Learnability, negative evidence and the L2 acquisition of the English passive

Abstract

An issue currently being debated in second language acquisition research is whether negative evidence (i.e., information to the learner that his or her utterance is ungrammatical) plays a positive role in the acquisition of the L2. Some researchers, such as White (1991a; 1991b) and Carroll and Swain (1993), have argued that negative evidence has positive effects while others (see, for example, Schwartz and Gubala-Ryzak, 1992) are sceptical about such effects. In this article, we report the results of a small-scale study that investigated the effects of formal instruction on the acquisition of the English passive by native speakers of Japanese. Japanese has both the ‘direct’ and ‘indirect passive’, whereas English only has the direct passive. The ‘indirect passive’ is possible in Japanese because of the dual status of the passive morpheme rare, which can function not only as a non-thematic (auxiliary) verbal element but also as a lexical/thematic verb. A learnability problem posed by the differences between Japanese and English is that Japanese ESL learners may initially treat the passive auxiliary be in English as being similar to rare in Japanese and thus assume that English, like Japanese, allows not only the ‘direct passive’ but also the ‘indirect passive’. Negative evidence will, therefore, be necessary in order to enable them to arrive at the correct L2 grammar. A group of Japanese ESL learners, who were pretested on the English passive, were placed in an experimental group and a control group. The experimental group was explicitly instructed on the impossibility of the indirect passive in English, whereas the control group was not. Following instruction, both groups were post-tested. The results indicated that the experimental group improved dramatically, whereas the control group did not.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 66 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.