The Chinese orientation word "down" is a high-frequency word.For native Arabic Chinese learners,"down" is a word with a certain degree of difficulty,which is mainly due to the differences between the two languages of Han and Arabic.At present,the academic research on " under" is not sufficient,so this paper plans to study this.The main work of this paper is two: first,it summarizes the semantic type of the bearing word " down" in Chinese,and starts from the "down" in different meanings,and compares it with the corresponding expression in Arabic.The second chapter of this article mainly completes the work in this aspect.Second,according to the comparative conclusion,the important differences between the Chinese orientation word "down" and the Arabic related expressions are summed up.The difference may be difficult for learning for Chinese learners in the background of native Arabic speakers.The difficulties were sorted out,and the test paper was formulated to investigate the acquisition of Chinese learners in the native Arabic language background,so as to understand the overall acquisition of learners’ orientation words "down".The third chapter of this article mainly completes the work in this aspect.As the basic meaning of the orientation word "down" is the meaning of "spaceorientation".In this sense,the Arabic orientation terms(tahata)and(asfala)correspond to "down".The meaning of "down" in "space-orientation" can be extended to the meaning of "boundary","time-order" and "condition",most of which are the metaphorical meaning of "space-orientation".The Arabic terms(tahata)"and"(asfala)cannot express the meaning of "boundary","time-order" and "condition".Through comparison,the paper predicts the difficulties of Chinese learners in native Arabic language background.Therefore,we can make a test paper to investigate the acquisition of Chinese "down" among Chinese learners in native Arabic language background.Through the statistical analysis of the questionnaire,the correct rate of the questionnaire is sorted from high to low,and the testers are divided into different levels of HSK,mainly divided into three parts,namely elementary,intermediate and advanced,and the correct rate of different semantics is calculated statistics.This paper analyzes the error types and causes of the " bottom" of the subject output,and finds that misgeneration is the main error type,and the most important cause is the influence of negative migration of mother tongue. |