Word order in natural languages and the theory of phrase structure

Abstract

Chomsky (1995)提出的最简方案里词的线性次序与核心运算系统毫无关系根据这个假设词序参数在最简方案的模式里已被否定本文介绍两个短语结构理论Kayne (1994)提出的线性对应定理和Takano (1996)和Fukui and Takano (1998)提出的线性化理论这两个理论有不同的结论线性对应定理认为指示语-中心语-补述语为语言的普遍词序而线性化理论则认为指示语-补述语-中心语 为语言的普遍词序移位是一种得出不同词序的方法以汉语为例虽然这两个理论都能够推导出汉语主动宾的基本词序但是本文指出根据线性对应定理而设计的短语结构 理论能够解释一些与词序无关的问题例如汉语句末助词的句法问题汉语似乎提供了一些支持线性对应定理的证据 In the Minimalist Program advocated by Chomsky (1995), linear order does not play a role i n the core computation and for this reason, word order parameter is eliminated. This paper reviews two theories of phrase structure proposed recently in the literature, namely linear correspondence axiom (LCA) by Kayne (1994) and linearization by Takano (1996) and Fukui and Takano (1998). These two theories have different consequences: the LCA predi cts that Specifier-Head-Complement (S-H-C) is the universal word order whereas the latter predicts that S-C-H is the universal word order . To derive the right word order, movement takes place. Although both theories can capture the basic SVO word order in Chinese, it is shown that the theory of phrase structure under the LCA can explai n some unrelated facts regarding t he sentence final particles in Chinese. Chinese data seem to lend some additional support to the LCA.