In this paper, we focus on Saikaku Ihara's posthumous works collections, especially Yorozu no humihougu (万の文反古) (A Scrapbook of Old Letters; 1696). Saikaku Ihara (井原西鶴) is one of the most famous writers of the Edo period (1603– 1868) in Japan. It is said that he wrote 23 works in 10 years. However, that achievement has not been fully verified.
Identifying authors of novels in this era is difficult, because typically their authors did not sign the books. In addition, Saikaku's posthumous works were edited and published from 1693 to 1699 by his student Dansui Houjyou(北条団水). It became more difficult to identify the authors of Saikaku’s posthumous works.
In our study, we focused on Yorozu no humihougu because many Saikaku’s researchers have raised questions about the authorship. Saikaku researchers have tried to identify his works by investigating their history, content, format and so on. However, it remains unclear which works are really by Saikaku. Meanwhile, the potential of quantitative analysis of textual data and the related field of the digital humanities have also dramatically advanced. For this reason, this study verifies the text of Yorozu no humihougu using a quantitative approach.
II. Database of Saikaku's works
To resolve the Saikaku authorship problem, we made a database of his works with Saikaku researchers, so this database is the only one on Saikaku's works presently and has a high degree of reliability. We made this database based on Shinpen Saikaku Zenshu (新編西鶴全集) (Figure 1).
Fig. 1: 17th century publication
Table.1 is a part of the database from Saikaku’s works used for this analysis. All sentences were divided into words. Moreover, information was added as required for the analysis. According to our database, the total words in 23 Saikaku works are about 578,617 words.
Fig. 2: Database of Saikaku’s works
III. Analysis and results
In general, Saikaku’s works are made up of many short stories (chapters), so we used information of each chapter in our analysis. In addition, we used four posthumous works other than Yorozu no humihougu as one group (Table 2). Then, we compared Yorozu no humihougu to four other posthumous works.
Fig. 3: Saikaku's posthumous works
At first, we examined the appearance rate of the seven principal parts of speech: nouns, particles, verbs, auxiliary verbs, adjectives, adverbs and adnominal adjectives. Figure 2 shows the results of the analysis on the appearance rate, using the principal component analysis (PCA) with a correlation matrix. The horizontal axis shows the importance of the first principal component, and the vertical axis shows the second. In this figure, indicating differences revealed by PCA, 17 chapters of Yorozu no humihougu is on the right and 83 chapters of four other posthumous works on the left.
Fig. 4: CA results for Yorozu no humihougu (万の文反古) posthumous works (These circles drawn on the figure are 95% confidence ellipse)
Upon examining the first principal component, we found that differences in the use of particles and verbs (Table 3). In Yorozu no humihougu, the appearance rate of verbs is higher and those of the particles are lower, compared with four other posthumous works.
Fig. 5: The result of PCA
Furthermore we examined by Welch’s t-test at the 0.05 significance level (Table 4). Significant differences were found in verbs and particles.
Fig. 6: The result of Welch’s t-test. (Note: a positive t-value indicates that the word is characteristically used in four other posthumous works)
Next, we examined the words of verbs and particles characteristically used in each work by using Welch’s t-test for determining whether there exists a difference between the averages of Yorozu no humihougu and four other posthumous works. Of the 1,625 types (15,113 words) of verbs, Welch’s t-test was performed on each of the 47 types (8,055 words) that appeared more than 51 times. Similarly, of the 53 types (29,994 words) of particles, Welch’s t-test was performed on each of the 31 types (29,842 words). We examined 10 words with the smallest p-values.
Table 5 shows that “sourou,” “mousu,” and “gozasourou” are the most frequently used words of verbs in Yorozu no humihougu, and “su,” “iu,” “naru,” “ari,” “yuku,” “sumu,” and “tamau” are in four other posthumous works. From these results, honorific words of verbs are assumed to be characteristically used in Yorozu no humihougu.
Fig. 7: The result of Welch’s t-test. (Note: a positive t-value indicates that the word is characteristically used in four other posthumous works)
Table 6 shows that “he,” “bakari,” “ni,” “nite,” and “ha” are the most frequently used words of particles in Yorozu no humihougu, and “zo,” “te,” “do,” “kasha,” and “tote” are in four other posthumous works.
Fig. 8: The result of Welch’s t-test. (Note: a positive t-value indicates that the word is characteristically used in four other posthumous works)
We conducted the comparative analysis among Yorozu no humihougu and four other posthumous works using a quantitative approach. Yorozu no humihougu was revealed to be characterized as having a higher the appearance rate of verbs and a lower the appearance rate of particles than in four other posthumous works. Furthermore, we analyzed the words of verbs and particles characteristically used in each work by using Welch’s t-test, in Yorozu no humihougu, honorific verbs are used more frequently than in four other posthumous works, while the particles “he,” “bakari,” “ni,” “nite,” and “ha” appear more often. These results indicate that Yorozu no humihougu and four other posthumous works are quite different in the appearance rates of parts of speech as well as the words of verbs and particles.
Among the works created by Saikaku, Yorozu no humihougu is the only work written in an epistolary style. Analyses on the characteristics of the epistolary style of writing should be performed in the future in order to clarify any doubt concerning the author of Yorozu no humihougu.
In addition to the appearance rate of parts of speech, other information such as the appearance rate of words, as well as the works of Dansui Houjyou, should be analyzed.
Noboru Asai et al. (2000), “Shinpen Saikaku Zenshu (新編西鶴全集)” vol.1–4, Benseishupan
Munemasa isoo (1969), “Kanazoushi kara Ukiyozoushi he (仮名草子から浮世草子へ)”, Shibundou
Yamaguchi takeshi (1929), “Saikaku meisakushu ge (西鶴名作集下)”, Nihonmeityo zenshu kankoukai
Teruoka yasutaka (1953) “Saikaku kenkyu note (西鶴研究ノート)”, Tyuuoukouronsha
Taniwaki masachika (1981) “Saikaku kenkyu ronkou (西鶴研究論攷)”,Sintensha
Nakamura yukihiko (1982), “Nakamura yukihiko cyogitsushu (中村幸彦著述集)”, Tyuuoukouronsha
If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.
Hosted at École Polytechnique Fédérale de Lausanne (EPFL), Université de Lausanne
July 7, 2014 - July 12, 2014
377 works by 898 authors indexed
XML available from https://github.com/elliewix/DHAnalysis (needs to replace plaintext)
Conference website: https://web.archive.org/web/20161227182033/https://dh2014.org/program/
Attendance: 750 delegates according to Nyhan 2016
Series: ADHO (9)