Contenu du sommaire

Abstract
On the basis of a critical review of previous research, this article shows that the counterfactual use of the “imparfait” — the classical example since Guillaume is Un instant après, le train déraillait “An instant later, the train would have derailed” — is not simply a variant of the so called “imparfait de rupture” (i.e. cases where the “imparfait” is in competition with the “passé simple”, e.g. Six mois plus tard, il se mariait, “Six months later, he married”) previously analyzed in Berthonneau & Kleiber (1999, 2000, 2003).In the case of the “imparfait de rupture”, the fronted adjunct can only be temporal, and must indicate posteriority. In the case of the counterfactual “imparfait”, it can point to an earlier moment (Tu arrives au bon moment. Un peu plus tôt, ce n'était pas prêt “You're just on time. A little earlier, it wouldn't have been ready”), and is not necessarily temporel. Beyond the classical case of Sans vous, je m'ennuyais “Without you, I would have been bored”, there are also less well known cases of changes in quantity (Deux kilos de moins, je rentrais dans ma robe “[Had I weighed] two kilos less, I would have been able to fit in my dress”, Moins cher, c'était du fil de fer “[Had it been] less expensive, it would have had to be wire”, etc.).The counterfactual reading of the “imparfait” is crucially dependent on the interaction between the “imparfait” and the fronted adjunct (this differenciates it from the “conditionnel passé” which is usually used to gloss it). Three sub-cases are analyzed. The first, Une minute plus tard, … “One minute later, …”, involves a change in duration between the factual situation and the counterfactual situation. The second, Une minute après, … “One minute later, …”, no longer takes as its reference point the factual situation, but the event referred to by après. Between these two cases, there is the type Sans vous, je m'ennuyais, or Deux kilos de moins, je rentrais dans ma robe, where there is no longer the idea of a failure of occurrence of something imminent. As in the first sub-case, this reading is based on a homogeneous variation in quantity, except that the quantity involved is no longer temporal.In all cases, the fronted adjunct makes an anterior situation salient. The counterfactual imperfect thus perfectly fits the anaphoric meronymic analysis of the “imparfait” proposed in Berthonneau & Kleiber (1993), though the detailed analysis of the different uses is initially carried out independently of the latter hypothesis.

Abstract
The aim of this paper is to put forward a unitary analysis of French predicate adjectives in the framework of the unaccusative hypothesis. If it is assumed that a predicate adjective is always syntactically connected to the direct internal argument of a verb, it is possible to explain why French predicate adjectives modify either the direct object of a transitive verb or the subject of unaccusative predicates such as intrinsically unaccusative verbs, copulative structures, passives or pronominals. This hypothesis predicts moreover that adjectives cannot be predicated of subjects of unergative intransitive verbs or subjects of transitive verbs because these subjects have to be analysed as external arguments of verbs. Finally, a comparison is made between French predicate adjectives and English resultative constructions that have been claimed to be sensitive to the unaccusative / unergative distinction.

Abstract
This paper deals with the type of anaphoric process associated with the adnominal adjective pareil in sentences such as Je ne m'attendais pas à un scandale pareil. It aims at specifying the kind of identity relationship expressed by this adjective. The data suggest that pareil does not entail a strict identity with the antecedent but rather involves a reference to the type associated with the antecedent. The adjective pareil also turns out to involve appraisal, as it indicates that the subject considers the referent as outstanding. It is also concluded that the prenominal vs. postnominal position of the adjective has a clear semantic impact. When the adjective is used prenominally (un pareil N), there is a focus on the categorization of the antecedent, hence the predicative content of the N is highlighted, and the article can be omitted. When the adjective is used postnominally (un N pareil) however, the focus is on the particular characteristics of the antecedent.

Abstract
Normative grammar defines the uses of the French expressions rien moins que and rien de moins que. The first has a negative meaning: Pierre n'est rien moins qu'intelligent means that he is not intelligent at all ; while the second has a positive meaning: Pierre n'est rien de moins qu'intelligent means that he is extremely intelligent. However, in practice, this distinction is far from being clear: rien moins que often has a positive orientation. Are the speakers mistaken? The problem with a normative standpoint is that it prevents one from questioning the rules. A linguistic analysis would permit understanding and explaining these variations. That is why we carried out a corpus study of these expressions, on the basis of which we provide a coherent description in conformity with linguistic reality and an explanation of their uses.

Abstract
Group adjectives are often considered, at least when accompanying deverbal nouns, as true substitutes for noun phrases, which means that they can bear a thêta role, as noun phrases do. The first aim of the article is to argue that that is not the case. The main argument is that these adjectives occur exclusively in one type of nominals, namely those of which the head noun, notwithstanding the fact that it is a deverbal noun, has no argumental structure anymore. In the other type, where the head noun retains the argumental structure of the verb, group adjectives are excluded. The second aim is to understand why it is that group adjectives, preferably to other denominal adjectives, are so often regarded as equal to noun phrases in semantic value as well as in syntactic function. The answer could be that they are particular in having almost (but not exactly) the same referential strength as that of noun phrases.