Séminaire de Recherche en Linguistique

The aim of the talk is twofold: on the one hand it provides a schematic overview of the historical evolution of clitic placement in the Romance languages; on the other, it elaborates on the emergence of clitic systems from the syntax of Latin ‘weak’ pronouns. 

            In Latin, pronouns with a “non-emphatic” interpretation usually occur after focalised constituents (denoting either contrastive or new information) or operators (such as wh elements or bare quantifiers) (Thurneysen 1892: 302; Krisch 1990: 68; Adams 1994a; Salvi 1991, 1994, 1997, 2004; Devine & Stephens 2006: 277-312; Spevak 2010: 214).

(1)          a.           Alium illa       amat, non illum                                              (Plaut. Bacch. 593)

                             ‘She loves another, not him’  

                b.         Quid tibi vis dicam nisi quod viderim?                               (Plaut. Miles 300)

                            ‘What would you have me say to you, but that I did see her?’

According to Salvi, these are instances of weak pronouns (in the terms of Cardinaletti & Starke 1999), which successively turned into clitics (Salvi 1991, 1994, 1997, 2004). In my talk, I will review Salvi’s analysis in the light of current minimalist assumptions. In particular, following Longobardi’s 2001 Inertial Theory, I assume that ‘linguistic change […] may only originate as an interface phenomenon’, while ‘syntax, by itself, is diachronically completely inert’ (Longobardi 2001: 277–278). Hence, change or variation affect the Lexical representation of functional elements, but never affect the principles of syntactic computation, i.e. the algorithm of set formation yielding phrase structure. If so, two main questions arise w.r.t. the emergence of Romance clitics:

-          Which interface phenomena caused the emergence of a double series of pronouns?

-          How are Romance clitics represented in the Lexicon across different languages or chronological stages?

-          How to model change/variation in phrase structure?