L'indexicalité du sens et l'opposition "en intension"/"en extension"
Institution:
Paris 8Disciplines:
Directors:
Abstract EN:
This work aims at founding a theory - "indexicality of meaning" - that postulates a same nature for meaning and referent. Meaning and referent are based on "relations" we have to objects. Conversely, the notion of "object" -which generalizes the notion of referent-is defined as the focal point, deprived of + substance ;, of a potential infinity of such relations. Being so, the object is perceived and conceptualized as such without the means of descriptive and/or objective properties, by way of "direct conformation" to the permitted relations. This phenomenological thesis, directly inspired from a criticism of referentialism, leads to substituting "direct conformation" for classification, which is made possible by virtue of the notions "in intension" and "in extension". These notions allow for a new approach of the "evolving referent" problem. They also help characterizing metonymy, metaphor, and some uses of the french pronoun "il'. The sense of this pronoun is defined, by way of a notion of "individuation". Also, the noun and preposition classes are characterized, without the use of schematicity and compositionality. It appears then that the noun inherently underdetermines the object. The proper name class also is given a definition that accounts for its main properties. Finally, the theory is completed by the integration of homonymy. I two applications of the theory are proposed : the definition of the sense of the french prepositions a and de, and the characterization of the two uses of the french verb commencer (analogous to : to begin doing something vs. To begin something). It appears that the methodological requirement of linguistic motivation accounts for the accuracy of the results.
Abstract FR:
L'objectif de ce travail est de fonder une theorie semantique - 1' "indexicalite du sens" - a partir du postulat selon lequel le sens et le referent sont de la meme nature, definissable par une notion de "rapport" - ou de "propriete extrinseque". Il est alors possible de definir la notion d' "objet" - qui generalise celle de referent - comme le point focal, vide de toute << substance >>, d'une infinite potentielle de rapports. Ainsi, l'objet est percu et concu en tant que tel sans la mediation de proprietes descriptives et/ou objectives, mais par "homologation directe" des rapports qu'il permet. Cette these phenomenologique, directement inspiree d'une critique du referentialisme, amene a substituer 1'"homologation directe" a la notion de classification, ce qui est rendu possible par une definition des notions de mise "en intension" et de mise "en extension". Ces notions permettent une approche originale du probleme des "referents evolutifs". Elles permettent egalement de caracteriser la metonymie, la synecdoque et la metaphore, ainsi que certains emplois problematiques du pronom il la signification de ce pronom est egalement definie, a travers une notion d' "individuation". D'autre part, les categories lexicales du nom et de la preposition sont caracterisees, sans faire appel a la schematicite ni a la compositionalite. Il apparait alors que le nom est fondamentalement une sous-determination de l'objet. La categorie du nom propre recoit egalement une definition expliquant ses principales proprietes. Enfin, la theorisation de l'homonymie clot la construction et la consolidation de la theorie. Deux parties applicatives permettent de confirmer l'ensemble des hypotheses : la determination des significations des prepositions a et de, et la caracterisation de la double construction - transitive directe et infinitive - du verbe commencer. C'est une exigence methodologique de motivation linguistique qui est responsable de la precision des resultats.