The analysis of characteristics of expressions in introductory lectures of liberal arts studies in Japanese universities (provided by a pair of male and female lecturers each in their 30s, 40s and 50s years of ages), student's activities centering around note taking in class rooms, and their notes that represent their understanding processes as well as understanding exercises that represent the results of their understanding was completed. As a result, followings became apparent :
1)Each lecture progressed as a structure planned by the lecturer in advance of the class (e. g., exercise setup, definition of terminologies, explanation of examples/cases, and arrangement in the order of compilation) and a spontaneous structure that is created through interactions between the lecturer and the students are concurrently being generated.
2)The spontaneous structure is the portion which is created as the lecturer changes the lecture contents and the method of transmission based on the observation of the students' responses, and which represents the portion the lecturer wishes to transmit most.
3)The characteristics of expression include citing expression, repetitious expression, referencing expression and use of metalanguage.
4)It was also observed through the analysis of the students' notes that there is a tendency among them that the contents immediately before and after these expressions are best noted.
5)Since a lecture is a speech that both those planned and spontaneous structures progress simultaneously, there are portions where important pieces of information are concentrated and portions that combine them. This was confirmed from the analysis of the lecture as well as from the analysis of the students' notes.
6)Based on these features, it is concluded that it is important for a curriculum for the first year foreign students of a university to be constructed in such a way as to help them understand the introductory courses better by teaching them the characteristics of expression that mark the planned structure and the spontaneous structure respectively so that they can attend the class anticipating what comes next in the lecture.
7)Moreover, it was noted that teaching how to write reports describing what they understood, exercising how to speak about the lecture contents using their own notes, and exercising how to take notes using the speech structure diagram preparation method that shows the structure of a lecture in a diagram are useful in refining the skill of understanding and expressing the lecture contents.
These observations suggest that it seems useful to prepare and use a lecture material with an approximate length of 30 minutes that presents the abovementioned two structures showing in various combinations in order to nurture the students' lecture understanding capabilities. The development of such a lecture material shall be our next target.

The Center of Japanese Language (Japanese Center), Waseda University was established for Japanese education and research. The Japanese Center provided the research and teaching team, which includes Japanese linguists and Japanese applied linguists, opportunities to discuss and consult with each other about the ongoing Japanese program and teaching materials designed for the short-term overseas students in the Japanese Center.
The present study aimed to develop an appropriate Japanese program and teaching materials to accommodate the above situation. The secondary aim of the present study is to build up the a new foundation and guidelines for Japanese education within Japan and overseas in the near future.
The procedures of the present study are described as follows:
In 2000, learner needs for the short-term overseas student in the Japanese program were inspected. Along with the results, published teaching materials were sampled searching for appropriate models for the present study. (See Japanese version for an example of data collection method.)
2001, The Graduate School of Japanese Applied Linguistics. Waseda University were established. The researchers in the present study coordinate the practical research seminars in their specialized field in order to gain practical knowledge for authentic Japanese education for overseas students.
The researchers in the present study met regularly and discussion were made about the ongoing teaching materials used in the Japanese program. Teaching materials for 'Reading Comprehension'. 'Oral Expression Practice' and 'Overall Competence Practice' for elementary Japanese were developed in 2001.
The final product of the present study is aimed not only be used at the Japanese Center, but also as teaching materials for the research demonstration for the Japanese education program.
The trial version of the teaching materials from the present study is hoped to be published and practiced within the next 3 year. The final goal of the present study is to accomplish opinions and cooperation from Japanese education organizations and institutes within Japan and overseas for the possibility of developing appropriate Japanese education curriculum.

