Abstract

AbstractAn essential task of contemporary witnesses in oral history interviews is to remember past events and to communicate these memories to others. This is all the more challenging, as it often involves narrative reconstructions of virtually ‘indescribable’, ‘unspeakable’ or ‘unimaginable’ events. The present study is interested in videographed narratives of French contemporary witnesses of National Socialist forced labour. By verbatim and thus explicitly indicating that they ‘remember something’ (in French they do this withje me souviensorje me rappelle, among others), the witnesses authenticate their memories and reveal their memory process to others. Our analysis shows two different form-function designs: softly articulated forms of memory markers with retrospective syntactic scopus have the communicative function of indicating what exactly is remembered and what is not, in order to be credible as a contemporary witness. Emphasised, clearly articulated memory markers, on the other hand, have a prospective syntactic scopus and have the communicative function of opening an episodic narrative sequence. If the multimodal analysis is extended to the physical movements of the eyewitnesses during narration, it can be shown that the different levels of expression (especially syntax, prosody and posture, direction of gaze as well as gestures) together give rise to an expressive multimodal gestalt.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call