How accurate are Japanese grammar books and other descriptions of Japanese? Traditionally, scholars have relied on their own intuitions in describing various aspects of Japanese, notably grammar. However, one person’s experience with Japanese, even a native speaker’s, is quite limited. Furthermore, opinions and views based on that limited experience tend to be subjective and biased. To complement the traditional method, a corpus-based, computational approach has gained popularity in recent years. In this method, a massive collection of Japanese written texts (written language corpora) and of transcriptions of spoken language samples (spoken language corpora) is analyzed computationally, with features of interest examined in context, extracted, and quantified. This lecture will conduct a broad survey of the approach through demonstrations of software tools and a review of recent findings obtained through the approach, showing how these findings help us toward a more accurate understanding of the Japanese language.
Atsushi Fukada is Associate Professor of Japanese and Linguistics and the Director of the Center for Technology-Enhanced Language Learning in the School of Languages and Cultures at Purdue University. His research areas include foreign language pedagogy, Japanese linguistics, computational linguistics, and CALL (Computer-Assisted Language Learning).
Sponsored by UW Japan Studies Program
For more information contact firstname.lastname@example.org.