對中文附加語及並列語孤島制約的測試

Testing Adjunct and Conjunct Island Constraints in Chinese

越來越多的語法學家在例句判斷上除了考慮自身語感，也開始用實驗方式蒐集並分析無相關背景之受試者對語句的接受度。本文即以實驗測試以下三項假說：（一） 從並列語孤島中提取成分較之從附加語孤島中提取成分更容易接受。（二）從附加語孤島中提取成分不僅在語言處理上有困難，而且也違反語法。（三）中文的關係化由位移生成，主題化則否。本文的實驗結果僅支持這三項假說中的第一項。此外，本文所使用的語句接受度實驗成效快且操作簡便。

A growing number of syntacticians are supplementing their own intuitions with formal experiments, collecting and analyzing acceptability judgments from theoretically naïve native speakers. This paper applies this experimental approach to test a set of interrelated hypotheses in Chinese syntax: that extraction from conjunct islands is more acceptable than extraction from adjunct islands; that extraction from adjunct islands truly violates grammar rather than merely affecting sentence processing; and that relativization involves movement while topicalization does not. The results support the first hypothesis but challenge the other two. The study also demonstrates how quick and simple formal judgment experimentation can be.