LONG DISTANCE REFLEXIVES AND ISLANDHOOD IN CHINESE: HEAD MOVEMENT VERSUS ADJUNCTION TO IP / 汉语的长程反身词与孤岛关系------------[主要语移分析] 与 [小句子加接分析] 之比较

The head movement analysis of long distance reflexives has been shown to provide an explanation of many of the properties of ziji in a wide variety of environments. The analysis, however, appears on initial examination to make the wrong predictions regarding the appearance of ziji in islands. In this article we show that these incorrect predictions do not occur if the head movement analysis is combined with an appropriate theory of Chinese phrase structure. We further show that the leading competing movement-based theory of Chinese reflexives, the IP adjunction theory, is not free of difficulties with respect to islandhood. We conclude that the totality of evidence provides strong support both for head movement of ziji and for a theory of Chinese phrase structure incorporating a number of functional projections. 汉语反身词 “自己” 长程约束的现象，讨论的学者相当多，其中主张移位 (movement) 分析，主要有两个：一个是［主要语 (head) 移位分析］，另一个是［小句子加接 (IP Adjunction) 分析］。在其它相关文章里，我们已证明主要语移位的分析能解释许多“自己” 句法上的特点，但却好像不能解释“自己” 为何可以出现在一些孤岛 (islands) 句型里。本文就此提出辩驳与说明：（一） 如采用 Tang (1990) 汉语名词词组结构之分析， 上述之问题将可解决。（二） ［小句子加接分析］亦无法解释相关的孤岛句型。以上讨论皆强力支持〔主要语移位分析〕和采纳若干 functional 投射的词组结构分析。