| dc.description.abstract | This thesis describes and analyses (i) Cinque's (1999) hierarchy, (ii) verb movement, and (iii) clitic placement in a selection of Italo-Romance languages spoken in Sicily (southern Italy). The local languages of Sicily I investigate in this thesis are the ones spoken in the following towns: Castelvetrano, Valderice (Trapani); Misilmeri, Palermo (Palermo); Modica, Ragusa (Ragusa); Capizzi, Itala, Mazzarr` a Sant'Andrea (Messina); Belpasso (Catania); Delia (Caltanissetta); Favara (Agrigento); and Buscemi (Siracusa). All the empirical evidence proposed throughout this thesis, unless explicitly stated otherwise, has been collected by the Author through fieldwork in loco between September 2021 and December 2024. As far as I am aware, there are not any works which focus specifically on Cinque's (1999) hierarchy, verb movement, and clitic placement in the local languages of Sicily. The three topics that I investigate are interrelated and shape the organisation of my thesis as well as the timeline of my fieldwork in Sicily. More precisely, in Chapter 1, I discuss the research questions, the research methodology, the fieldwork, and lay out my theoretical assumptions. Then, in Chapter 2, I aim to test theoretically Cinque's (1999) hierarchy and build my own hierarchy for the local languages of Sicily. I test combinations of AdvPs, which occupy the specifier of a specific functional projection, to determine their order and map the inflectional domain in Sicily. In Chapter 3, I focus on verb movement: I test theoretically the analysis of verb movement put forward by Schifano (2018) and investigate the placement of finite lexical verbs, the perfective auxiliary HAVE, and (active) past participles across Sicily. In conducting this task, I use, as an empirical base, the hierarchy of functional projections that I built in Chapter 2 for the languages of Sicily. In Chapter 4, I aim to test theoretically clitic placement theory, as well as describing and analysing formally the distribution of pronominal clitics in the languages of Sicily. Specifically, Romance clitics are weak pronominals which must attach either proclitically or enclitically to a verbal host, i.e., they must either precede or follow a verb. In Chapter 3, I determined the position of verbs in the inflectional domain and, consequently, use these findings as an empirical base to investigate the placement of clitics. The main findings of this thesis can be summarised as follows. In Chapter 2, I present Cinque's (1999) hierarchy modelled on the results of my fieldwork in Sicily which provides strong evidence in favour of Cinque's (1999) claims. In Sicily, adverbs such as sempri and ancora can have different readings and thus lexicalise distinct functional projections, both in the HAS and the LAS. I also discuss the negative adverb manco 'not even' and the preverbal negative marker. The first seems to behave as a focaliser, lexicalising distinct positions in the IP, whereas the latter occurs in Neg (i.e., a syntactic head X) right below T(Anterior)P. I use these findings to test the placement of verbs in Chapter 3. More precisely, I test the validity of Schifano's (2018) analysis of verb movement in Romance and argue, following new empirical evidence, that verbs across Sicily consistently target the A(SPECT) field. Specifically, finite lexical verbs vacate the verbal domain and occupy a position sandwiched between the right of T(Anterior)P and the left AspPerfectP. I detected microvariation in the movement of the verb around AspContinuativeP, where I grouped Sicilian towns into two types. I also observed that the perfective auxiliary, which I claim to be base-generated in vAux, does not exhibit higher movement than finite lexical verbs as it cannot occupy the HAS, and lands right below T(Anterior)P. As for past participles, these must remain structurally adjacent to the auxiliary (with some exceptions). I also explained the 'unusual' word orders that I detected with the adverb chiu 'no longer' in SpecAspTerminativeP which seem to provide some counterarguments to Cinque's (1999) assumptions. I argued that this adverb is a NPI and that its licensing requirements override verb placement. In Chapter 4, I account for clitic placement by using verb movement as a diagnostic, since clitics must attach to a verbal host. Across Sicily, clitics must attach proclitically to finite lexical verbs, proclitically to the perfective auxiliary, and either proclitically or enclitically to restructuring verbs. I argue that an analysis of cliticisation based on Agree and defective Goals` a la Roberts (2010) accounts for the Sicilian data and, more importantly, explains why some word orders are disallowed (e.g., enclisis to the finite lexical verb). The clitic attaches (or, incorporates) to its verbal host in the verbal domain, as a consequence of Agree between a series of functional heads vs and the clitic, which is claimed to consist (only) of a bundle of valued phi-features, hence lacking phonological/lexical specification. In addition, this analysis accounts for the lack of adverb interpolation, PCC effects, and the order of clitics in a cluster. | en |
