Spatial Framing Adverbials : Linguistic and Psycholinguistic approaches – SFA

1. Scientific context and objectives of project. In linguistic and psycholinguistic research, it is generally acknowledged that a coherent interpretation of discourse depends on two types of relations, namely semantic relations and referential relations, respectively signeposted (when they are signeposted) by connectives and by anaphor. The project submitted will propose an alternative hypothesis: introducing a new type of discourse relation, namely that of indexation (or forward-labelling) - as opposed to the connection relations backed up by connectives and anaphor. This hypothesis stems from research conducted over a number of years at the LATTICE laboratory (in collaboration with other research teams) into the way that some adverbials, we called framing adverbials function. Such adverbials (adverbs, prepositional groups, noun phrases, etc.) in sentence-initial position focus on a semantic criterion (spatial, temporal, mediative, thematic, etc.) which indexes the utterance in which they appear and extends, by default, to subsequent incoming utterances, unless certain closing markers contradict it. This remote indexation capacity involves scope relations which are directed downstream in the discourse, as opposed to the upstream relations signposted by connectives and anaphor. The project submitted is intended to support and further this conception of discourse relations. The analyses undertaken will focus on spatial adverbials and involve two types of investigation: linguistic corpus studies, and psycholinguistic experiments. 2. Description of project, methodology The first part of the submitted project consists of a number of studies on adverbs and prepositional non argumental constituents introduced by either static or dynamic spatial prepositions. These studies aim to show that, although prepositional constituents with static spatial prepositions are better equipped than those with dynamic spatial prepositions to extend their scope beyond the host sentence, the latter are nevertheless capable of assuming such functions in contexts involving verbs of perception. The second part, will more generally defend the idea that the initial positioning of non-syntactically integrated adverbials sets in motion a grammaticalisation process which can operate in two opposite directions: the connectional path which ends when the group becomes a connective (with an initial stage in which it functions as an ellipsis or an anaphora) and the framing path (in which lexicalisation and desemanticisation are globally less developed and less compatible with anaphoric type relations). A clearer insight into factors likely to explain these changes (which lead to polyfunctional expressions in a given evolutionary state of language) is all the more interesting in that these two paths are divergent and that very few detailed studies of these forms have been undertaken up to now. The third part will involve two series of psycholinguistic experiments designed to test the effect of adverbial positioning on reading comprehension. The general assumption underlying these experiments is that placing an adverbial at the head of the sentence gives rise to a special kind of processing, with the subject having to retain in his/her memory the semantic criterion emphasised by the adverbial and control the closing and the attachment of the frame(s) that it introduces. The linguistic investigation will be carried out on different corpora, which will be annotated as appropriate for the hypothesis tested. The corpora will be collected with a view to studying modern-day written or spoken French or with a view to making comparative studies: present-day French/old and middle French and French and other romanistic langages and Serbian. The psycholinguistic experiments will make use of online methods (self pace prensentation in a first stage). The linguists and psycholinguists involved in the project already have experience of working together. 3. Expected results T.