On scalar readings of French propre 'own'

摘要

The goal of this paper is to account for scalarity effects typically arising in the presence of French propre 'own'. I argue for the presence of an implicit, focus-sensitive operator E akin to overt 'even', which has previously been proposed to combine with certain NPIs and minimizers. My argument is based on empirical and theoretical considerations: the E operator hypothesis correctly predicts that the domain for the scalarity effect does not depend on the position of propre and that the simultaneous presence of overt focus particles gives rise to intervention effects; this hypothesis is moreover economical since it makes it possible to provide a single lexical entry for propre under all its readings and to analyze all cases of scalarity effects by using a single mechanism, namely E. Furthermore, I argue that the association between E and the focused possessive DP containing propre is pragmatic. It is because propre is a maximizer of possessive relation-i.e. it characterizes the possessive relation it modifies as most specific-that it is relevant in the association with E of the focused DP containing it: the possessum is ranked higher than its alternatives on the scale of relational specificity introduced by propre, and this scale pragmatically correlates with the scale of unexpectedness required by the semantics of E. In sum, the case of propre supports the existence of E on new empirical ground and thus clarifies the pragmatic conditions giving rise to it.