Corpus Taurinense (CT) is the POS tagged version of ItalAnt Corpus, an electronic corpus of Old Italian texts (between 1251 and 1300). In this article we aim to describe the approach followed in CT for the annotation of multiword units (MWU). MWU in our work is a set of two or more graphic words which receive (also) an overall POS tagging because this set of words is in paradigmatic relation with one word lexical unit with the same POS.Our POS tagging confirms that most of the Modern Italian compound conjunctions at that time were not lexicalised. The order of the components is already the Modern Italian order but they can still be interrupted by occasional elements.

This article aims to investigate experimentally the links between prosody and the various domains of linguistics (syntax, semantics, pragmatics). Since the function of prosody is to convey the interpretation of the meaning the speaker wants to give to his utterance, the method that is proposed is to define linguistic models that can be quantified in such a way as to both yield an interpretive strategy of the linguistic contents and to predict a melodic structuring of the lexical framework. Experimental evidence shows links between on the one hand constraints related to the text and the context, and on the other the distribution of models and melodical indices, and allows us to define prosody and more in particular melody, as a subjective locus, where whatever belongs to the linguistic realm enters the realm of pragmatics.

Based on recent studies that put into question the concepts of prototype and typicality, this study treats the processes of categorization by sufferers from Alzheimer's disease, in the light of two themes: clotting of the prototype into stereotype and distinction between, on the one hand, the representations of common objects developed by an individual in a very early stage of his development and, on the other hand, the socialized attainments acquired subsequently. We analyse the processes of categorization in five groups of subjects (English-speaking students, French-speaking students, sufferers from Alzheimer's disease and witnesses) paired by age, sex and level of studies. Our aim is to study two effects on the probability of categorical inclusion: the effect of constraints and categories (putting in contrast four types of categories - two "natural" ones and two "artefactual" or cultural ones), and the effect of typicality of examples, in the different groups of subjects. By this analysis, we aim to demonstrate specific behaviours shown by the group of sufferers.

The majority of lexical databases and computerized thesauri are organised on the principle of a system of lexical 'heritage' based on taxonomic relationships (IS_A). This relationship is perceived as the channel through which the lexical information is passed. We claim, however, that the transfer of information in a lexical thesaurus can also take place through other kinds of relationships. In this respect, we analyse the 'heritage mechanism' based on the mereonymic relationship COMPOSED_OF. The main object of this paper is to characterise the framework of a lexical system based on a system of mereonymic relationships, i.e. a system which will allow for a whole to inherit the information of its parts. Further, we want to demonstrate that this type os heritage allows for a model of metonymic interpretation of polysemic nouns.

Many more dictionaries are being updated than are written from scratch. Taylor-made computational linguistic support for lexicographers should thus not only provide corpus-derived data, but, along with this, a comparison with the information given in the targeted dictionary. We are developing a system for German, which provides this kind of comparison, for a set a macro- and microstructural data. We report about the main lexicographic and computational aspects of the approach.

Reference to a well-known proverb or phrase by altering one of its components is a widespread phenomenon in the British and American press. Since such modifications can impede a non-native speaker's understanding of newspaper or magazine articles, a system that could identify them and refer learners of English as a second language to the original wording of such expressions might be useful in the conception of an on-line comprehension assistant. Using a data base of 10 500 titles from an American news magazine, we analyze the various types of modifications that come into play in the use of shared cultural references. Through the comparison of our data base with 800 English proverbs, we test various ways in which such modifications can be automatically detected.

The free and online encyclopaedia project Wikipedia has become in less than six years one of the most prominent commons-based peer-production example. The way the project works and evolves is now at stake for academics eager to explore auto-organized structures. Although many studies have been led on the connections between contributors, the linguistic properties of Wikipedia productions remain almost unexplored. In this article, we focus on the way sciences are represented within the project and examine the general and epistemic lexical characteristics of the articles thanks to the comparison of a set of corpora extracted from Wikipedia’s category system.

Verbal lexicon has key functions at textual level. Competence in second language - and specially narrative competence, with which we are dealing here - can be influenced by lack of verbal lexicon, by partial acquisition or idiosyncratic acquisition of it. A textual and conceptual approach of some narrative productions in French and Spanish as first and second languages (FLM, ELM, FLE and ELE) will allow us to compare both cases in these two languages in order to make clear how oral narratives are built up at propositional level (mainly, the explicit reference to processes, space and temporality) and at suprapropositional level (cohesion and coherence). Finally, we will study the influence of real oral input in the learners' oral productions.

Unlike two-word collocations, multi-word discourse markers have until recently suffered comparative neglect in lexicology and lexicography. The present article aims to remedy this deficiency for French. After giving an operational definition of the lexical items in question, the author proceeds to classify them by functional criteria. He concludes his article with a detailed survey of suggestors. The great frequency of this type of marker would seem to belie the assumption that academic language is free of subjectivity.

Spoken word recognition involves automatic activation of lexical candidates compatible with the perceived input. In running speech, words abut one another without intervening gaps, and syllable boundaries can mismatch with word boundaries. For instance, liaison in 'petit agneau' creates a syllable beginning with a consonant although 'agneau' begins with a vowel. In two cross-modal priming experiments we investigate how French listeners recognise words in liaison environments. These results suggest that the resolution of liaison in part depends on acoustic cues which distinguish liaison from non-liaison consonants, and in part on the availability of lexical support for a liaison interpretation.

Following an empirical study, this article deals with a specific type of errors in vocabulary acquisition of French as a second language: intralingual lexical confusion. 480 English and Greek speaking learners (beginners, intermediates and advanced) of French were given a L2-L1 translation task of French words. A large number of errors revealed confusions due to formal and/or semantic similarity between the target words and other French words known by the learners. The formal and/or semantic comparison between error and target word allowed a narrower description of intralingual lexical confusion highlighting some precise sub-categories of what the author calls "lexical confusibles".