Contenu du sommaire : Aux sources de la polysémie nominale, sous la direction de Pierre Cadiot et Benoît Habert

The phenomenon of systematic polysemy offers a fruitful domain for examining the theoretical differences between lexicological and lexicographic approaches to description. We consider here the process that provides for systematic conversion of count to mass nouns in English (a chicken /chicken, an oak / oak etc.). From the point of view of lexicology, we argue, standard syntactic and pragmatic tests suggest the phenomenon should be described by means of a single unindividuated transfer function that does not distinguish between interpretations (rabbit = « meat » vs. « fur »). From the point of view of lexicography, however, these more precise interpretations are made part of explicit description via the inclusion of semantic « licences », a mechanism distinct from lexical rules.

Lexical signification is often conceived, at least partially, in terms of a description or, in our terminology, in terms of intrinsic properties (IP). In this article we explore an opposite hypothesis that words (most notably nouns) invoke directly those types of relations : 1 . which a given word allows the loculor to establish with various referents. 2 . which these referents have (or arc perceived as having) with their environment. The meaning itself of nouns must be identified with these relations or extrinsic properties (EP). In this way, polysemic, metonymic and metaphoric usages shed their marginal status, becoming the basis of a lexical signification which purely denominative usages conversely tend to obscure.

Starting from the « indexical » conception of words, this work tries to lay some theoretical foundations taking accoung of polysemy and its boundaries. In the first part, the necessary notions — including a new meaning of in intension/in extension — are defined in the nominal domain. The second part makes use of these notions for the analysis of traditionnally problematic uses of the french pronoun il.

Towards an archeology of figurative meaning Cognitive semantics postulates that specific mental mechanisms belonging to the human imagination, namely metaphorical and metonymical mappings, are involved in semantic changes. Nevertheless, a thorough examination reveals the weakness of this theory. I therefore advocate a radically different analysis which allows us to do without these postulated cognitive mechanisms. Systematizing Meillet's observations about the role of « discontinuity in transmission » in semantic change, I propose, for each case studied, to look for what circumstances in collective experience can make the discrepancy between the new and the former meaning imperceptible to the speakers themselves. I argue that a semantic change at its decisive moment must most often perform in a covert way, without its instigators' knowledge. The efficiency and heuristic fruitfulness of this method is illustrated in this paper by the study of certain Latin polysemous words expressing not only a physical bond but also a juridical obligation (obligare, obligatio, nectere, obstringere, solvere, absolvere). Cognitive semantics postulates that specific mental mechanisms belonging to the human imagination, namely metaphorical and metonymical mappings, are involved in semantic changes. Nevertheless, a thorough examination reveals the weakness of this theory. I therefore advocate a radically different analysis which allows us to do without these postulated cognitive mechanisms. Systematizing Meillet's observations about the role of « discontinuity in transmission » in semantic change, I propose, for each case studied, to look for what circumstances in collective experience can make the discrepancy between the new and the former meaning imperceptible to the speakers themselves. I argue that a semantic change at its decisive moment must most often perform in a covert way, without its instigators' knowledge. The efficiency and heuristic fruitfulness of this method is illustrated in this paper by the study of certain Latin polysemous words expressing not only a physical bond but also a juridical obligation (obligare, obligatio, nectere, obstringere, solvere, absolvere).

The prominence of the noun in the discussion of language results from a privilege accorded to the perceived tangibility of its referent. However, the physical properties of a referent are an insufficient basis for a semantic analysis of the word which designates it. An examination of the polysemic noun « key » in French and English, demonstrates this deficiency, providing an opportunity to develop a semantic account of the distribution of its use in the two languages.

Multiple descriptions and object categorization Little research has yet been done, except in theory of perception, on the multiple ways physical objects are represented when they are used. By showing how the same object affords several various descriptions that capture diverse components of its use, Marr shed light on the semantic role of intermediary representations. This analysis shows when an object is used, visual properties are represented in a specific way without requiring object recognition and object naming.

La sémantique lexicale est d'abord inférentielle Lexical Semantics is concerned with the meaning of words ; but how to express a meaning ? In words ? With « atoms of meaning » ? With logical predicates ? With mathematical functions ? We discuss the weaknesses of these solutions, try to identify their causes and conclude that the problem is basically ill-defined. Actually, what matters are the inferences which are warranted by the use of a word in its context ; the existence of a referent in some universe of discourse is merely a possible by-product of these inferences. We finally explain that the inferential behaviour obeys general principles that can be represented in a logic, provided it is a non-monotonic one.

Semantic « fixism » is the view according to which non-indexical words and phrases possess Fregean senses conventionally, that is, in a fixed and context-independent manner. I show thai fixism cannol accounl for systematic polysemy in a satisfactory way. To account for systematic polysemy, we must acknowledge that the sense of an expres sion (its condition of satisfaction) is generated rather than merely selected. In the third part of the paper I discuss two models of sense generation. According to « segregalio nism », the sense of a word is generated through the interaction of its context- independent meaning with the context-independent meanings of the other words in the sentence. According to « contextualism », the sense of an expression is generated through the interaction of that expression's meaning with the context, both linguistic and extralinguistic. Semantic « fixism » is the view according to which non-indexical words and phrases possess Fregean senses conventionally, that is, in a fixed and context-independent manner. I show that fixism cannot account for systematic polysemy in a satisfactory way. To account for systematic polysemy, we must acknowledge that the sense of an expression (its condition of satisfaction) is generated rather than merely selected. In the third part of the paper I discuss two models of sense generation. According to « segregationism », the sense of a word is generated through the interaction of its context- independent meaning with the context-independent meanings of the other words in the sentence. According to « contextualism », the sense of an expression is generated through the interaction of that expression's meaning with the context, both linguistic and extralinguistic.