This article discusses the great potential of the current movement toward open data in the field of English language education. For this purpose, I introduce and illustrate an online system specially designed to aid instruction of constructional patterns in English. The system, the TED Corpus Search Engine (TCSE), uses a set of open data of English presentations from TED Conferences and provides many functionalities, including full-text utterance retrieval of both transcript segments and corresponding videos using various search operators. The article shows that the open data movement, alongside technological innovations, could greatly benefit English language educators if they increase their awareness and make necessary efforts. I also present and discuss the fact that by making use of the appropriate data, it is possible to bridge the apparent gap between work in theoretical linguistics and the practice of language education. The TCSE is equipped with several features that help the user to extract instances of constructional patterns of English from the TED Talks corpus. Their designs are based on observations and theorizations from the field of cognitive linguistics.

View full abstract

[in Japanese]

View full abstract

The present study describes the acoustic and phonetic characteristics of Japanese (L1) and English (L2) vowels produced by Japanese ESL learners, and examines the learning models in L2 pronunciation. Eleven native speakers of Japanese took part in the experiments, and the measurements were made for F1 and F2 formants of the vowels in both L1 and L2, and the examination was made to characterize the features by male and female groups to reduce speakers' variations. The results of the analysis indicate that the vowel space in L1 and L2 differs in male and female groups in the acquisition of English vowels, and the learners tend to use L1 sounds whose acoustic qualities are close to L2 ones, i.e., Japanese female learners tend to use their own vowel regions to produce American English (AE) vowels which are similar to L1 sounds. From these observations, it can be said that they seem to apply perceptual assimilation model in acquiring L2 vowels.

View full abstract