Pictorial Marks, Anamorphosis and the Nature of Depiction

Theories of depiction – notwithstanding their competing positions – seem to agree on one general statement: images have to be analyzed from a twofold perspective. On the one hand, there is the material component of images: images are real surfaces, covered by oil, temperas, charcoal or some other coloring material. On the other hand, a picture is a kind of representation: it depicts X because it is a surface capable of occasioning a visual experience as of X. From this theoretical point of view, it seems to be obvious that pictorial marks belong to the material component of the picture. Therefore, it follows that if we want to focus form and nature of pictorial marks we have to divert attention pro tempore from the depicted content and to gaze to the picture’s material component. Notwithstanding the twofold nature of picture perception, But is this really the way things are? I don’t think so. Let us assume, for the sake of argument, that pictorial marks belong to the material component of the picture. In this case there are two different positions about their nature, but I think that both of them are wrong. First, it is possible to claim that pictorial marks are conventional graphemes which pertain to the pictorial page – a thesis which was explicitly uphold by Goodman, but which is still at work in Gombrich’s “making and matching” theory. According to Gombrich, pictorial marks are “useful tricks” and we need them in order to encode and decode whenever information is embodied in depiction. I don’t think that pictorial marks “can be described as simple methods of coding” (Gombrich), and as a matter of fact it is simply not true that in order to perceive a Renaissance perspective painting you need to know the laws of perspective or in order to see the figurative content of Seurat’s La grande Jatte you need to know what chromoluminarism is. The separation of colors into individual dots instead of mixing pigments is a possible solution to a figurative problem because it works – because it is in nature of color dots to interact optically. It follows that to mark with color a pictorial surface is not the same as to encode depictive information in symbols. According to the second thesis we hinted at, pictorial marks belong to the material component of the picture, but are not conventional in nature; on the contrary, they must be understood from the standpoint of those technical rules the painters – but not the spectator – has to know if he wants to modify the pictorial surface in a way which is designed to control the appearance of a particular figurative content. As Hyman suggests, technical rules mediate “between the marks on the surface of a picture and its content”. Technical rules must be effective: the viewer need not to know what the rules are, but he has to see in the picture what the picture is about. Hyman is right in suggesting that there is an asymmetry between viewer and spectator, but I think he is wrong in describing pictorial marks as if they were means to a goal. To boil water you can use a gas or an electric cooker, but the outcome is the same – boiling water. Different technical devices can lead to the same outcome; on the contrary, no perceptually different composition of pictorial marks can realize the same pictorial content. You cannot have the same whole connecting differently different parts. So, the separation of colors into individual dots is not just a way to achieve a pictorial goal – greater luminosity, according to Seurat’s misconception of the optical law of : it is the way in which we perceive the depicted scene to be. We do not see the depicted scene thanks to pictorial marks we cannot completely neglect, as well as we cannot completely divert our gaze from the canvas or from the picture frame. Things are different: looking at The Grande Jatte, we see a world which is figurative in nature and whose nature we grasp in its being perceptually made by individual color dots. Pictorial marks as such do not belong to the material component of the canvas and it would be a mistake to identify them with the material dots of pigments which alters the appearance of the pictorial surface: they are part of the figurative content as such and that they greatly contribute to its perceptual nature. These remarks are in some respects relevant for a better understanding of the nature of depiction. Seeing a picture of a landscape is in important ways different from seeing a landscape, but this is not tantamount to saying that in order to appreciate this difference we have to focus the pictorial surface and the material component of the picture: it means that we see a painted landscape, i. e a virtual object (as Gibson puts it), whose perceptual nature is essentially tied up with its being made of pictorial marks. Pictorial marks are an essential ingredient in the phenomenology of depicted objects as such – a consequence which I would like to clarify by analyzing two borderline cases: on the one hand I will discuss (though very sketchy) trompe l’oeil paintings in which the perceptual presence of pictorial marks seems to fade, on the other I will dwell more on anamorphic images – that is, images which seem to vanish in the mere presence of pictorial marks.