Phonological phrasing in Spanish

This article investigates the role of syntactic and prosodic markedness constraints on the construction of phonological phrases (φ- or p-phrases) in Peninsular Spanish. The data come from a reading task of a corpus composed of 85 utterances with a wide variety of structures and constituent lengths. Four speakers read each sentence at three different speech rates (normal, slow, and fast). It is shown that the construction of prosodic structure in this language cannot rely solely on syntactic information but has to refer to prosodic markedness constraints which regulate the size and balance of phrase constituents. The proposal will be cast in a constraint-based OT approach (McCarthy & Prince 1993a), where the notion of edge alignment from Selkirk (1984) and constituent wrapping from Truckenbrodt (1995, 1999) are considered to be ranked and violable constraints. Specifically, phonological phrasing in Spanish is determined by the interaction of right-alignment of syntactic and phonological phrases (Align-XP,R) with a maximal requirement on the length of p-phrases (Max-Bin) and a minimality constraint on the prosodic parsing of utterances (Min-Bin). Other Romance languages (and English and recently Egyptian Arabic) have also provided critical evidence in favor of the importance of prosodic restrictions on phrasing prediction (see Ghini 1993a, 1993b for Italian, Prieto 2005 for Catalan, Sandalo & Truckenbrodt 2002 for Brazilian Portuguese, Selkirk 2000, 2005 for English, and Hellmuth forthcoming for Egyptian Arabic; see also Elordieta, Frota, Prieto & Vigário 2003, D’Imperio, Elordieta, Frota, Prieto & Vigário 2005).