• 1 Biederman and Kim 2008.
• 2 Kennedy and Ross, 1975; Cavanagh 1999.
• 3 Shepard and Metzler 1971.
• 4 Tarr and Bülthoff 1998.
• 5 Almost, as some features of pictures, such as resilience to viewpoint change, at least within a cer (...)
• 6 Luminance describes the light that reaches our eyes. Reflectance is the disposition of objects to s (...)
• 7 Kennedy 1974.
• 8 Casati and Pignocchi 2008.

1Drawing — and I speak here of outline drawing that uses just lines: monochrome, with no particular concern for what fills the spaces left between the lines — is a human artifact as ancient as it is mysterious. Even a simple enumeration of facts about it is bound to arouse interest and theoretical curiosity. Here are a few. Drawings are just as old as the oldest human representations known to us. The painted animals in the caves of Lascaux and Chauvet are surrounded by thick lines that have a remarkable degree of pictorial sophistication1. A primordial instrument of representation, drawing has virtually no history: it has remained unchanged through the ages — it's hard to imagine huge innovations in comics, for example — and the differences between drawings from different times and places are incomparably less salient than the differences between the styles of representation in paintings that uses shapes and colors and not just lines. The universality of drawing also hints at a substandal difference from the other great graphic achievement of our species, writing. On the one hand, leaming to draw (in a rudimentary way) is much easier than leaming to write, and on the other hand, the content of drawings can be deciphered by pre-linguistic children, by people who live in cultures without drawing, even by some animals2. At the same time drawing and writing are related: they both rely on action routines, the simplest of which are common to both (drawing circles, strokes) and are learned in the same way. The way we learn to draw — for example, the fact that it is easier and faster to learn to draw by copying other drawings than by drawing from life — indicates that visual perception can retrieve a large deal of motor information about the action that produced the drawing. Modern society is tributary to drawing: it was not always so, but most of the drawings are today prescriptive drawings (how-to leaflets, projects), and conversely, almost all of the objects that are produced today are the result of a drawing that is was used to design them. Nor is it clear whether things will change dramatically with the forthcoming synergy between three-dimensional digital modeling and the diffusion of extrusion 3D printers. It is not at all evident that the new systems will allow designers to sidestep drawing, which provides an initial view on which to build up a project, since it is not clear whether it is possible to think of complex objects in 3D — and if it is possible3 2D vision is still crucial for object recognition4. The cognitive mystery of drawing stems from the fact that its ability to represent objects is all but evident. Consider that paradigm of iconic representation, the photorealistic image: a high resolution color photo. A simple theory of its functioning may almost5 not even make mention of psychological facts and mental events, as it may consist of a description of the way the picture structures the optical array centered on a “well situated” observer, together with the realization that this array would be exceedingly similar to the one that would be structured if instead of the picture there was in front of the viewer the scene that the picture represents. (The simple theory, as a tribute to his supporters, and to the fact that it requires an observer who uses only one eye and a fronto-parallel piane, might be called “cyclopean-Albertian-Gibsonian”.) The theory may not be very appealing, but to its credit one can say that it would be an extraordinary fact if likeness of optical arrays had no role to play in the explanation of how the realistic picture represents. There is no simple theory of drawing in this sense, i.e. a theory that could settle for a purely ecological component (the description of the represented environment) of a theory of perception. The relevant lines are not found, except incidentally, in the “outside” scene that is represented. My real profile is not a black line a few millimeters thick, which the artist could reproduce on paper. To this fact must be added that if drawing, given a liberal interpretation of the concept of “line”, tried to represent all the visual discontinuities in a scene, typically exemplified by differences in luminance6, and then in particular chromatic discontinuities, it would be in the norm illegible. This is due to the fact that differences in luminance are ambiguous: they can correspond to (or supervene upon) differences in reflectance (stable properties of objects) or to differences in illumination (ephemeral properties of scenes). It does not seem to an accident that drawings, historically, abhor profiling shadows, which are purely illumination accidents. This neglect of course may depend on various factors, such as a lack of general interest for shadows in pictorial representation. It is also the case that line-profiled shadows can be recognized as shadows7. At the same time lines that indicate a difference in reflectance (e.g. the border between two patches of different colors on my sweater) are not automatically the best candidate to be represented as lines in a drawing: tracing them all also makes the drawing hard to decipher. And on the other hand there are lines that do not correspond to a difference in reflectance (as when one depicts a white jacket against a white wall), which are most commonly and succesfully represented in a line drawing. Only a few lines are then usefully represented in drawing: the lines that normally correspond to actual or potential occlusion profìles of an object. These are the lines that visually separate the figure from its ground. This visual separation has nothing trivial in a three-dimensional world: the concept of line is very articulate here, since it depends on the relative positions of the observer and the thing seen, and since you can represent profìles that are only potential (like the line that indicates the edge of a nose or a body even when the nose or the edge are not seen in profile). These are not purely visual discontinuities; they have a conceptual component to them. Some of these features of drawing and their uses are certainly contingent. If our fingers left sufficiently permanent traces in air when we moved them, the practice of drawing would have gone a different way8: free from two-dimensional support, but unable to represent the occlusion contours, we would have had very different “drawings”.

