Uncovering verb-second effects. An interface-based typology

除现代英语外，日耳曼语言的特点是有限动词经常出现在主句的第二位置，这一特性被称为“动词第二”（V2）。在经典方法中，这种结构被认为是由于有限动词被吸引到最高的小句头部(C)，同时在其外缘出现一个元素。随着更丰富的句子左边缘模型的发展，包括一系列与话语相关的功能投射，这种观点变得有问题，因为没有一个类别是唯一对应于c语言的。然而，这种发展使得将日耳曼语V2视为发生在许多其他语言中的更广泛的V2效应的一部分。例如，古罗曼语的特点是遵守一种限制，这与日耳曼语的限制非常相似，即限定动词通常不出现在子句的开头位置，而是至少出现在第二位置。该项目有三个基本目标：一方面，它旨在扩大与V2类型相关的语言的范围，除了那些相对较少的被定性为V2的语言（例如克什米尔语（印度雅利安语）、印古什语（高加索语））外，还包括具有基本主-宾-动（SOV）顺序的语言。特别是，在许多类型学文献中建立的SOV顺序与原位h-短语和焦点之间的相关性将被批判性地重新审视。根据这一观点，动词与前一个wh短语或SOV语言中的焦点之间的邻接关系对应于前语元素在VP内的配置，即原位。有人会说，在巴斯克语这种真正的SOV语言中，这种邻接关系更像是一种接近古罗曼语类型的句法结构，也就是说，它被用来显示V2效应。在突厥语言中同样的相关性的证据将被探索。为此，将研究三种突厥语言（吉尔吉斯语、哈萨克语和乌兹别克语），并辅以更彻底的土耳其语，并评估与它们可能的V2状态有关的证据。第二个目标是评估语言接触是否可能对产生V2效应产生重大影响，这在欧洲西南部是一种普遍现象，在那里巴斯克语、西班牙语、法语和加斯孔语（奥西顿语）都表现出V2效应（在不同程度上）。这项研究将由索布族（斯拉夫语）和raeto -罗曼语的研究补充，两者都表现出V2效应，并与德语密切接触。最后，该项目的第三个目标是探索一种非统一的方法来处理V2现象。具体来说，我们将研究V2是否依赖于一组独特的形态句法机制，或者更确切地说，它与一个线性概念有关，因为涉及到来自多个语法领域的众多因素，从而导致V2的不同类型。

Germanic languages, with the major exception of Modern English, are characterized by the regular occurrence of the finite verb in second position in main clauses, a property known as ‘verb second’ (V2). In the classical approach, this configuration is assumed to result from the attraction of the finite verb to the highest clausal head (C) along with the appearance of an element in its outer edge. With the development of a richer model of the sentential left periphery, which involves an array of discourse-related functional projections, this view becomes problematic, as there is no category that uniquely corresponds to C. Yet, this development has allowed looking at Germanic V2 as part of a wider set of V2 effects occurring in many other languages. Old Romance languages, for instance, have been characterized as obeying a restriction that is highly reminiscent of the Germanic one, namely that finite verbs generally do not appear in clause-initial position, but rather in at least second position.This project has three basic objectives: on the one hand, it aims at expanding the perimeter of the languages relevant to the V2 typology, by including languages with a basic subject-object-verb (SOV) order other than those relatively few examples that have been characterized as V2 (e.g. Kashmiri (Indo-Aryan), Ingush (Caucasian)). In particular, the correlation, established in much of the typological literature, between SOV order and in-situ wh-phrases and foci will be critically revisited. On this view, the adjacency between the verb and a preceding wh-phrase or focus in SOV languages corresponds to a configuration in which the preverbal element is within the VP, that is, in-situ. It will be argued that, in Basque, a bona fide SOV language, this adjacency relation rather corresponds to a syntactic configuration close to the Old Romance type, i.e. it is taken to display V2 effects. The evidence for the same correlation in Turkic languages will be explored. To this end, three Turkic languages (Kyrgiz, Kazak, and Uzbek) will be studied, complemented with the more thoroughly described Turkish, and the evidence in relation to their possible V2 status will be evaluated. The second objective is to assess whether language contact might have had a non-trivial impact in giving rise to V2 effects as an areal phenomenon in South-Western Europe, where Basque, Spanish, French, and Gascon (Occitan) all display V2 effects (to a different extent). This study will be appended by that of Sorbian (Slavic) and Raeto-Romance, both exhibiting V2 effects and being in intense contact with German. Finally, the third objective of the project is to explore a non-uniform approach to the phenomenon of V2. Specifically, it will be investigated whether V2 relies on a unique set of morpho-syntactic mechanisms or rather relates to a linear notion, in that a multitude of factors from several domains of grammar are involved, resulting in a differentiated typology of V2.