Film, like the novel, is intrinsically temporal and good at telling stories. Both succeed by absorbing us into an illusion of comprehensive life – worlds, societies, relationships in which we live intensely for the duration of the telling. Both succeed insofar as they persuasively select what we need to know, and jettison what we do not, and in such a way that we never miss what is not there. That said, these completely absorbing worlds are built from different elements: the one from motionless words; the other from moving pictures. Unlike the novel, film reality is visual; unlike film, novel reality is imagined: that is, the pictures we make as we read are unconstrained by the mechanics of seeing. In Mansfield Park the narrator tells us that, with Mary Crawford’s help, Fanny’s dress to be worn at the ball held in her honour was ‘settled in all its grander parts’ (MP 2:8:299). Two chapters later, she writes of the dress’s ‘neatness and propriety’, of Sir Thomas Bertram’s approval of its effect when he speaks ‘of [Fanny’s] beauty with very decided praise’ (2:10:316), and the rest is left to our imagination – to our mental picture-making. Not so on screen, where the film-maker must decide whether the dress is white or cream, plain or spotted, muslin or silk, with what particular neckline and bodice; and the viewer, no longer free to imagine, takes what she sees into what she knows about Fanny, the eye determining interpretation. While this is a general distinction in the narrative systems of the two media, telling us something about the relationship of content to form in each, it is compounded in the case of Jane Austen’s novels by a strong distrust of visual understanding. Within the sensory and cognitive means available to the novel form, Austen confirms the activity of the eye, of imaginary seeing, less readily than most.