现代汉语“得”字句的语段推导及线性化  

The Db P derivation and linearization of modern Chinese “De”sentences

作　　者：杨军[1] 李琳琳[1] YANG Jun LI Linlin(School of Foreign Languages, Nanjing University of Posts and Telecommunications, Nanjing 210023 , China)

机构地区：[1]南京邮电大学外国语学院,江苏南京210023

出　　处：《南京邮电大学学报（社会科学版）》2016年第4期98-104,共7页Journal of Nanjing University of Posts and Telecommunications(Social Science Edition)

摘　　要：基于边缘特征等于可合并性这一假设,尝试在边缘特征驱动的语段推导框架下,证明功能语类"得"在现代汉语"得"字句语段推导中的必要性,以及"动-得"线性紧邻的必然性。"得"字句动词复制的原因,在于当"得"前动词带宾语,宾语必须递出,这就在C(Complementizer)的移动路径上形成两个语障,即限定词短语和该限定词短语所在轻动词短语;作为一种句法后修复手段,话题化使整个宾语所在轻动词短语在CP(Complementizer Phrase)之上再合并,经过链删除,消除较低的语障,但动词却不能删除(这造成动词复制),原因是在尽可能删除较低副本的条件下,保留递出单位的核心是不违反强制曲拱原则的最优选择。This paper attempts to show, assuming EF = mergibility, that in EF-driven Derivation byPhase of modem Chinese De sentences, De , as a functional category, is derivationally necessary, andthat the first predicate verb is linearly adjacent to De （i. e. , no word（s） should appear in-between） , which explains why De must adhere to V. And we arrive at the conclusion that when pro-De V takes a DP object, which derivationally constitutes a phase like the vP it proceeds to merge, two phases （ i. e. , barriers） appear on the path of C moving to the top of the sentence. As a repair strategy, the whole vY is topicalized, that is, remerged above CP so that one of the two barriers （the object DP） can be removed via Chain Modi-fication just to make the whole derivation linearisable. Pre-Z）e verb doubling, in this case, is forced as the optimal solution to avoid violation of OCP （ ＊ ee） because there would be two adjacent empty spell-outs if the lower vV is removed entirely.

关 键 词：“得”字句 语段推导 边缘特征 线性化 

分 类 号：H1-0[语言文字—汉语]