Developmental Study of English Acquired by Japanese-English Bilingual Children Raised in an Artificial Bilingual Environment

本研究对两名在人工双语环境中长大的日英双语儿童的话语进行了发展分析。首先，对英语动词用法的发展进行分析，发现了以下几点。孩子们通过复制父母的输入开始使用动词。他们开始用未经分析的词块替换一个词。他们开始通过替换主元图式中的主语、宾语或补语来进行创造性的使用。他们很少犯语序错误，大概是因为他们在4岁时说话的创造力不高;其次，分析了儿童在言语中区分日语和英语的能力。他们在2岁以后开始区分，并在2岁时开始生产。在他们开始区分这两种语言之前，同义语被使用了几个月。自我纠正行为在他们表现出这种能力后开始。滥用和混合减少3岁左右; 0。0岁以后，他们开始大量使用同义句。这是因为他们从事项目学习和存储未分析的块。当不能用英语表达日语时，他们会采取以下策略：使用语义相似的替代表达或回避;句子构建失败，或表达出受日语影响的英语句子。这是由于项目学习所储存的英语库存不足。上述发现与基于使用的模型中的论点一致。

Developmental analyses were conducted on utterances produced by two Japanese-English bilingual children raised in an artificial bilingual environment. First, the analyses of the development of English verb usage revealed the following findings. The children started to use verbs by reproducing parental input. They proceeded to replace a word in unanalyzed chunks. They began to be creative by replacing the subject, object or complement in the slot of pivot schemas. They made very few word-order mistakes presumably because they were not highly creative in their utterances by age 4;0. Second, analyses were conducted on the children's ability to distinguish Japanese and English in their utterances. They started to make distinction after age 2;0 and became productive by age 2;6. Synonymous utterances were used for a few months before they start distinguishing the two languages. Self-correction behavior started after they displayed the ability. Misuse and mixing decreased around age 3;0. After age 3;0,they started to use synonymous sentences in large quantity. This is because they engaged in item-learning and stored unanalyzed chunks. When unable to produce the English equivalent of Japanese utterances, they resorted to such strategies as, using semantically similar alternative expressions or avoidance; or they failed in sentence construction, or produced English sentences revealing the influence of Japanese. This will be due to the shortage of English inventory stored by item-learning. The findings above agree with contentions made in the usage-based model.