L'Inserzione vocalica nelle lingue romanze, con particolare riguardo per le varietà  italo-romanze

The title of my doctoral dissertation is "Vowel Insertion in Romance Languages, with Special Reference to Italo Romance Varieties". The main purpose of my work is to propose a principled classification of non etymological vowel insertion sensitive to the phonetic phonological context (henceforth vocal insertion) in the Romance languages in a diachronic perspective. The traditional classification of vowel insertion is based on the positions in which the phenomenon occurs. Hence, three positions are generally recognized: prosthesis in word initial position, epenthesis in word medial position, and epithesis in word final position. In the linguistic literature, this classification is broadly accepted a priori, yet the nature of vowel insertion in Romance languages is better captured if we consider other factors, as segmental and phonotactic context. Thus, we propose a revision of that tradizional classification. An essential bibliographic contribution was the work of Sampson (2010), Vowel Prosthesis in Romance, where he identifies three categories of "prostheses", providing various examples from many Romance languages. My classification was obtained by applying the criteria chosen by the scholar to other positions. The result of my research was the identification of five major categories, which are called InsVoc S, InsVoc R, InsVoc C, InsVoc Oss and InsVoc Nat respectively. InsVoc stands for vocal insertion; the following letters represent the relevant context in which each category operates. Thus S represents the initial etymological cluster; R represents the presence of the rhotic or sonorants; C represents the final consonant and Oss represents oxytone. Vocal insertion has operated mainly adjacent to the aforementioned contexts in diachrony. The four categories just mentioned have worked in words derived directly from Latin, while InsVoc Nat has only targeted loan words, as vowel insertion is one of the most frequently observed strategies for adapting the loan words to the phonological structure of the receiving language, i. e. nativization, which is represented by Nat. The dissertation is divided into two main parts: the first part discusses the general properties of vocal insertion (mainly its activation and diffusion within a linguistic system) as a diachronic process from a cross linguistic point of view, with special attention to the Romance domain. In the literature, vocal insertion is widely known by scholars of the eighteenth century (eg Schuchardt 1867) and several attempts have been made so far to describe its behaviour and to explain why it occurs in various theoretical frameworks. Nevertheless, after a century and a half of studies, the nature of vocal insertion is still unclear in all its aspects. The purpose of the first part of the thesis is to propose a possible synthesis of vocal insertion studies in the literature and to discuss a possible coherent and global interpretation of its phenomenology. In the second part I consider the properties of each major category of vowel insertion in Romance languages. I discuss various aspects of the major categories such as diatopic distribution, possible causes and possible ways of spreading within a language system, based on data obtained in different Romance varieties. As I mentioned above, the nature of vocal insertion has not yet been clarified in all its aspects, and in the future, new proposals and interpretations will certainly come forward. The interpretation I proposed in the first part of my work could be denied in future studies, but irrespective of its theoretical account, the fact remains that vowel insertion in the Romance words inherited directly from Latin works canonically in the contexts described above, represented in my work by the letters S, R, C and Oss. Any future attempt at interpreting this phenomenon would have to consider these contexts, and the identification of these contexts is precisely the main scientific contribution of my dissertation