On the Role of Formal and Semantic Features in the Formation and Interpretation of Syntactic Chains

假设句法移位，如wh-移位，会产生一条记录派生历史的链。本研究的目的是通过对操作者移位、NP移位和日语乱码的比较，来考察连环的形成机制和解释。首先，在对绑定和作用域重构进行详细研究的基础上，提出了在着陆点复制移动物品的所有特征形成链，然后在起始点或着陆点通过删除每个特征来解释链。这一建议使我们能够以精确的术语制定操作员移动和NP移动的解释程序，更重要的是，使我们能够为重建日本乱七八糟的复杂模式做出贡献。此外，在随后的研究中，我们还解释了这三种移位与省略的不同互动方式。在此基础上，考察了NP移位和错位在被移项的主观性及其参与一致性方面的差异，并得出结论：不仅T而且Vis都具有触发NP移位的固有的EPP特征。本文还考察了日语左边缘句的话题和焦点解释，并与意大利语进行了比较，证明了所提出的链形成和解释机制在解释句子的语篇属性方面是相当有效的。最后，基于日语完全缺乏句法一致性的事实，笔者提出了一个假设，即一致性特征的存在与否不仅影响移位操作，而且还决定了论元省略在语言中是否可能存在。这项研究目前正在发展为我计划承担的下一个重大项目。

It is assumed that syntactic movement, such as wh-movement, produces a chain that records the history of derivation. The purpose of this research project was to investigate the mechanism for the formation and the interpretation of chains, comparing operator movement, NP-movement, and Japanese scrambling. First, based on detailed research on binding and scope reconstruction, I proposed that chains are formed by copying all features of the moved item at the landing site, and then are interpreted by deletion of each feature either at the initial site or at the landing site. This proposal allows us to formulate the interpretive procedures for operator movement and NP movement in precise terms, and more importantly, enables us to amount for the complex pattern of reconstruction Japanese scrambling exhibits. In addition, it was shown in subsequent research that the different ways in which the three types of movement interact with ellipsis are also explained as a consequence of this proposal.Further, I examined the differences between NP movement and scrambling with respect to the subjecthood of the moved item and its participation in agreement, and arrived at the conclusion that not only T but also vis equipped with an inherent EPP feature that triggers NP movement. I also investigated the topic and focus interpretation of the Japanese left periphery, in comparison with Italian, and demonstrated that the proposed mechanism of chain formation and interpretation is quite effective in the explanation of the discourse properties of sentences. Finally, based on the fact that Japanese completely lacks syntactic agreement, I presented a hypothesis that the presence/absence of agreement features not only influences movement operations but also determines whether argument ellipsis is possible in the language. This research is currently being developed into the next major project that I plan to undertake.