Observational drawing is drawing from a perspective. However, artists could not transfer a perspective image on their retina onto a pictorial surface. The first and foremost challenge to accurate observational drawing stems from the extremely limited capacity of the mind to retain visual details. In theory, there are two ways to overcome the limitation in observational drawing: to take frequent looks, and to improve understanding of what one is looking at. Accordingly, art students and novice artists are advised to take frequent looks at their models as well as to study the structure and anatomy of whatever they depict, be it human figures, trees, architectures, or clouds. On the other hand, advanced and experienced artists do not necessarily take frequent looks when they draw from observation. When they take a look at an object they are depicting, they are likely to be ascertaining what they expect the object to look alike from their perspective.
Here lies the key theoretical issue in vision science that bears on accurate observational drawing: How much accuracy in drawing can be derived from a close look at a scene? In the Helmholtz-Gregory-Marr-Rock constructivist framework for theorizing vision, there is a distinction between perception and visibility: A perspectival image is a means to the end----the construction of a mental model of a distal object. Accordingly, the perspectival image must be encoded and processed to serve a purpose of everyday vision, but can remain invisible in itself. Indeed, philosopher are still debating whether perspectival images are perceived or imagined (for a recent and extremely viewpoint, see Schwenkler & Weksler (2019), Phenomenology and Cognitive science, 18, 855-877). In observational drawing, the goal of vision has to shift from comprehending and interacting with 3D objects and environment to perception of 2D pictorial relationships in order to simulate them on a pictorial surface. Although retina images are invisible, artists do share the compelling experience of the “innocent eyes”, best summarized in the following quote by John Ruskin (1957, p. 27): “The whole technical power of painting depends on our recovery of what might be called the innocence of the eye; that is to say, of a sort of childish perception of these flat stains of color, merely as such, without consciousness of what they signify—as a blind man would see them if suddenly gifted with sight”. Although a “mere flat stain of color” may not properly characterize the artists’ percept even when they aim to see things as such, nor even the visual perceptual capability of an infant or a blind person suddenly gifted with vision sees, it does provide a summary of what many artists are trained to see. Many instruction books on drawing advise beginners to see geometric shapes in human figures and still-life set-ups and use short straight line to depict global shapes or shapes that “contain” a figure in an imaginary frontal-parallel plane. Common to such useful techniques are the requirement to see 2D imaginary shapes from 3D scenes. They are imaginary first of all because they are projections on an imaginary rather than a real frontal-parallel plane (Alberti’s window), and secondly because they could be any subjectively imposed figures or shapes conceived from a projected image, including figures that delineate negative spaces. The capability for “imaginary seeing” is the essence a culturally learned and continuously expanding mode of vision (Lou, 2018) that involves reconfiguring the visible in a perceived (or identified) scene with voluntary attention and imagination.
0 Comments
I took a workshop of Kenneth Cadwallader several years ago. Ken paints beautiful sunny impressionistic paintings that remind me of Spanish master Sorolla's. HIs art education includes being tutored by the late master painter Richard Schmidt. One of the lessons I learned from Ken's workshop is a powerful rule on color temperature relationships: When the lighted side of an object appears cold, its shaded side and shadow will appear relatively warm despite of reflections from ambient objects, and vice versa. Ken insists that this relationship in color temperature contrast is an objective truth waiting to be confirmed by any serious and attentive painters. He insisted on the need to meticulously inspecting all the color relationships in the last phase of painting, adjusting and fine-tuning each one to achieve a consistency according to the rule.
