Sens de l'enonce et carte neuronale

It is by no means unusual for linguistic theories to dispense with constraints from adjacent disciplines, and to do so, on perfectly legitimate methodological grounds. However, where is a need to refer to sub-personal aspects (such as “cheminement interprétatif” or “topos”), linguistic theories are particularly at risk of postulating entities and processes that may not be supported by cognitive or neurocognitive findings. Is this the case for Anscombre and Ducrot’s argumentation theory (TADL)? I will argue that despite its anti-cognitive stance, the TADL presents some systematic correlations with Damasio’s model of brain functions. Its conception of utterance meaning, in particular, bears a strong resemblance to Damasio’s brain maps for objects, in that in both cases we are dealing with a tripartite structure which includes references to the context of production (or perception), and to appropriate responses, alongside information concerning the object of representation. The hypothesis being put forward is that utterance meaning inherited this structure from brain maps, and is subject to the same mode of memory encoding, one construed in terms of convergence-divergence zones. This level of correlation is not surprising if argumentative phenomena, as construed by the TADL, were to be looked upon as linked to survival.