Lingvisticæ Investigationes - Volume 33, Issue 2, 2010

The three inversion types of French are put in a synchronic comparative perspective with respect to Dutch and Italian. One, complex inversion, is unique. Pronominal inversion not only shares the syntactic derivation with Dutch, but also the conditions of use, with one lexical exception, anteposed aussi. Nominal inversion appears to not only share the syntactic derivation with Italian, but also the focus property of some non-interrogative inverted structures. On a more abstract general level, we argue in favour of an hypothesis according to which the actual use of both inversion types in French form proper subsets of their actual uses in Dutch and Italian respectively.

In this paper, we propose a contrastive corpus study of French car and parce que and Italian ché and perché, meaning “because”. Our goal is to assess the importance of subjectivity in grammaticalization in general, and in the renewal of causal conjunctions in particular. The evolution of these two pairs of conjunctions is quite similar: on the one hand, the most grammaticalized items of each pair, car and ché, are also the most intersubjective; they tend not to change meaning and to fall into disuse. On the other, the less grammaticalized items, parce que and perché, are also the less intersubjective, but gradually change meaning, possibly acquiring (more) intersubjective uses, and seem to have become the default causal conjunctions in Modern French and Italian — more clearly so in the spoken than in the formal, written varieties. There thus seems to be a link between the degree of grammaticalization and that of (inter)subjectification.

In this chapter, we examine control configurations in French gérondif clauses in which the referential properties of the implicit subject argument (PRO) have to be determined by contextual (linguistic and extra-linguistic) factors. We compare the control facts with parallel Norwegian translations. As a Germanic language, Norwegian does not have equivalent gérondif constructions, hence leaving it to the translator to resolve the control facts left unspecified in the French original. Traditionally, it is held that the implicit subject of the gérondif has to be the same as the subject of the matrix clause. It will be shown that this generalisation does not hold. In fact, PRO may have an explicit controller that is either a subject, an object or a prepositional complement of the matrix. The controller may also have an arbitrary reference. As a result, the resolution of PRO in the traditional sense will be regulated by control theory and discussed with respect to obligatory and non-obligatory control contrasts (OC vs. NOC) in generative settings. The data is taken from the Oslo Multilingual Corpus (OMC). This database enables us to compare parallel texts from Norwegian, English, German and French. We focus here primarily on Norwegian translations of French originals.

This paper is an exploration of similarities and differences concerning absolute constructions in French, German and Norwegian. In the first part, I have examined a more general question raised by these constructions: the connections between these types of absolute constructions and the matrix subject. I have shown that the means by which the absolute constructions are related to the subject can be morphosyntactic, semantic and pragmatic. The second part contains a purely contrastive analysis. Two issues have been examined: on the one hand, the absolute constructions and their congruent and non-congruent correspondences, on the other, the use of determiners. Essentially, French is different from the two Germanic languages, but similarities also exist between French and German, which are the center of a European Sprachbund.

The French spatial preposition sur (‘on’) has recently developed new spatial usages. It has evolved from expressing a spatial configuration of superposition to also expressing extent and even a location which is merely relational. The aim of our study is to provide evidence for the hypothesis of the grammaticalization of sur. This task is carried out by comparing these new spatial usages of sur with their Dutch translations. Eighteen attestated cases of sur were selected from a unilingual French corpus, which were then translated by ten native speakers of Dutch. The analysis of these translations showed, first of all, that the new uses of sur are rendered by a wide range of Dutch prepositions. Second, when expressing a location which is merely relational, i.e. when sur is used as a synonym for à (‘to’), the only translation proposed by the native speakers of Dutch is in (‘in’). It comes as no surprise that this preposition is also the most frequent translation of à, which is the French desemantized preposition par excellence.

In this paper I try to find the ways of expression of Inalienable possession (IA) in Bulgarian, in comparison with French. I discuss the morphology and the interpretation of the determiners of these two languages, likely to function as A-bound morphemes according to Guéron’s (1985, 2005) binding hypothesis concerning the IA construal, based on the presence of phi-features on the determiners. I use an additional condition for a determiner to be A-bound (which comes from Vergnaud & Zubizarreta 1992), namely the possibility for a determiner to have expletive uses. I claim that only the French definite article and the Bulgarian null determiner are subject to binding, the Bulgarian definite article being excluded due to its strong referential power.

This article reassesses N. Ruwet’s (1988) attribution of the label “unaccusative” to weather verbs across the board. Through a cross-linguistic analysis of NP expansions (in French, English and Russian), we argue that only metaphorically used French weather verbs and English weather verbs in personal constructions, followed by a directional particle or a PP, behave as genuine unaccusatives and can have their NP expansions qualify as internal arguments. All other NPs of weather verbs function rather as adverbial adjuncts, modifying the meteorological predicate. Moreover, our study shows that unlike a widely accepted belief, the class of weather verbs is far from being homogeneous.

In this paper, we intend to show that grammaticalization can effectively be used as a parameter in language comparison by proposing a corpus-based account of the difference in distribution between demonstratives and definite articles in French and Dutch. Taking cases of literal but non-equivalent translation as a starting point, our study revealed that French demonstratives are very often translated by definite articles in Dutch, indicating a different semantico-grammatical function of demonstratives and definite articles in both languages. We propose to account for this bleaching of the French demonstrative by hypothesizing that Dutch demonstrative determiners have a much stronger ‘defining function’ than their French equivalents, and that the latter are in fact evolving towards definite articles, which would be an instantiation of the cyclical evolution from demonstrative to article, also found in the development from Latin to French.

In this article we zoom in on the Verbs of Instrument of Communication (B. Levin 1993) in English and French and address a number of interesting issues. We examine the structural possibilities of the old and new Verbs of Instrument of Communication in comparison with one another and across the two languages. First, despite the creative uses of these verbs, some structural differences between the languages still prevail in the old and new Verbs of Instrument of Communication. For instance, while English allows non-pronominal realisation of object and recipient in the so called double-object construction, this is not the case for French. Secondly, it can be observed that some verbs are more frequent in one construction than in another. For instance, while ‘fax someone something’ does occur with significant frequency in English and French, ‘telephone someone something’ occurs much less frequently. The observed cross-linguistic syntactic differences are explained in terms of a different conceptualisation of the communicative event where the structural possibilities correspond to different semantic frames.

From the 17th to the 20th century, the interpretation of savoir comme S (to know how/how much S) has evolved: these constructions had first an interrogative meaning, and then became mainly exclamative. This paper aims at describe one of the most important contextual modifications this evolution is correlated with, namely changes concerning savoir (to know). Semantic factors seem to be very important to identify the meaning, so one can conclude that a single syntactic structure is involved in those sentences. The study is based upon examples coming from Frantext corpora.