Contenu du sommaire : Grammaticalisation et lexicalisation : la formation d'expressions complexes

The goal of this contribution is to determine the phenomena involved in the formation of complex prepositions. We first discuss whether complex prepositions are to be considered a subclass of prepositions. A diachronic corpus study of a sample of such structures in Old and Middle French (a chief de lit. ‘at (the) head of', a hore de ‘at the time of', por l'amor de ‘for the love of', au lieu de ‘instead of', au-dessus de ‘above') then enables us to show that lexicalization plays a major role in the first stages of their development. Grammaticalization appears only for a small number of already formed complex prepositions, which thereby acquire a more general meaning, with reduced phonological weight (e.g. du côté de > du côté > côté evolving from its original ‘lexical' meaning ‘on/to the side of' to its modern meaning ‘about', ‘talking of').

This article intends to demonstrate that verbal phrases such as prendre garde, estre en garde, avoir garde, etc. (lit. ‘take, be in, have care') in French might not be as rigid in terms of syntactic pattern as one could think, especially in Old French (11th – 13th centuries). This study aims to collect as many occurrences as possible of the different structures of verbal phrases using the noun garde in the structure : [verb + (preposition) + garde]. Several points are evaluated : 1/ what is the consequence, if any, of the lack of determiner in the verbal structure ? ; 2/ how does it affect the semantic evolution of the whole structure ? We show in this article that, in Old French, many syntactic verbal structures which apparently could be considered as verb phrases because of the use of a bare noun do in fact develop a specific meaning (general versus particular), but should not be seen as syntactic phrases.

In this article, I challenge the generally well-accepted idea that constructions with support verbs in Medieval French are “frozen expressions”. My objective is to show that these support verb constructions in Middle French (14th-15th centuries) are, in fact, cases of free syntax, even if it is observed that the nominal predicate appearing in these structures is often a bare noun. Only in some cases can a support verb construction appear to be frozen : to be precise, when the article preceding the noun cannot vary, and when the noun's meaning is opaque.

Only pour sûr (‘for sure') is used in Present-Day French as a modal adverbial (although essentially in regional or popular use), but in Medieval French and even later, pour vérité, pour voir, pour vrai and pour certain (lit. ‘for truth', ‘for true' and ‘for certain') were likewise used to indicate the speaker's attitude regarding the truth of his proposition. While most modal adverbials originate in a clause-internal adverbial, the data suggest that the adverbials studied here originate in prepositional phrases which originally functioned as object predicatives (tenir/dire pour vrai/vérité... que P ‘to hold/give for true/truth... that P'). Most mechanisms usually involved in grammaticalization occurred in the shift from object predicatives to modal adverbials (reanalysis, decategorialization, generalization, subjectification) and, in some respects, modal adverbials do assume a grammatical function ; nevertheless, if we are to refer to Traugott's most recent works, we cannot consider this evolution to be, strictly speaking, a grammaticalization.

This paper focuses on modal adverbials in en/à la vérité (lit. « in/to the truth », meaning « to say the truth »). This variability (preposition en « in » or à « to ») is interesting because it does not appear for other adverbials with a seemingly parallel construction : en toute franchise, en un mot ( « to be frank », « in a word »), etc. which do not present the same variability (*à toute franchise, etc.). The authors examine the emergence of en/à la vérité in the 15th to 17th centuries, and show that there is in part coexistence of those adverbials' various uses ; in particular, textual and enunciative uses seem to appear simultaneously.

This article analyzes in detail the emergence of topic markers in Clasical French : en/pour ce qui concerne, en/pour ce qui regarde, et en/pour ce qui touche (lit. « in/for what concerns, looks, touches », all meaning « concerning »). These constructions followed the usual evolutionary path, from linked adverbial to thematic function, then to topic function. The author examines in detail the contexts in which these constructions undergo reanalysis, and focuses on two criteria, morpho-syntax and topic/comment. It can be shown that the opposition between en and pour is meaningful : pour is found indifferently for all uses, while those containing en are usually restricted to intrapredicative postverbal uses.

The aim of this paper is to give an account of the emergence and evolution of the French « topicalization marker » à propos de (‘about'), as well as two other phrases : à propos and à ce propos. This study is corpus-based and deals with texts in Middle and Pre-Classical French (14th to 16th centuries). As we try to refine the chronology of the evolution, we show that all three expressions have always displayed both postverbal and preverbal (even initial) positions. While the expressions in postverbal position have always functioned as complements, strongly linked to the verb, those in initial position go through an important change : from verbal complement to topicalization or enunciative markers. As we explain in our paper, this evolution roughly corresponds to the three-staged movement described by Traugott in the global framework of grammaticalization, namely from a more objective to a more subjective value, or in other words, from a propositional to an expressive function, through a textual one.