This study conducted a corpus-based investigation of collexemes for active-passive alternation found in the English part of an English-Japanese parallel corpus as an attempt to use them as metrics for distinguishing native English and non-native English. The results show that some verbs in the data are used more often in the active voice than the passive voice, and vice versa, and the differences are statistically significant. However, these verbs are not the same as those found in a previous study. This fact supports the claim that activepassive alternation constitutes a lexico-semantic phenomenon that is sensitive to various factors, such as differences in genres and type of the authors of the text.
|Number of pages||6|
|Publication status||Published - 2019|
|Event||33rd Pacific Asia Conference on Language, Information and Computation, PACLIC 2019 - Hakodate, Japan|
Duration: 13 Sep 2019 → 15 Sep 2019
|Conference||33rd Pacific Asia Conference on Language, Information and Computation, PACLIC 2019|
|Period||13/09/19 → 15/09/19|