Presented by Romance Linguistics in the Antiopodes (RomLA)

Spanish clitics are special clitics (Zwicky 1977). They have been analysed as agreement markers (Jaeggli 1982, Suñer 1988), stem-level inflectional affixes morphologically attached to the verb (Andrews 1990, Spencer & Luís 2012), and independent syntactic elements adjoined to functional heads (Kayne 1990; Uriagereka 1995) among others. However, there is general consensus on clitics being an interface phenomenon.  In this paper I focus on the morphology-syntax-information structure interfaces in clitic doubling in language contact situations, where divergence at the interfaces results in hybrid/split or new clitic systems. The transition from co-reference in standard varieties to object marking in contact varieties (Andean Spanish) is marked by feature and case erosion or case extension, including locative doubling challenging the existing typologies. In order to link these grammaticalized clitics to information structure roles, I propose a fine-grained distinction between first and second person clitics and third person indirect and direct object clitics on the one hand, and the ‘featureless’ third-person clitic lo, also known as strange lo, on the other hand. In this novel classification (Toivonen 2003), which is based on the two parameters syntactic projectivity and phonological dependence, all Spanish clitics but the accusative clitics lo/la can be analysed as phonologically dependent and non-projecting. I will argue that accusative lo/la are also phonologically dependent but contrary to all other clitics they show weak projecting use as determiners (Mayer 2010, in prep). These clitics are still subject to strict locality constraints to the left edge of the verb and satisfy the Kayne (1975) and Zwicky & Pullum (1985) criteria for clitichood.   

Join us for drinks and nibble afterwards.