What meets the eye. Cognitive narratology for audio description

Cross-modal (intersemiotic) translation is the general task of audio description but not the specific problem faced by audio describers of narrative texts. Their specific task for this textual subgenre is to select those discourse elements which produce narrative force (narrativity), and thus to attain a degree of narrative equivalence between source film and audio described film. This task is potentially difficult for two reasons: because narrative force is not just the realized action on screen: it is the receiver's state of mind, induced by the assumably realized events and by the discursively suggested hypothetical events; and because the suggestive, 'intentional' discourse triggers can be small. The following problem-solving procedure is proposed: equipped with a cogent definition of narrative force, audio describers can learn to recognize their narrative states of mind and become aware of the (possibly small) discursive triggers that generate those mental states. The second part of this paper argues that audio describers may also attempt to measure the spatial equivalence between the film images and the visual imagery that their descriptions produce -- because such equivalence may be relevant for impaired people with residual mental imagery. Wallace Chafe's Pear Film is used to illustrate these ideas.