The present research on the function and structure of the semantic paragraph "dan" in Japanese written and spoken discourse, is based on a theoretical investigation of previous research and results of a survey on summaries of an essay which was heard or read.
(1) The immediate constituents of written discourse and spoken discourse, i.e., the largest linguistic units in Japanese, are "dan" units which have a topic unifying function.
(2) "Dan" in principle are composed of 1) "tyuusinbun" (topic sentence (s) ), which express the topic of topic expressions and predicate expressions, and 2) the supporting sentences which are unified by these topic sentence (s) .
(3) Tere are 6 types of Japanese discourse. These discourse types are classified according to the firequency and position of the thesis sentence (s) .
(4) In Japanese discourse, bikatugata 'main point final' type (where the topic sentence occurs at the end) is the most common.
(5) We conducted a survey of two groups in which the surveyees wrote summaries based on their hearing or reading of a main point final type argumentative essay. Results showed that there was a tendency for the summaries of both groups to reflect the main point final structure of the original text. The summaries provided evidence for the structure and function of the dan in the original text, and there was a tendency for elements of the thesis sentence and topic sentences to be used in the summaries.

The goal of this research is to investigate the relation between patterns of expression and procedures for evaluation of Japanese summaries. We analyzed similarities, differences and relevant factors based on a comparison of the patterns of expression used in the summaries of two bikatugata 'main point final' argumentative essays by non-native foreign students and native Japanese university students and evaluations of these essays by teachers of Japanese as a foreign language (JFL), teachers of Japanese as the national language (Japanese) and Japanese university students. Based on these results, we determined principles for writing summaries of Japanese argumentative essays and drew the follwing conclusions related to the writing and evaluation of summaries written in Japanese.
1. Based on an analysis of the patterns of expression used to combine essential parts of the original essays we showed that while summaries written by native Japanese students tended to be written in the same main point final structure as the original essays, some of the summaries written by foreign students varied from this structure and had were problems related to content and expression.
2. There was a correlation between the patterns of expression used in the summaries and the evaluations made by the JFL teachers, Japanese teachers and Japanese university students. The more similar the patterns of expression used in a summary were to the original essay, the higher the evaluation and vice versa, the less similar, the lower the evaluation.
3. The average of the Japanese teachers'evaluations of summaries written by Japanese university students tended to be slightly higher than the average of the JFL teachers'evaluation of summaries written by non-native foreign students. A similar tendency was observed in Japanese university students'evaluations of the summaries written by these two groups of students.
4. The patterns of expressions used in the summaries related to the evaluation given and the categories used to evaluate the summaries by the JFL and Japanese teachers. In general, summaries were rated higher when they included (1) the conclusion and main points of the original essays, (2) did not have particular examples or opinions of the summary writer of content that was not included in the original essays, and (3) had well-organized connectives, cohesion, and discourse structure. We view these features as principles and techniques for "good" summary writing.
5. The Japanese teachers based their evaluation of summaries written by native Japanese students primarily on conciseness and ability to express the discourse structure. In contrast, JFL teachers' evaluations of summaries written by non-native foreign students focused on correctness of content and words, grammar, characters, etc.
6. The rating that JFL teachers gave for "good" summaries written by non-native foreign students was higher and the rating they gave for "bad" summaries was lower than that given by Japanese teachers and this was statistically significant as indicated by the stndard error of the mean.
Topics for future study include further investigation of the nature of evaluation by increasing the number of evaluators so that more statistically significant patterns can be found. We also plan to use this future study as a basis evaluators so that more statistically significant patterns can be found. We also plan to use this future study as a basis for determining more ideal categories of evaluation.

Present report contains the findings of the study of summary surveys by Korean learners of Japanese language in 1988. This study was done by comparing the data with the data of a similar survey done by Japanese university students. The report mainly concerns typology of the content structure of the summaries, and the error analysis.
1. The criteria for definition of "content identification units" (CIU), used to identify the elements of the original text in respective summaries, were thoroughly reconsidered and reclassified into 10 classes containing 23 different criteria. The order of application of the criteria was also established. Next, the data of Korean and Japanese summary surveys of the 6 original texts were reanalyzed with the help of the newly defined CIUs. Thus "content survival frequencies" for each CIU were obtained and reexamined using two kinds of chi-square method.
2. As the result of 1, three types of CIUs emerged : the "obligatory", the "supplementary" and the "other" CIUs. On the basis of "obligatory" CIUs and their distribution (presence or absence) in the "initial", "middle" and "final" segments of the summaries, a classification of the content structure of the summaries was attempted. The result-including the comparison with the Japanese data was as follows. A majority of the Japanese summaries of the 6 original texts display rather clearly defined patterns of the "obligatory" CIU distribution over the three types of segments. Compared with this, ddata of Korean learners display different patterns, showing that the capacity of Korean learners for understanding the originals has in general not been sufficient.
3. The errors found in Korean learners' summary data were classified and stored in a database, with the help of which the trends in error making were further examined. There seems to be some connection between the type of errors and the type of content structure of summaries according to the classification in 2. The summaries of Korean learners display different characteristics both in expression and understanding of the original texts, compared with the data of Japanese students.