The rule works. By religiously following the rule and trying to see the color relationships as I was told to exist objectively and paint them as such, I was able to produce prettier and more "mature-looking" paintings. Nevertheless, I kept wondering whether I was truly painting the colors that I see or simply being a blindly conforming to the rule. Objectively or physically speaking, the color temperature of a "patch" in a scene to be depicted is first of all influenced by the dominant wavelengths of the reflected light. Cool lights are simply lights dominated by the shorter wavelengths in the visible spectrum, and warm lights are simply lights dominated by the longer wavelengths in the visible spectrum. Without considering the reflections from ambient objects, the degree in which longer or shorter wavelengths dominate the reflectance from a shaded side of a laterally illuminated object would be same as that from its lighted side. However, for two reasons, the warm-light-cool-shade and cool-light-warm-shade rule could be valid in many circumstances. First, the rule may have ecological validity in two types of scenes favored by artists: 1) overall warm indoor ambient reflections with cool light from a north-facing window impinging on objects to be depicted; 2) sunny outdoor environment, where the shaded and shadowed areas of objects receive reflections more from the cool and diffused sky light than from the direct sunlight. Secondly, the perception of color temperature can be conflated with the perception of brightness so that a reflectance in the shaded side is perceived not only darker but also warmer in contrast to its appearance in the lighted side, and vice versa. It becomes clear from this analysis that the rule is unlikely generalizable to objects under all possible lighting conditions, and it is questionable at least in some cases whether the warm-cool contrast is genuinely perceptual and not conflated conceptually. Despite its lack of empirical support, the rule regarding color temperature contrasts seems aesthetically effective and appealing. Paintings done by following the rule religiously appear organized, harmonious, and even beautiful, depending on the skill level of the artist. It reminds me of another powerful rule taught by many art teachers: Get values or value contrast right and colors will take care of themselves, or at least not stray far off. The rule has been backed up by recent discoveries in neuroscience as suggested by Margaret Livingstone (2002), the eminent Harvard Medical school scientist. That is, the primary means for obtaining information from the visual world by human and other mammalian visual systems is by extracting and enhancing value (luminance) contrast, and that information provided by the color is secondary and piggybacked on a neuronal architecture that is first and foremost concerned with value contrast, in artists' terms. For painters, especially painters of various "colorist" styles, color is of paramount importance, not necessarily in its contribution to object recognition----the primary goal of everyday vision as proposed by late computational vision scientist David Marr (1981), but in constituting a means for conveying the emotional and motivational connections between what one sees or imagines to have seen and what is to be depicted. As such, I suspect that the primary function of the rule of color temperature contrasts is similar to that of value contrasts. Namely, it offers an organizational principle (a rule of thumb) that appeals human visual aesthetic preference for structures. In other words, it is likely more of a principle for painting what most painters deem pleasant to see rather than what they actually see. Different artists and even the same artist at different times drawing or painting from the same perspective of a sitter or a still-life setup typically end up with different pictures. It is convenient to say that every artist sees things differently and they end up depicting whatever they see. Although such an answer is satisfying to psyches of an individualist culture, it can prove to be more useful to try to unpack what it means:
1) There is a gap between what one intends to depict and the what emerges on the pictorial surface. Some media, such as oil and color pencil, are more controllable than others, such as watercolor and air brush, and artists who have practiced more and are more comfortable with their favored media are more likely to obtain what they envision. 2) Even with "perfect" control, there remains a gap between what the artist sees and what is achievable on a pictorial surface with the constraints on a given painting medium and the painting tool (for example, a paint brush vs. a painting knife) used. In other words, it is impossible for anyone to depict exactly what she sees but possible to depict a representation of what she sees that incorporates the knowledge and skills associated the painting medium and tools used. Because seeing involves visual representation of the world----a consensus in mainstream vision science, painting has to be considered, in the best scenarios, the materialization of representations of visual representation, or more plainly, visual imagination. As I will argue later, the realization that observational painting involves second-order representation or imagination can be inspiring and have pragmatic consequences. 3) Seeing as visual representation is inherently selective----a consensus among vision scientists based on decades of empirical researches. Most notably, the experimental findings of the "change blindness" and "inattentional blindness" have provided convincing evidence that 1) conscious visual awareness at any given moment is extremely limited and almost entirely driven by what one is interested in, whether one is aware of the interest or not; and 2) short-term retention of visual information is extremely limited, and almost entirely driven by what one is interested in, whether one is consciously aware of the interest or not. In fact, Most visual artists would find nothing surprising in these two conclusion, which are totally consistent with their experiences in trying to fill the gap in paintings between what they see and what they believe they see. Combining these two conclusions together, however, leads to the insight that variations in depiction of the same scene from the same perspective must arise because of the variations of attention and interest of the artist beholding the scene in front of their eyes. |