A constructional account of French-clé 'key' and Dutch sleutel-'key' as in mot-clé / sleutelwoord 'key word' 1 1 This paper is a revised version of a conference paper given in French (Amiot & Van Goethem 2010). 2 Since bound forms such as stok-'lit. stick' and-boer 'lit. farmer' still have an independent counterpart, contrary to true affixes, authors such as Meesters (2004) and Booij (2009) prefer treating them as semi-affixes or affixoids: "morphemes which look like parts of compounds, and do occur as lexemes, but have a specific and more restricted meaning when used as part of a compound" (Booij 2009: 208). However, such an analysis raises at least one fundamental question: are the nouns stok and boer still the same "morphemes" as stok-and boer-as components of, respectively, the stok-A / N-boer constructions? Semantically, they do not seem to have any common feature. For instance, the noun boer means 'farmer', while the bound form-boer appears in complex nouns denoting sellers/ grocers (1b).

Cet article porte sur un problème de sémantique lexicale appliquée à un domaine particulier, celui des adjectifs préfixés par pré. Il s'agit en effet d'étudier la construction du sens de ces adjectifs, celui-ci ayant pour particularité de se construire sur deux niveaux. Les adjectifs dérivés s'interprètent en effet avec un sens d'antériorité mais aussi avec ce que j'appelle les valeurs ‘continue’ et ‘discontinue’, cette double structure interprétative étant caractéristique de l'ensemble des mots préfixés par pré- (adjectifs mais aussi noms et verbes). Il s'agira donc de mettre en évidence les mécanismes de construction du sens d'antériorité lui-même, ainsi que les principes qui régissent l'attribution des valeurs ‘continue’ et ‘discontinue’. Ceci nous permettra alors de distinguer ce qui, dans le sens des adjectifs dérivés, est à imputer à des mécanismes purement linguistiques et ce qui relève plutôt de l'ordre du culturel, i.e. des représentations et des connaissances que nous pouvons avoir sur le monde. Les adjectifs prétuberculeux et précolombien me serviront d'exemples privilégiés tout au long de ces analyses.

Summary The aim of this article is to study the relationship between certain adverbs, prepositions, prefixes and conjunctions of subordination in French. A distinction is made between two classes of morphemes. The first one includes prefixes and ‘governed’ prepositions (which introduce arguments of the verb, such as the indirect object) : both types of expressions can be said to be ‘linked’, the first to the prefixed element, the second to the verb. The second one includes ‘not-governed’ prepositions (those which introduce adverbial phrases), certain adverbs and conjunctions of subordination ; these morphemes are not linked in the sense mentioned above and can thus said to be ‘free’. This distinction could suggest a grammaticalization channel leading from the free elements to the linked ones : from the adverbs via the related non governed and governed prepositions, in that order, to the prefixes. It is shown, however, that some Latin adverbs grammaticalize directly into preverbs, without an intervening prepositional stage, and that some prepositions directly inherit ‘governed’ as well as ‘not-governed’ uses from their Latin predecessors. Nevertheless, the hypothesis that there is a relation between the governed or non governed nature of some prepositions and the possibility to use them in order to create conjunctions of subordination, could explain whyavant queandaprès queare still used as such conjunctions, whereassur ce que/sour ce quehas disappeared.

Within the medical field, very specialized terms are commonly used, while their understanding by laymen is not always successful. We propose to study the understandability of medical words by laymen. Three annotators are involved in the creation of the reference data used for training and testing. The features of the words may be linguistic (i.e., number of characters, syllables, number of morphological bases and affixes) and extra-linguistic (i.e., their presence in a reference lexicon, frequency on a search engine). The automatic categorization results show between 0.806 and 0.947 F-measure values. It appears that several features and their combinations are relevant for the analysis of understandability (i.e., syntactic categories, presence in reference lexica, frequency on the general search engine, final substring).

No abstract