Corpus-Construction of Spoken English and its Analysis-- For the Improvement of English Language Education for Communicative Purposes

我很自豪地告诉大家，我已经建立了一个由大约11,000,000个单词组成的英语口语语料库，但不幸的是，由于版权问题，这个语料库将不会提供给公众使用。我广泛使用了这个语料库，并在期刊和会议上公布了各种关于英语口语的研究成果。在这项研究的过程中，我明确指出，教科书和词典在对自然口语的描述和分析方面有待改进。长期以来，对英语的分析，特别是在这个国家，一直是以书面材料为基础的(《英语会话》和《英语交际》课程的教科书提供了口语材料，但它们没有被用作语言分析的重要材料来源)，因此，材料本身以及以语法、教科书和词典的形式呈现给学生的那些材料的语言分析结果主要是书面材料。就我国英语教育的现状而言，无论是为了研究目的，还是为了使教材转向英语口语，口语语料库都是不可或缺的。这项持续而广泛的研究让我得出了两个主要发现：(1)英语口语受到一套不同的规则的支配，这些规则在书面英语中是找不到的--接近和线性等。(2)教材中和英日词典中出现的许多规则与从英语口语语料库中获得的材料不太匹配。必须补充的是，一些常用的口语表达没有出现在教科书和词典中。目前的研究人员相信，该语料库将为语篇和会话分析提供非常重要的材料，以及话语标记语。

I am proud to report that I have constructed a spoken English corpus consisting of approximately 11,000,000 words, unfortunately, though, this corpus will not be offered for public use because of copyright problems. I have made extensive use of this corpus and made public various research results on spoken English in journals and at conferences. In the process of this research, I have made clear that textbooks and dictionaries are waiting for improvement in terms of the description and analysis of natural spoken" English, or English uttered spontaneously in natural situations. The analysis of English, especially in this country, has long been based on written materials (textbooks for "English conversation" and "English communication" classes present spoken materials, but they have not been used as important sources of materials for linguistic analysis), and, as a result, the materials themselves as well as the results of linguistic analysis of those materials presented to students in the form of grammars, textbooks and dictionaries consist mainly of written materials. Regarding the present situation surrounding English language education here in this country, spoken English corpus is indispensable for research purposes and for the purpose of making textbooks veered to spoken English. "The continuous and extensive research led me to two main findings', (1) spoken English is governed by different set of rules that are not to be found in written English - proximity and linearity, among others. (2) Quite a few of rules stated in textbooks and presented in English-Japanese dictionaries ＆ not go well together with the materials obtained from the spoken English corpus. It must be added that some frequently used spoken expressions have not found their way into the textbooks and dictionaries.The present researcher is sure that this corpus will present very important materials for the discourse and conversational analysis, as well as discourse markers.