论英语方位倒装句的动词限制条件

Constraints on Verbs in English Locative Inversion Construction

方位是人类语言表达的基本范畴之一。PPloc+V+NP的格式在许多语言中用于表示方位,所不同的是,这一句法格式在有些语言中属于正常语序,但在另一些语言中却属于非正常语序。英语是后一种情况。英语倒装句引发的争议一直以来是围绕这一结构中动词的限制条件展开的。在相关研究中先后出现了三种颇具影响力的假说,它们分别是Hoekstra和Mulder的小句分析说、Culicover和Levine的轻倒装与重倒装说以及Birner的动词弱信息说。前两种假说是讨论动词的句法-语义限制条件,后一种假说则关注动词的语用限制条件。这三种假说都试图解决前人研究遗留下来的问题,却都有自己的研究盲区。英语倒装句的动词限制涉及句法、语义和语用多重因素。因此,综合考虑句法、语义和语用的限制条件,是探究英语倒装结构中动词性质的正确途径。

One of the controversial issues regarding the locative inversion construction （LIC） in the English language relates to the constraints on verb types. There are three influential hypotheses in this area, known as Hoekstra and Mulder＇s analysis, Culicover and Levine＇s analysis and Birner＇s analysis respectively. The early understanding assumed that it was the unaceusative verbs, which were permitted in LIC, whereas, in the most cases, the unergative verbs were not permitted, except for the subtype of motion, such as walk, enter, when they were used together with a directional PP. Hoekstra and Mulder （1990） propose that LIC embeds a small clause, which licenses the so-called process of unaecusativization -- the shift of unergative-type to unaceusative-type, when an unergative verb of motion occurs in a directional construction. However, Culieover and Levine （2001） argue that it is the distinction between light inversion and heavy inversion, rather than the distinction between unaccusative verbs and unergative verbs, that predicts the acceptability of LIC, due to the fact, noticed by Levin and Hovav （1995）, that many other unergative types may occur in LIC, including swim-type, glitter-type, flutter-type, and even work-type, along with the unergative verbs of motion. Culicover and Levine＇s analysis suggests that the acceptability of a unergative verb should be highly associated with the co-occurrence of a heavy subject in the same sentence. But the further data research, conducted by Birner （1994; 1995）, Birner and Ward （1998） and Hovav （1995）, support the evidence that subjects in the majority of LIC are heavy, since they have to carry complex new information. Instead of the syntactic-semantic approach, Birner＇s analysis adopts the pragmatic approach and explores the pragmatic constraints of informationally-light verbs. The above three analyses shed light on the conclusion that it is the integration of syntactic, semantic and pragmatic factors that contributes to the constraints on verbs in LIC. Why is the sentence ＂Into/Out of the room walked John＂ more acceptable, while ＂ To the room walked John＂ is less acceptable？ It is the syntactic-semantic explanation which permits the unergative verb walk to occur in LIC through the process of unaccusativization. The difference between the ＂into/out of the room＂ and ＂to the room＂ is that, the former emphasizes the bipolarity of motion, while the latter the process of motion. The pragmatic explanation makes it clear that the initial position in a sentence must provide the topicalized information, which can be offered by the bipolarity of motion, but not by the process of motion.