Volume 6 (2011)

Noémi Boubel
La tâche de détection automatique de la polarité ou de la valence affective d’un texte devient actuellement importante dans le domaine du traitement automatique des langues. Nous nous intéressons au contexte associé aux termes positifs ou négatifs et proposons une méthodologie permettant d’extraire automatiquement des structures syntaxiques ayant un impact sur la polarité d’un terme tout en identifiant cet impact. Dans cet article, nous présentons et discutons les résultats de cette méthodologie appliquée au cas des adverbes associés à un adjectif positif ou négatif.

Steven Delarue
In this paper, I discuss the polemic regarding the use of standard language and tussentaal (‘intermediate language’) in Flanders from an educational perspective: what language variety (or which varieties) can be used by teachers in primary and secondary schools in Flanders? The language policy of the government demands the use of the standard in all school situations, but the linguistic practice of both teachers and pupils shows a mixed use of standard language, tussentaal and dialect. In addition to that gap between language policy and linguistic practice, there are also internal contradictions to be found in the governmental language policy, where the final attainment levels of the Department of Education of the Flemish government seem more permissive than the demand for standard in the policy documents, as well as in the linguistic practice, where teachers seem to consider themselves as role models who always speak standard language when teaching – which is not the case in reality. In this paper I set out this intricate web of contradictions, to sketch the paradoxical double gap between policy and practice.

Anne-Sophie Ghyselen
In this paper, it is shown that the language repertoire in Flanders is in need of thorough macro-research. In the last decades, many researchers have focused on the form and function of parts of the diaglossic language repertoire, such as the Flemish dialects, the standard language and what is called tussentaal (‘in-between-language’). While these studies have yielded interesting sociolinguistic insights in aspects of the linguistic repertoire, they often tend to neglect the macro level, hence leaving several questions about the language repertoire in Flanders and by extension about repertoires in general unanswered. The present paper tracks some key lacunae in the current literature on the structure of the Flemish diaglossic situation, continuing with a status quaestionis about its dynamics. This double overview will ground a plea for more empirical research into language repertoires as macro-entities. To conclude, a concrete example will be given of how such empirical research might be conducted, taken from a project at Ghent University on the Flemish language repertoire.

Thomas Hoelbeek
Dans cette contribution, je propose un cadre interprétatif illustré de l’usage qui était fait, au XVIe siècle, de la locution française à travers (de). Je m’inspire, entre autres, des travaux de Stosic qui a décrit de façon approfondie l’emploi spatial d’à travers en français moderne en le contrastant avec celui de par. La notion de « guidage » qu’il introduit, et qui serait centrale dans les emplois spatiaux d’à travers, a l’avantage d’être moins ambiguë que la notion d’« obstacle ». Tout en me situant dans une perspective à la fois diachronique et cognitive, j’adopte une approche fonctionnelle dans la tradition des travaux de Vandeloise (1986) et autres. Cette analyse s’inscrit dans un projet de recherche plus ample, qui a pour objectif de dévoiler les changements qui ont caractérisé l’évolution de toutes les prépositions analytiques contenant le mot français travers.

Agnès Provôt
Peut-on définir un invariant sémantique du connecteur si ? Face à la logique et la linguistique qui ont tenté de répondre à cette question selon diverses approches, nous proposons une analyse basée sur le concept de référentiel temporel, déjà utilisé pour la formalisation du temps, de l’aspect et de la dimension énonciative du langage. Nous commençons par présenter une carte sémantique des valeurs du connecteur si, inspirée d’Aptekman (2006), Stage (1991) et Corminboeuf (2009), et nous la détaillons à l’appui de quelques exemples illustratifs. Des représentations en diagramme de ces énoncés montrent comment le connecteur si construit un référentiel distinct du Référentiel Énonciatif et comment les deux situations p et q s’inscrivent dans ce nouveau référentiel pour s’actualiser ou non dans le Référentiel Énonciatif. Les différentes valeurs de si construisent différents types de référentiels mais ceux-ci peuvent se subsumer sous la notion du « possible », qui se spécifie en épistémique ou aléthique et s’applique aux faits, à l’énonciation ou à la relation entre deux situations. On peut donc ainsi dégager un invariant abstrait du connecteur si, en tant que constructeur d’un « Référentiel des Possibles ».

Lauren Van Alsenoy
This paper discusses negative indefinites in African languages on the basis of a representative sample. I will explain why the term negative indefinites can be decieving, since many languages do not use true negative indefinites, like nobody in English. This particularly holds for African languages where the most frequent pattern is sentential negation combined with a bare generic noun. This runs against the recurring claim that negative concord, the pattern whereby a true negative indefinite always cooccurs with sentential negation, is very widespread. Another claim regarding negative concord is that a language has to have negative concord in order to develop double negation as in French ne…pas. It will be suggested that this may only be true under certain conditions which are not satisfied in the languages from the sample.

Daniël Van Olmen
This article is a corpus-based study of the further developments of the imperative of saying. It first examines say in contemporary English. Its uses are then compared to those of its Dutch equivalent. The similarities, which include the functions of attention-getter and approximator, and the differences, which include the more subjective character of clause-initial say and the use of zeg maar as a hedge, provide a new, dynamic synchronic perspective on the ways in which the imperative of saying evolves.

Willy Vandeweghe
Prepositional objects in Dutch tend to combine with specific semantic predicate classes. Prominent among these are psychological verbs requiring an Experiencer role for the subject, and relegating the more causative or agent-like role to the prepositional object. A second class groups the verbs of communication, which are more agentive as a rule, and a third more heterogeneous class comprises verbs that express relationality in one way or another. The paper ends with an overview of the factors pertinent to the diachrony of this emerging class of prepositional objects.

Deze bijdrage behandelt het voorzetselobject (VzO), meer bepaald de predicaten die met dit specifieke type object gecombineerd worden. Doel is een categorisering op basis van de bestaande lijsten, niet alleen om beter inzicht te krijgen in de relationele aard van het VzO, maar ook om enigszins dichter te komen bij de motor die achter de expansie zit van wat door Duinhoven (1989) een ‘zinspatroon in wording’ werd genoemd. Daarom wordt bovenop de synchrone beschrijving ook een stap in de diachronie gezet.

Frens Vossen
Since Dahl (1997) the term ‘Jespersen Cycle’ is commonly used to refer to a process in which a clausal negator is accompanied by an element which acquires negative meaning and eventually takes over as the sole clausal negator. The reference in the label is to Jespersen (1917) and the most often cited illustration comes from French, which, simplifying somewhat, once had ne, now ne…pas, at least in standard written French, and just pas in colloquial spoken French. The process is a universal in the sense that it can, in principle, occur everywhere, but in fact it does not, yet when it occurs if often seems to be contagious across family boundaries and the details can differ significantly from the ones illustrated with French. Up to now, the process is documented best for Western Europe, less well for the southern Mediterranean, and now also for Sub-Saharan Africa (especially Bantu). Yet, suggestions for the occurrence of a Jespersen cycle have also appeared for South East Asia and Oceania, most prominently perhaps for Northern Vanuatu with Early (1994a, b) and François (2003). The present survey reports on the whole of South East Asia and Oceania. The paper focuses on what is peculiar about these Asian and Oceanic manifestations and hence relevant for a general understanding of the process.