• 9 Kennedy 1975.
• 10 Cavanagh 1999.
• 11 Flores d’Arcais 1994, Pignocchi 2008.

2These are the facts, I said, or at least the most salient ones. How are we to explain them? The hypotheses that have been put forward as to how drawings work, i.e. as to how they are tailored, as artifacts, to a cognitive system that has our possibilities of vision, conceptualization and action, are as fascinating as the data they attempt to explain. It is reasonable and plausible to argue that drawings have not been invented, but discovered9 — unlike, presumably, other types of images, bar sculptures. Indeed some items (like leaves fallen on rocks) leave traces that correspond to their profile, thus indicating that it is possible to use the bare profile of occlusion for recognizing an object. Humans thus found that certain select lines — those signaling occluding boundaries — can induce recognition. The main explanatory hypothesis is that if drawings work as they do, this is because the visual system at a certain intermediate stage of information processing is actually “drawing”, i.e. it represents objects in a format that is compatible with the format of line drawing10. We also know that the visuo-motor system automatically extracts from drawings a wealth of information on how they have been produced11. This means that a certain type of information is encoded so that it can be “read” both as visual information and as information on action, on gesture. Another large hypothesis is that writing stems from drawing not only in the simple sense that construes it as a form of iconic abstraction, but because it uses visual routines available to a brain that can recognize and execute drawings. According to various authors, most notably Karmiloff-Smith and Dehaene, the brain is able to appropriate informational pre-existing structures to create new representational vehicles.

• 12 Kennedy 2003.
• 13 The Discovery Channel documentary on Kennedy and his work with blind artist Esref Armagan is worth (...)

3Finally, the fact that people with congenital blindness are able to draw by exploiting cues that would appear as a purely visual, such as foreshortening or occlusion12 also suggests that drawing has internail rules that involve mainly the system of spatial representation, and not only the visual system. What next? We still lack a general theory of performance: a theory that looks at all the phenomena that are relevant to the execution of drawings, such as the unraveling of attention, planning, implementation steps, the online visual control of execution (try to draw a circle with eyes shut), the possibilities of simplification (the discoveries of “mental compression” of an image), the strategic decisions, negotiated step by step during the execution, which could achieve likelihood by sacrificing objectivity. Neither there is a comprehensive theory describing the philosophical possibility of drawing as an activity that essentially involves the exercise of attention (drawing as a form of meditation). We also lack a theory that links the characteristics of drawing to its use in different cultural contexts. It is to be expected that because of the simplicity of its implementation and understanding drawing can do the many things that it has been asked to do in history: representing, educating, planning. But details are still forthcoming. Given the complexity of the phenomenon, a plea for interdisciplinary research is the motivation of this issue of the Rivista di Estetica, which collects contributions from several quarters. Patrick Maynard, the author of Drawing Distinctions, the only philosophical monograph devoted to drawing to date, proposes a Gombrichian approach, according to which the artist taps into a cognitive toolkit to create images — and in particular drawings — that must be appreciated, among other things, just as artifacts, not only as attempts to “deceive the eye”. John M. Kennedy, the psychologist who most worked on drawing, and Sherief Hammad, study the case of a blind adult, Ben, who represents cubes as unfolded structures, displaying some puzzling foreshortening. Based on a compari- son with the drawings of children around eight years, and noting improvements of Ben with practice, it is suggested that the acquisition of the drawing is no different for the sighted and the visually impaired13. Alessandro Pignocchi, author of a monograph on the automatic attribution, by the viewer, of intentions to the artist, sees in the perception of drawings the prototpical case of this cognitive mechanism that allows one to establish a direct link between perception and action, and defines a research area for studying the interaction between art and cognitive science. Giuseppe Di Napoli, author of the monograph Disegnare e conoscere, offers a wide-ranging vision of the drawing activity, centering on the representative properties of the line. Drawing, in this sense, is akin to thinking. Indeed, a large chapter ought to be opened on the nature of drawing as an activity controlled by moment-by-moment decision-making. Drawing is not a mere visual recording, it is chiefly to plan and to execute. In this sense, thought is a key component of drawing. That drawing and thinking are interwoven is discussed in the text by Eduardo Côrte-Real and Susan Oliveira. The evaluative judgments on drawings change over time, depending on the qualities that are preferred in different periods (distinctess or clarity vs speed and the play of chance). The last two contributions, by Agostino De Rosa and Manlio Brusatin, study the role of drawing in design and its uncertain future due to technological advantages which give an advantage to digital models directly creating illusionistic renderings. Commenting on our theme are drawings by some of the authors and a graphic reflexion by artist Matteo Pericoli. The differences in style and content of these graphic comments are in a sense a tribute to the theoretical complexity of drawings.