During the past several decades, the role of instruction (especially negative evidence) in the second language (L2) acquisition has attracted the ever-lasting interest and attention of researchers. Studies are focused on the following aspects, where considerable controversy exists: 1) Whether learners who receive formal instruction achieve higher levels of L2 proficiency than those who do not. 2) Whether formal instruction works on the accuracy with which learners employ specific items and rules. 3) Whether formal instruction produces the short-term or long-term effects. Almost all the researches focus on the L1 French or German learners of English. There has been very little research aiming at the learners whose mother tongue belongs to a different typology from English. Research on the L1 Japanese learners of English demonstrates the positive role of negative evidence. Up to date, there has been almost no research with L1 Chinese learners of English as subjects.This thesis first attempts to establish the positive effect in the L2 acquisition by exploring the theory of learnability and the model of L2 acquisition. Besides, the negative transfer of the native language (L1) in the developmental process of L2 as they progress to full competence in the target language is examined as well. Then, the author conducts a one-year experiment to examine the positive role of instruction in the acquisition of English passive for the L1 Chinese learners. The experimental group is offered negative evidence while the control group receives none. All the subjects are required to take three tests: the pretest, the first post-test and the second post-test. All the data are analyzed by software SPSS. The results indicate that to a great extent, the topic-comment structure of Chinese exerts a negative influence on the acquisition English passive; the negative evidence can facilitate the acquisition of English passive in the short run as well as in the long run.