大学学部留学生の日本語の講義理解力を向上させるために、人文学系講義の「談話型」の表現特性と、3種の講義の日本人学生と外国人留学生による受講ノート、要約文、インタビューの｢理解類型｣と「表現類型」を、｢話段｣という観点から｢日本語機能文型(FSP)｣と｢情報伝達単位(CU)｣を用いて分析した。その結果、1)講義の談話には複合文型や話段の多重構造がある、留学生の理解データは2)｢情報伝達単位(CU)｣の残存傾向にばらつきがある、3)具体例や身近な表現が残存し、専門的で抽象的な表現が避けられる、4)講義のテーマや全体的構造の把握が不十分なこと等が、日本人学生との比較により明らかになった。

本研究では、人文学系の講義全12資料における「日本語機能文型」の用例を検索し、2種の講義A、Bの「情報伝達単位(CU)」「節」による文型記述をして表現特性を分析した。また、日本国内の大学・大学院の外国人留学生35名を対象とした講義AのVTR視聴による受講ノートと要約文の理解調査を実施し、日本人大学生28名の調査結果と比較して、留学生の日本語の講義理解における談話の全体的構造の認識不足の問題を明らかにした。「日本語機能文型」739項目を検索した12種の講義で、検索対象外とした114項目(15.4%)には1，2レベルの初級文型が100項目(87.7%)あり、6～8レベルの上級文型はなかった。機能文型の使用頻度は、検索した625項目(84.6%)中に、「異なり文型」の使用が399項目(54.0%)、不使用が226項目(30.6%)であった。上級文型ほど不使用率が高く、検索保留を加えると、初級・中級文型がさらに多用される傾向がある。つまり、日本語の講義の談話に多用される機能文型の習得が、外国人留学生の講義理解力の向上にとり有効だということである。講義A、Bの談話の全文を「情報伝達単位(CU)」に分類し、「文末・節末叙述表現」からなる「節」ごとに、全機能文型を記述した結果、検索保留の初級文型が多用されていた。講義の談話の文型記述により、接続詞や指示語を含む連文、話段、連段等の統括機能の及ぶ範囲や多重構造を客観的に認定し、「談話型」のより精緻な分析が可能になった。受講ノートと要約文の表現の文型記述により、接続・指示・省略・反復・引用・参照・転換表現等のCUを指標に、原話の話段の統括機能を客観的に検証することが可能になる。本研究により留学生の日本語の講義理解力を向上させる教材開発の可能性が得られた。佐久間まゆみ研究代表者(2006)『２００５年度重点研究 研究成果報告書「日本語機能文型」教材開発のための基礎的研究』早稲田大学日本語研究教育センター

