Contenu du sommaire : Indéfinis, définis et expression de la partition, sous la direction de Catherine Schnedecker et Anne Theissen

The aim of this contribution is to show that the classical response in "sous-espèce" terms brought to the problem of the conversion du N de matière → un N de matière modifié is only partially satisfactory. This solution meets different types of difficulties, of which the principal one, passed unnoticed hitherto, concerns interpretation : the interpretation under type is not the possible only one. Besides the taxinomie interpretation and the particular one, it is necessary to distinguish a third interpretation, that we named lecture individualisante, of which we will analyze the reification model.

This paper deals with the linguistic explicitation of the relations of partition existing among the temporal referents. The referents of the names année/mois/ jour... are indeed connected by relations of order and inclusion (semaine can be described as a part of mois, just like volant is a part of voiture). Therefore, those names should satisfy the test-frame N de N. However, if it is possible to find the structure le volant de la voiture, it is odd to say les semaines du mois or en juin de l'année. The aim of this article is to put in evidence the lexical features of the temporal nouns and some constraints governing the linguistic translation of the whole-and-part relationship between the temporal referents.

The aim of this paper is to analyze the semantic properties of the quantifier beaucoup de in subject NPs (Beaucoup de A sont В) and in il y a- sentences (II y a beaucoup de A qui sont B). The analysis accounts for four readings of beaucoup de: existential (cardinal, "weak") reading, proportional reading with respect to the set |A| ("strong" reading), proportional reading with respect to the set |B| , and a reading which is called predicative quantificational. The hypothesis defended is that beaucoup de contains an indefinite component on all its readings. Consequently, it always satisfies the criteria for qualifying as a weak quantifier. The arguments in favour of this hypothesis are based on a distinction between an extensional approach to the "weak" reading according to which the item beaucoup of the quantifier beaucoup de is a cardinal predicate (Ils sont beaucoup) and an intensional approach according to which the item beaucoup is treated as a quantificational predicate (II y en a beaucoup).

The constructions un des N and un N enter in competition when the simple indefinite phrase un N is employed partitively as the complex phrase un des N. Our contribution aims to resolve the questions that put this parallel. In a way more precise, we will try to measure the semantic and syntactic resemblances and differences of the two partitive formulas, to evoke certain of their employment constraints and to put in value the originality of the partitive way borrowed by each one.

This article tries to put forward on the basis of historical consideration that the pronoun les uns is a pronoun with whole share, quite fixed, original from the point of view of its internal plural. In one second part, one studies the referential behavior frame of the pronoun, under angle of his relative (in)definitude and his partitive nature, which one considers certain consequences in the discursive plan

The goal of the paper is to show that morphologically complex pronouns such as quelques-uns, celui and chacun are in fact determiners within a DP that contains an empty noun. This claim is supported by several arguments. First, several syntactic phenomena, such as the use of dislocated nouns introduced by de or the combination with the quantitative pronoun en, suggest that pronouns such as quelques-uns, celui and chacun are in fact determiners within a DP in which the noun can be dislocated or can be replaced by the pronoun en. Second, to respect Kayne's (1994) Antisymmetry theory, for the partitive construction (the first part of which can be one of the pronouns quelques-uns, celui or chacun) a structure is proposed that necessarily contains an (empty) noun, which can account for agreement facts. Finally, the noun position containing an empty category is also required to account for restrictions on nounless DPs. It is proposed that the empty category pro is only licensed in the nominal position by determiners or adjectives with a partitive meaning and only if it can be interpreted as a topic.