Romance Syntax [antikvár]

1 PAOLA BENINCA GIAMPAOLO SALVI INTRODUCTION I. The relevance of Romance syntax for the development of Generative Syntactic theory The chief aim of this collection is to provide students of Romance linguistics with certain fundamental essays as a basis for outlining the syntactic patterns of the better-analysed Romance languages, with particular reference to the theory of functional structure. In this introduction, we also want to emphasise that the study of Romance languages has been central to the development of generative syntactic theory in many respects. The study of French, started by Kayne (1975), substantially broadened the empirical basis of syntactic theory. As French is a member of a large family of well-documented languages, syntactic comparison could subsequently deal with minimally different varieties and concentrate on more precise details of the theory. It was Kayne's monograph on French, in fact, which inaugurated in 1975 the systematic comparison of English, the best-described language at the time, with another language. French was different enough to render the comparison challenging for the theory, but at the same time similar enough to render the comparison a theoretically significant one. Only because the two languages were similar, was it possible to perform a kind of experiment whereby the main variables were kept under control and the differences could be correctly interpreted by inserting them in a common grammatical scheme. It soon became apparent that English and French were too distant to enable one to test precise hypotheses concerning the properties that were to be linked to other properties as consequences of a single abstract difference. For example, French has agreement of the past participle and a set of object clitic pronouns, where English has not. A first set of hypotheses could be made exploiting this difference, but in order to narrow them and test them, we soon need to 'go into' the phenomena and break them down into smaller components. In French, the past participle agrees with a surface subject of unaccusative verbs and passives, with relativised objects and clitic object pronouns, it does not agree with the surface subject of transitive verbs, nor with an NP object. We need to know if these characteristics are an essential part of the phenomenon, or derive from other characteristics of the language we are dealing with. To be able to answer this question, we have to compare French with a language which is minimally different from French, yet has a different kind of past participle agreement. This language cannot be English, it has to be a Romance language. This way of approaching the study of syntax laid the basis for the study of Italian and pushed the comparison towards the observation of languages closer and closer to each