@InProceedings{Crysmann:1999,
AUTHOR = {Crysmann, Berthold},
TITLE = {Licensing Proclisis in European Portuguese},
YEAR = {1999},
BOOKTITLE = {Empirical Issues in Formal Syntax and Semantics. Selected papers from the Colloque de Syntaxe et de Sémantique de Paris (CSSP'97), October 16-18},
PAGES = {255-276},
EDITOR = {Corblin, F. and Marandin, J.-M. and Dobrovie-Sorin, C.},
ADDRESS = {Paris, France},
PUBLISHER = {Thesus},
ABSTRACT = {In this paper, I will address the interaction between quantification and linearisation in the grammar of European Portuguese (EP) clitic placement. In particular, I will suggest that a licensing relation holds between a subset of the natural language quantifiers identified in Generalised Quantifier Theory (GQT) and the order in which the clitic and its host must surface. More specifically, I will argue that the class of proclisis licensors is best described in semantic terms (i.e. in terms of logical entailment), whereas the relation between proclisis licensor and licensee should be conceived of as entirely surface-syntactic. It will be shown that approches which mediate the licensing relation by means of syntactic movement (Barbosa, 1996; Duarte, 1983; Madeira, 1992, cf. e.g.) are faced with both motivational and empirical problems. Instead, I claim that surface-syntactic linearisation constraints will relate clitic placement directly to a class of lexical items, which in turn is defined on the basis of semantic properties. Therefore, an integrated model of syntax and semantics is called for which builds on highly articulate lexical information. The analysis will, thus, be carried out in the framework of Head-driven Phrase Structure Grammar (HPSG) (Pollard and Sag, 1987; 1994), using multiple inheritance type hierarchies and linearisation constraints. The paper is organised as follows: in the first section, I shall briefly describe the basic empirical observations regarding EP proclisis. In section two, I shall review the empirical problems faced by previous (mostly syntactic) approaches. Section three provides the details of the proposal, starting with a semantic typology of proclisis licensors. In the remainder of the section, I outline the surface-syntactic constraints which define the phenogrammatical relation between licensor and licensee, analysing EP proclisis in an essentially similar way to English negative polarity items (NPI) (cf. Ladusaw, 1996, and reference cited there).}
}
|
