This study examines the functional verbs of the passive voice. Functional verbs of the passive voice do not seem to have the same function, but considered to have their unique characteristics. Therefore, I try to clarify the difference in characteristics of functional verbs by investigating their descriptions in dictionaries and the corpus. I first investigate the characteristics of functional verbs by conceptual consideration, then I perform quantitative consideration based on the result. In the quantitative consideration, three following viewpoints were useful; distribution of the functional verbs from the appearance field, noun part of the functional verb, and combination of the nominative and the active subject. By the above-mentioned consideration, the difference in characteristics of functional verbs such as the passive type, the characteristic of the noun part, the writing style and the sentiment polarity became clear.

View full abstract

Although semantic roles of the Japanese case particle de have been studied from varied angles, relatively few studies have attempted to incorporate a quantitative approach into examination of its usage. The author, therefore, conducted a quantitative analysis of the Balanced Corpus of Contemporary Written Japanese (BCCWJ) and assigned twenty-four semantic role codes to seven hundred text samples taken from the corpus. Then, the author investigated (1)what kind of features each of the twenty-four semantic roles had, (2)how often each of them occurred in the attested data, (3)how they were classified, and (4)to which extent each of them deviated from the prototypical semantic role. The frequency-based investigation has proven that (1')each semantic role is embodied by a highly limited set of collocating nouns, (2')place-related semantic roles occur most often, and they are followed by measure/ situation-related semantic roles, (3')semantic roles can be classified into a core group (place, measures, reasons, state) and the others, and (4')the distances from the prototype are roughly in accordance with those suggested in the radial category model.

View full abstract

In this paper, we extract grammatical items using N-grams and show that grammatical items or functional expressions can be taught efficiently by extracting expressions using the word N-grams from the corpus. First, we review the literature on corpus and grammar research from the viewpoint of Japanese language education. Then, we examine the collocation expression list of Lee & Sasaki (2015) and construct a functional expression list. For the purpose of verifying the validity of these extracted functional expressions, we have compared them with the frequency data of YNU corpus. The result indicates that the numbers of intermediate-level functional expressions tend to increase as learning proficiency increases. we argue that analyzing quantitatively learner corpora enables us to grasp the whole picture of the learner language.

View full abstract

This manuscript aims to clarify the changes of Koji Uno’s writing styles after he suffered mental illness in 1927. In order to understand the differences in Uno’s works before and after his illness, a quantitative analysis using correspondence analysis and statistical hypothesis testing was employed. By analyzing the textual measurements, we found that his writing style after recovering from the illness is different from the previous. Both the frequency of Chinese characters and the vocabulary richness increased in the works written after the illness. These results, after further analysis, mainly consist of an increase in the use of nouns in later writings. Consequently, the overall writing style changed from spoken to written language. Furthermore, the overuse of the comma has an impact on the loss of the garrulity and fluency, which led to the changes in the style of the writings after the illness. Additionally, the work named ‘Nichiyobi’ which had been published before the disease treatment is more similar with the works written after recovering, indicating that the writing style had been changed even before the illness.

View full abstract

This paper analyses linguistic features of primary school textbooks for the penultimate and final grades (grades five and six) and junior high school textbooks for the first and second grades (grades one and two), in order to reveal factors affecting the so-called "primary-secondary" learning gap. We chose science textbooks and analysed their complexities from the points of view of lexical and syntactic characteristics. The analysis revealed that there is a clear gap in syntactic complexities of sentences between primary and secondary school textbooks. The analysis showed that the factors constitute the gap between primary and secondary school textbooks.

View full abstract

我々は日本語を第二言語とする日本語学習者が大学などで論文・レポートなどアカデミックな文章を作成する作文支援システム構築を目指している．オーセンティックな大規模日本語コーパスを利用して，学習者の作文中の表現と，コーパス中の表現とを比較できる「表現辞典」を構築し，学習者の表現をその目的にふさわしい表現に言い換える提案をするものである． 本稿では様々な言語要素の中から，「接続表現」に焦点を当て，「接続表現辞典」構築を目指す．はじめに文章全体を見据えた視点から「接続表現」という語を定義した後，大規模コーパスのテキストデータから人手とシステムの両方の方法で「接続表現」568項目を選定し，自動判定の必要性と問題点を述べた．最後に「接続表現」568項目からレジスター判定を人手で行い，学習者が目的とする論文にふさわしい表現が「接続表現辞典」の構築によって獲得できる可能性を示した．

View full abstract