Comments on W. A. Adams’ Analysis of J. M. Kennedy’s Seven Features of Pictures

Leonardo, Vol. 11, pp. 295-297. 0Pergamon Press Ltd. 1978. Printed in Great Britain. COMMENTS ON W. A. ADAMS’ ANALYSIS OF J. M. KENNEDY’§ SEVEN FEATURES OF PICTURES Ramona J. Demander" John M. Kennedy has listed seven features of pictures that are not found in non-pictorial objects. He claims that viewers must develop a special skill before they can interpret these features [l]. William A. Adams has critically analyzed the features [2], but I believe his analyses are faulty, as I shall indicate below. The features are considered in the order found in Kennedy’s and in Adams’ article. 1. Kennedy stated that pictures are flat surfaces that depict things other than themselves and that viewers must be able to hold the flatness of a picture’s surface in abeyance and treat the information for the pictured world somewhat separately. Adams takes this to mean that viewers must disregard the flatness of a picture’s surface. Surely, ‘to treat. ..somewhat separately’ does not mean to disregard! 2. Kennedy stated that the flatness information must not be forgotten entirely or one will suffer trompe l’oeil and possibly be unable to compensate for viewing a picture from incorrect viewpoints, because one will not be able to recognize that one is examining a depicted scene via a canvas at a particular location (see also point 6 below). Adams recasts the statement to read: ‘The disregard must not be total, else the picture would be mistaken for a window.’ This is correct. However, Adams then says: ‘There is no danger that pictures will be mistaken for real scenes, unless they are viewed monocularly from a fixed position.’ This is not correct, for, although it is true that binocular vision allows one to distinguish more accurately between a real object and a picture of it in most cases, it has been shown by Kennedy and Magee [3] that, even with binocular viewing and physically unrestricted viewpoints trompe I’oeil can occur. This was demonstrated by showing to a few persons an outline drawing of four children facing an open central design and a pencil above them to the right (Fig. 1).When directed to look at the picture and to complete it by drawing something in the open central design, the persons, after gazing at the picture for a while, on occasion would reach for the depicted pencil! A trompe l’oeil effect like this was probably due to the fact that the depicted pencil was at the person’s periphery of vision while attention was concentrated on the central part of the picture, as directed. If instead of a pencil an object cognitively irrelevant to the task had been depicted, the trompe l’oeil effect probably would not have been produced. *Psychologist,Life Sciences Dept., Scarborough College, University of Toronto, West Hill, Ontario, MIC 1A4, Canada. (Received 12 May 1978.) Normally, when viewing figurative pictures binocularly , one can interpret flatness information by knowing from which location the depiction was observed. Presumably such knowledge, if it is valid, is needed for one to compensate appropriately for viewing a picture from an incorrect or less than optimal viewpoint (say the picture was upside-down). In addition, one’s own final decision regarding what the viewpoint is, can depend on several factors: the viewer’s skill in interpreting pictures, the amount and kind of depicted detail present in the picture, the degree of depicted slant, whether the edges of the picture are concealed, effects of illumination, etc. Thus, contrary to Adams’ statement, there is a danger that some pictures will be mistaken for real scenes even when viewed binocularly. D. R. Topper has also commented on Adams’ article as regards the topic of trompe I’oeil [4]. 3. Kennedy stated that pictures arenot replicas, or scale models. They necessarily leave out information. The skilled viewer must know the implications of omitted things, like absent details in a line drawing or no colors in a black and white photograph. He calls this selective inattention . Here Adams points to the importance of phenomenological invariants (e.g. viewers’ acts of attention)when relevant pictorial information is sought. But this does not contradict Kennedy’s statement, since ‘selective inattention...