The English dative alternation, which involves an alternation between the double object construction (DC) and the adpositional construction (AC), has been known to be one of the most problematic areas for L2 learners. This paper investigates the factors influencing the Korean English as a foreign language learners’ choices between the two constructions. The Korean section of the TOEFL11 corpus was used and sentences with dative constructions were extracted from the corpus. Nine factors were then encoded and analyzed with a generalized linear mixed-effect model. From the analysis, the following facts were observed: (1) Korean EFL learners’ choice between the DC and the AC was influenced by five factors: animacy of theme, definiteness of theme, pronominality of recipient, length difference between theme and recipient, and individual verbs; (2) despite some similarities in choice of alternation between Korean EFL learners and ENL speakers, the two groups also exhibited differences, and (3) these differences originated from the different effects of these factors.

Present study investigates whether the syntactic constraints of English reflexives and pronouns originally proposed in Standard Binding Theory (SBT; Chomsky 1980, 1981, 1986) are well-reflected in written corpus data in British English by using the ICE-GB corpus. Four linguistic factors including structural relations between the reflexives/pronouns and their antecedents were used to analyze 1,000 sentences (400 reflexives and 600 pronouns) extracted from the ICE-GB corpus. The results demonstrated the following: i) The linguistic factors related to binding conditions showed structural differences between reflexives and pronouns in English; ii) English reflexives showed more cases of discourse binding (i.e., binding outside the sentential boundary) than the original expectation of SBT; iii) With the domain of sentential binding, structural constraints such as c-commanding and binding domain (Governing Category: SSC and TSC) were nearly violated with reflexives.

Yong-hun Lee, JeeHee Yu, and Tae-Jin Yoon. 2017. Predicting the Occurrence of the English Modals Can and May Using Deep Neural Networks. Studies in Modern Grammar 96, 167-189. This paper tries to provide a computational modeling of language processing using deep neural networks. For this purpose, the corpus data in the ICE-USA was used. After all the sentences with can and may were encoded with eighteen linguistic factors, the annotated data were fed into the deep neural networks (DNN). The DNN was constructed with three layers, and each layer contained seventeen nodes. After the DNN was constructed, the learning process was performed with a training set. Then, the performance was measured with a test set. The processes were repeated one hundred times, and it was observed that the DNN had the classification accuracy of 91.5%. The results are promising in that reliable methods can be used in automatically classifying the frequently used modal auxiliary on the basis of the deep learning system.

The study investigates the validity of ‘subjecthood diagnostics’ in Korean proposed in the previous studies. Based on the studies on subjecthood diagnostics in Korean Single Subject Constructions (non-MSCs) and Multiple Subject Constructions (MSCs), the role of two subjecthood diagnostics―Honorific Agreement (HA) and Plural Copying (PC)―in these two constructions are examined through an empirical syntactic experiment. Seventy Korean native speakers participated in the experiment using acceptability judgment task designed to probe their intuition. The results show (i) that in the Non-MSCs, Honorific Agreement (HA) reliably seems to choose the subject (i.e., the only Nom-marked NP), whereas Plural Copying (PC) does not seem to pick out the subject in a less consistent manner; and (ii) that in the MSCs, neither Honorific Agreement (HA) nor Plural Copying (PC) seems to be a reliable subjecthood diagnostic. Possible reasons for these results are discussed.

Parasitic gap is one of the interesting topics in syntactic literature since its existence depends on a real gap in the sentences. Both English and Korean have parasitic gap constructions, but their syntactic behaviors are a little different. Lee (2010b) discussed how parasitic gap constructions in these two languages are different and provided Combinatory Categorial Grammar (CCG) analyses for the constructions. The operation functional substitution plays an important role in the CCG analysis of the construction. The goal of this paper is to provide computational implementation based on Korean Type-inherited Combinatory Categorial Grammar (TCCG) system (Lee, 2010a), which is a combination of Steedman’s CCG and the Linguistic Knowledge Building (LKB) system. Two types of output data will be provided as outputs of the implementation. One is the syntactic parse tree, and the other is the semantic representation. In the parse trees, category combinatorics is encoded. For the semantic representations, Minimal Recursion Semantics (MRS) is adopted, and all the semantic relations are represented in the MRS structure.