Processing two types of ditransitive sentences in Turkish: Preliminary results from a self-paced reading study

摘要

In Japanese sentence processing, Miyamoto and Takahashi (2002, 2004) showed that the [NOM>DAT>ACC] order is easier to process than the [NOM>ACC>DAT] order, and argued that the filler-gap dependencies which is driven by the case markers can explain the processing difficulty of scrambled ditransitives. On the other hand, Sakai et al. (2009) showed that the animacy has also impact on the processing of ditransitives in Japanese. Yet, it is unknown whether case markers and animacy affect the processing of ditransitive sentences and their word order alternations in other languages as well. In the present study, by manipulating the animacy of DAT-noun in Turkish, we first investigated how ditransitive sentences are processed through two self-paced reading experiments. We then evaluated whether the case markers and animacy affect the processing of ditransitives and their word order alternations in Turkish. The results of Experiment 1 showed that when the DAT-noun was animate, the participants read the [NOM>DAT>ACC] order faster than that of the [NOM>ACC>DAT]. This result is consistent with Miyamoto and Takahashi's, and the filler-gap dependency can account for. On the other hand, when the DAT-noun was inanimate, the [NOM>ACC>DAT] order was read faster than the [NOM>DAT>ACC] order. This result cannot be accounted for by the case driven filler-gap dependencies. Overall, our study indicates that semantic factors such as animacy also have an impact on the processing of ditransitive sentences in Turkish.%Miyamoto and Takahashi（2002，2004）は日本語で［が＞に＞を］語順の方が［が＞を＞に］より処理し易いという結果を報告しており，これを格助詞に駆動される埋語と空所の依存関係で説明できると述べた。一方で，Sakai et al．（2009）名詞の有生性も日本語の二重目的語構文の処理に関与するという結果を示した。しかし，日本語以外の言語では格助詞と有生性が二重目的語構文とそのかき混ぜ文の処理に影響を及ぼすか否かについて，まだよく知られていない。本研究では，トルコ語における二重目的語構文の与格名詞の有生性を換作し，二つの自己ペース読文課題を通して，二重目的語構文とそのかき混ぜ文がどのように処理されるのかを調査し，格助詞と有生性はトルコ語の二重目的語構文の処理にも関与するのか考察した。与格名詞が有生であった実験1では［主格＞与格＞対格］語順の方が［主格＞対格＞与格］より処理時間が短かった。この結果はMiyamoto and Takahashiの結果と一致しており，埋語と空所の依存関係で説明できる。一方で与格名詞が無生であった実験2では，［主格＞対格＞与格］語順の方が［主格＞与格＞対格］より読み時間が短かった。この結果は，格助詞に駆動される理詰と空所の依存関係によって説明できず，名詞の有生性のような意味的な要因もトルコ語の二重目的語構文に処理に影響を及ぼすことが示唆された。