Subject Realization in Japanese Conversation by Native and Non-native Speakers: Exemplifying a New Paradigm for Learner Corpus Research

In the field of Learner Corpus Research, Gries and Deshors (Corpora 9(1):109–136, 2014) developed a two-step regression procedure (MuPDAR) to determine how and why choices made by non-native speakers differ from those made by native speakers more comprehensively than traditional learner corpus research allows for. In this chapter, we will extend and test their proposal to determine whether it can also be applied to pragmatic and grammatical phenomena (subject realization/omission in Japanese), and whether it can help study categorical differences between learner and native-speaker choices; we do so by also showing that the more advanced method of mixed-effects modeling can be very fruitfully integrated into the proposed MuPDAR method. The results of our study show that Japanese native speakers’ choices of subject realization are affected by discourse-functional factors such as givenness and contrast of referents and that, while learners are able to handle extreme values of givenness and marked cases of contrast, they still struggle (more) with intermediate degrees of givenness and unmarked/non-contrastive referents. We conclude by discussing the role of MuPDAR in Learner Corpus Research in general and its advantages over traditional corpus analysis in that field and error analysis in particular.