Prosodic mapping of empty syntactic categories

• Grant number: 12/23900-4
• Support type: Scholarships abroad - Research
• Effective date (Start): July 15, 2013
• Effective date (End): July 14, 2014
• Field of knowledge: Linguistics, Literature and Arts - Linguistics - Linguistic Theory and Analysis
• Principal researcher: Raquel Santana Santos
• Grantee: Raquel Santana Santos
• Host: Myriam Uribe-Etxebarria Goti
• Home Institution: Faculdade de Filosofia, Letras e Ciências Humanas (FFLCH). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Research place: Universidad del País Vasco (UPV), Spain

Abstract

Traditional descriptions of the interaction among grammatical components were generally restricted to morphology and phonology. This picture changed significantly with the emergence of prosodic theories, which propose that different grammatical components (morphology, syntax, semantics) are mapped in the phonological component creating domains where phonological rules apply. With respect to the phonology-syntax interface, in particular, there have been many works arguing, for instance, that prosody can disambiguate between readings in structurally ambiguous sentences (e.g. Lehiste 1973, Lourenço-Gomes 2003, Finger & Zimmer 2005, Magalhães & Maia 2006) and that some phonological processes can identify which empty syntactic category is present in certain structures (e.g. Lightfoot 1976, Chomsky & Lasnik 1978, Santos 2002, 2003, Nunes & Santos 2009).My work on this topic has concentrated on phonology-syntax interface in Brazilian Portuguese, investigating whether phonological empty syntactic categories are interpreted by phonology. The results indicate that pro blocks stress-shift, while traces do not (cf. Santos 1998, 2002, 2003, 2004). However, no study has yet discussed how exactly pro blocks stress shift. That being so, the aims of this project are: - to analyze how pro affects the prosodic structure. That is: if phonology has an indirect access to syntactic information (through the prosodic mapping), how does pro interfere with this mapping, in a way that traces do not? More specifically, is pro an element to be mapped in the prosodic structure or is it only computed in the mapping, but does not appear in the output?... (AU)