Special talk, 3/25 – Caroline Féry

It has repeatedly been reported in the literature that French prosody reacts in a different way to changes in information structure as compared to Germanic languages (Delais-Roussarie 1995, Post 2000, Jun & Fougeron 2002, vander Klock, Portes et al 2014, Goad & Wagner 2018, among others). But all authors do not agree as to how to analyse this difference. Some propose that it is just a matter of degree across those languages (see the authors above), and thus the same prosodic tools can be used in French and in English. I propose that French has a different intonation system altogether (Féry 2014), the most important clues being the absence of pitch accents and the emphasis on the boundaries of prosodic constituents. I will show two experiments on French prosody in collaboration with Emilie Destruel. The first one compares post-verbal given and new objects and adjuncts and finds that the phonetic correlates of phrasing are larger for adjuncts than for objects. The second one investigates pairs of post-verbal objects and adjuncts in different information structural conditions: all-new, only one of the two constituents is focused, or both are (dual focus). In both experiments, it is the correlates of phrasing that are variable, but these correlates do only a poor job in unambiguously expressing information structural roles. The reason is that information structure cannot change the syntax-based phrasing, and the role of phonetic prominence is not clear in French. I will also briefly discuss vander Klock et al.’s semantic proposal and assess it in comparison with my intonational one.