本研究は、日本語の講義と講演の全体的構造、すなわち、独話の「談話型」について、受講者の理解調査の結果から解明することを、主な目的としている。講義と講演は、いずれも、単独の表現主体（講義者・講演者）が一まとまりの知的な情報内容を一定時間（60～90分間）内に多数の理解主体（受講者・聴衆）に向けて発話するという、独話としての共通点と、それぞれの伝達目的に応じた固有の表現特性を併せ持っている。 今年度は、講義の談話に関する共同研究の分析結果を再検討し、講演の談話の本格的研究の準備段階として、資料収集と予備調査により、両者の「談話型」の異同を解明するための方法論的可能性と問題点を探ることにした。 大学の講義の談話分析と受講者の日本人大学生を対象とする理解調査（ノート・キーワード・要約文）の結果との関連性を考察し、以下のような日本語の独話の談話型の表現特性と理解の諸相を解明した。特定の受講者との相互作用が積極的な講義は、「対話」性がより強く、不特定の聴衆との双方向の交流の少ない講演は、「独話」性が顕著に認められた。これは、大学生対象の講演と一般向けの講演との違いにも、共通する傾向だといえよう。 １．講義の理解調査の残存認定の尺度として、10種の「情報伝達単位（ＣＵ）」を用いて、原話・ノート・要約文を分析し、受講者の理解に基づく講義の談話型を解明したが、講演も、CUを適用した分析が有効かつ不可欠であることが確認された。 ２．２種の講義の受講者２集団約160名によるノートと要約文の調査結果から、講義の表現特性を分析したが、講演の場合は、聴衆の理解過程の調査が困難なことから、講演の録画資料による理解調査を後日実施する必要があるが、具体的手順の検討を要する。 ３．講義の「話段」の多重構造に基づく「談話型」の基本は、講演にも共通している。

本研究は、大学の講義の談話の表現特性について、受講者の理解を通して解明することを目的とするが、今年度は、準備段階として、日本語の講義の談話資料と大学生のノートや要約文等の資料を収集し、独話資料における各種の言語単位の認定方法を検討した。講義の談話と各種の理解データにおける情報の伝達過程を解明することが不可欠である。　 １．要約文の原文残存認定に有効な「情報単位」10種の分類基準を修正して、講義の談話とノートや要約文を分析し、受講者の理解に基づく講義の談話の表現特性を解明した。 ２．学部学生２集団約160名による講義の理解内容の要点を記したノートとキーワード、要約文を「情報単位」に分類し、その出現傾向から、講義の談話の表現特性を分析した。 ３．上記２の講義２種を含む、大学の人文学系の講義資料計11種の表現特性を比較した結果、「話段」の多重構造に基づく「談話型」を有するということが明らかになった。

本研究は、日本語の文章の読解と談話の聴解による要約文の表現類型の解明を目的とするものであるが、以下の方法を用いて、日本人の大学生と社会人の要約筆記者による要約文の資料を収集し、データの処理と分析を行った結果、次のような成果が得られた。　１．文章の読解による要約文の調査データの中から、日本人大学生計６８名による尾括型の論説文を原文とする要約文の原文残存認定作業の結果を修正して、原文残存率の統計的検定と表現類型の分類を行ったところ、原文中の重要な内容を表す必須成分と、原文の文章構造に基づく要約文の尾括型の表現類型が解明された。 ２．声優の朗読テープを用いた１と同じ原文の聴解による要約文のデータ（１９９７－９年度調査の日本人大学生２集団約計８４名と社会人要約筆記者２集団計１２１名）について、原文の要素の残存認定作業を実施し、統計的検定を施して、各要約文の「必須成分」の残存形態に基づく要約文の表現類型を分析した。特に、大学生の読解要約文と社会人要約筆記者の聴解要約文の表現特性の異同を検討し、両者の共通点とそれぞれに特有な表現特性を解明した。 ３．上記１と２の大学生２集団による読解と聴解による要約文の表現類型を比較した結果、いずれも原文の尾括型の文章構造類型を反映するという共通点が明らかになった。一方、聴解要約文のみに原文の具体的な内容を表す要素の残存が認められた。 ４．聴解要約文の調査に関して、読解要約文とは異なる方法論を開発する必要性が生じた。まず、複数の集団による要約文の原文残存認定作業の結果から、いずれを必須成分とするかについて、新たな判断を要したが、読解要約文を基準として聴解要約文の特性を把握した。また、原文と同様の「情報単位」に各要約文を区分した上で、原文残存認定と表現類型を分類することによって、従来よりも分析の精度が高まり、分類結果の妥当性が検証された。