French liaison has long been a favourite testing ground for phonological theories, a situation which can undoubtedly be attributed to the complexity of the phenomenon, involving in particular phonology/syntax, phonology/morphology, phonology/lexicon interfaces. Dealing with liaison requires stepping into all the components of the grammar, while at the same time tackling the quick sands of variation. The data on which a number of formal analyses are based have often been a source of concern since liaison, in part because of its intrinsic variable character, requires extensive and robust data. In the wake of the results from the study of other corpora, we present here extensive results based on the PFC database (Phonologie du français contemporain: usages, variétés et structures) and point to their implications for models of linguistic structure. While we do not believe that a motivated theoretical account can be mechanically extracted from the data, we conclude that future analyses will have to take explicitly into account the results of extensive corpus work as well as sociolinguistic surveys, acquisition studies, experimental phonetics and (neuro-)psycho-linguistic investigations, including the relationship between speech and writing. As stressed in Chevrot, Fayol and Laks (2005), these analyses will have to acknowledge that French liaison is not a homogeneous locus but a multi-faceted phenomenon requiring us to accept, without demur, the crossing of disciplinary boundaries.

Lobular histology and age less than 55 years are found to increase the risk of CBC, while adjuvant chemotherapy significantly decreased the risk of CBC. The progressive rise in the annual incidence rates of CBC, together with the absence of a link between clinical prognostic factors of the first cancer and CBC, suggested that CBC can be considered as a second primary breast cancer.

La thèse doctorale de Sanford Schane (1965) a lancé une longue tradition de travaux où la liaison est envisagée premièrement comme une question essentiellement phonologique, deuxièmement comme le résultat passif du non effacement de segments sous-jacents, troisièmement comme facilement explicable si on a recours aux outils théoriques pertinents (traits, frontières, règles ou contraintes, par exemple). Dans cet article, nous présentons une autre méthodologie pour étudier la liaison fondée sur le programme PFC (Phonologie du Français Contemporain : usages, variétés et structures) . Nous soutenons qu’une base empirique plus adéquate fournit une image différente de la liaison comme un phénomène multi-factoriel et multi-niveaux largement influencé par des effets de fréquence. Nous présentons un ensemble de résultats allant de l’enchaînement au rôle de l’âge, du niveau d’études, du sexe ou de la localisation géographique.

No abstract

No abstract