in terms of detailed information characteristics, and differs significantly from the late Qing and early Republican texts, indicating that the translated text has explicit features and may be influenced by translation commonalities. Dimension 2 reveals that translated texts have high interactivity and are similar to monolingual author texts, while early texts have obvious informative features. In dimension 3, the static narrative features of the translated text are relatively low, similar to MA, indicating normativity. In dimension 4, the translated text is similar to the bilingual author's text, with early texts mainly featuring dynamic narration. The data distribution of translated texts in multiple dimensions is lower than that of original texts, indicating that translation has a significant impact on original Chinese, especially in terms of language features and development trends.

5. Discussion

A comprehensive analysis was conducted on Chinese translated texts, single bilingual author texts, and late Qing and early Republic of China texts before and after 1930. Six main dimensions were extracted and four of them were explored in depth. Research has found that translated text is a "restricted language" that differs significantly from original Chinese in multiple dimensions, such as higher information elaboration and interactivity. The translated text exhibits a leading role in the English to Chinese translation direction, undergoing changes before the original text. The translated text is more standardized in some features, which may affect the text output of monolingual writers. The translated text and monolingual writer's text are similar in multiple language features, and even without English exposure factors, monolingual writer's text will be influenced by the translated text. The data distribution of translated texts and bilingual author texts in multiple dimensions is more concentrated than that of monolingual author texts, indicating that both involve processing and limitations of both languages. The translated text is more likely to have an impact on the target language in terms of communicative discourse features, while it is less evident in terms of narrative features. At specific historical stages, language changes triggered by translation may be more diverse than general language exposure changes. The changes in social and historical events, literary trends, and narrative modes after the May Fourth Movement have had a profound impact on the development of modern Chinese. When monolingual writers come into contact with translated works, they are more likely to directly absorb and imitate their style expression, while bilingual writers have a higher activation threshold when expressing themselves in a bilingual environment. The restricted features of translated text affect the increase and decrease of the target language, and their explicit or implicit features may affect the output of language contact contexts involved in translation.

6. Conclusion

This chapter summarizes the comparative analysis results of three types of Chinese texts before and after 1930, and finds that translated Chinese has "restricted" features, such as information elaboration, interactivity, and instant information processing. Translated texts differ from original texts in terms of leading and normative aspects, and are closer in language features to monolingual author texts, which may have a greater impact on the development of the target language than language contact. The "restricted" characteristics of translated texts may affect the development of the target language, while factors such as social and historical events, literary narrative patterns, cognitive salience, and text standardization can affect the magnitude of this change. Research suggests that translation, unlike general language contact, has a greater impact on the communicative discourse characteristics of the target language and may trigger more diverse language changes during specific historical periods.