Contenu du sommaire : Sélection et sémantique. Classes d'objets, compléments appropriés, compléments analysables, sous la direction de Jacqueline Giry-Schneider

This article aims at clarifying the concept of « object classes » in natural languages and its usefulness in describing verbal predicates. Syntactic features such as human, concrete, abstract are not accurate enough to account for the polysemy of the operators. Object classes can be put to other uses, some of which are outlined here.

This article purports to illustrate the concept of class of objects, derived by Gaston Gross from computer science. About 1500 nouns share the common characteristic of being directs objects to the verb lire (to read). According to what verbs can accept what objects, it is possible to subdivide this wide class into many semantically homogeneous subclasses. These are classes of objects, they are defined by the type of syntactic relations, that they have with verbs called appropriate operators.

Some verb-noun combinations require specific studies. Here, we shall present a set of such combinations which are composed of a verb admitting sentential complements and a predicative noun. These combinations which we shall call appropriate do not share the properties generally observed. Indeed, unlike the majority of other groups of verb and predicative noun, some of these combinations do not allow the application of general transformations. Also, in some cases, the predicative noun of these combinations can be dropped from a sentence without producing any loss of information. Using these two criteria, we shall give a precise definition of these appropriate combinations.

The aim of this paper is to account for the relationship between NP complements and infinitival complements for a particular class of verbs, that of so-called (semi-auxiliaries. The data examined belong to two Romance languages, viz. French and Spanish. For all verbs, both in French and Spanish, a definitional property is that they select an infinitival complement but never a that -clause (que P). However, whereas in French only few verbs also subcategorize for a NP complement, in Spanish the latter property holds for more than half of the verbs. This empirical fact is not only worth noting in a typology of Romance languages (section 1), but it also raises the theoretical question of how the universal category of auxiliaries ougth to be defined (section 2).

The object of this paper is to explore the syntactic structures of the English report verbs and to compare them with the French verb structures. Among these structures we single out for special attention the verbs to say and to tell, as well as the particular noun complements they accept. It will be shown that, although they share the same defining structure NO V That S to N2, they do not accept the same noun complements. Different derivations will be proposed to analyze the various noun complements which are allowed after report verbs. The table shown in the appendix illustrates the way lists of appropriate nouns for specific verbs are made out. In the second part of this paper, the comparison between appropriate nouns of the French verb dire and the two corresponding English verbs to say and to tell brings out the differences and similarities of the two systems. The linguistic phenomena are very similar, e.g. the use of support verbs and appropriate nouns. Each verb however has specific uses —whether in English or in French. A possible answer to these specific use is the careful working-out of a lexicon-grammar.

We study here the noun complements of French speech verbs (dire-say, proclamer-proclaim, etc.), and in particular the appropriate noun complements. We have tried to shaw that these complements are related syntactically to these containing the nominalization of the verb in a complement clause.