Acceso usuarios
, Jon Franco Elorza
Most of the literature on null objects in Romance languages such as Rizzi (1986), Authier (1988) and related work, has focused on determinig the categorial status of these empty constituents. There are two competing analyses in this regard wich read as follows: (a) null objects are variables bound by abstract operators; (b) null objects are instances of pros. Thus, Raposo (1986) an Authier (1988) subscribe to the former in regard to Euorpean Portuguese, Spanish and French respectively, whereas Rizzi (1986) for Italian and Farrel (1990) and Maia (1991) for Brazilian Portuguese subscrihe to the second one. This paper shows that the two hypotheses are not mutually exclusive, if we consider, unlike the investigations above, that the possibility of having both types of empty categories in a single language depends on the existence in that language of the two basic kinds of semantic interpretation that null objects may have, i.e. arbitrary and referential interpretation. Interestingly, Basque Spanish (the variety of spoken in the Basque Country) provides evidence for the need of a distinction of null categories based on their interpretation since this variety makes a syntactically and semantically defined cut between arbitrary and referential null objects, as far as their possibilities of ocurrence are concerned.
Using a series of tests developed within the Principles and Parameters framework for the classification of empty categories, we first demontrate that the asymmetries in syntactic hehavior between arbitrary null objects and referential ones are due to their different status as empty categories. Specifically, we claim that, in Basque Spanish, arbitrary null objects behave like variables bound by an abstract operator, whereas referential null objects behave like pros. Second, we provide an analysis that accounts for the licensing and semantic identification of these empty categories, wich are the two necessary conditions to fulfill the Theta Criterion and the Projection Principle.
