Skip to main content
Open Access Publications from the University of California

UC Berkeley

UC Berkeley Electronic Theses and Dissertations bannerUC Berkeley

Looking and Learning: Pictorial Representation and Visual Skill


Seeing can be difficult. This dissertation aims to bring out the philosophical significance of this commonplace fact. By examining the roles of difficulty, agency, and skill in visual experience, it sheds light on two related sets of philosophical questions: one about our ability to perceive the world, the other about our ways of representing it.

The first two chapters get at the basic structure of the experience of seeing. I approach this by concentrating on the notion of an opportunity to see. This notion is a familiar one, but upon examination it makes trouble for a traditional and tenacious idea, that seeing is the automatic upshot of a certain mechanical process – a causal chain linking things in our environment to our perceptual organs and our minds. Having presented this conflict, I examine the alternative framework for characterizing visual experiences that the notion of opportunity invites. I agree with other discussions that opportunities to see are defined by contextual conditions like lighting and the placement of occluding objects, and the degree of salience things have in virtue of the kinds of objects they are. But I argue that in order to understand the difference between an opportunity to see and an occasion of seeing, we must recognize skill on the part of the viewer as an additional explanatory factor.

The second chapter turns to a puzzle about how it is possible to see the intrinsic features of things in contrast to the way that they look in certain circumstances – e.g. the color of a painted wall, as opposed to the way shadows darken parts of its surface, or the volume of a cardboard box, as opposed to the square shape of its facing side. This puzzle construes lighting and opacity as obstacles to seeing, and my response emphasizes that illumination is an enabling condition for vision, and a thing’s opacity (the fact that one cannot see past its surface) is what makes it visible in the first place. At the heart of the puzzle is the assumption that whenever we see, we see how things look, and the idea that seeing how things are shaped and colored must involve more sophisticated skill than seeing how they look in the conditions in which we encounter them – if it is possible at all. I argue that this gets it backwards. Seeing how things look – e.g. how the play of light affects the appearance of a thing’s color, and how different parts of a thing’s surface appear from different vantage points – is the more sophisticated business.

The rest of the dissertation focuses on the art of pictorial representation, as a practice that exemplifies the skill involved in discerning the way things look. The third chapter takes up Richard Wollheim’s influential thought that seeing a picture is an experience whose nature is twofold, an experience of both a marked surface and a depicted object. I pinpoint the insight in this idea by explaining the duality in terms of the distinction between a marked surface and the way it looks. This allows us to avoid thinking of the experience of the depicted object as an illusion, or as a “vision” in an occult sense, of something in a mysteriously distinct realm. It also brings out the fact that pictures are things whose visual appearances are designed to be especially salient, and are products of an intention to show how something looks.

The fourth chapter turns to the relationship between depiction and visual resemblance. A central motivation for thinking that the concept of depiction in some way implicates that of visual resemblance is an interest in capturing the difference between pictures and linguistic expression, as two forms of representation. Nelson Goodman criticizes this way of comparing them, in part by arguing that it relies on a flawed conception of the experience of seeing, according to which it always involves an awareness of a determinate configuration of patches of color. I examine attempts to refine the resemblance theory of depiction to avoid Goodman’s objections, and argue that they too build in problematic assumptions about visual experience. They require that depiction always presents viewers with forms of appearance that they are already familiar with – aspects of shape and color that they already associate with things of a certain kind. But we should not think of depiction as constrained in this way. The role of shape, line, and color in depiction is not to remind viewers of looks they’ve seen before, but to get them to see, possibly for the first time, how things can look.

The final chapter attends to the activity of drawing from life, or creating a depiction based on a model. This activity serves as one final illustration of the idea that visual skill is developed with effort. The first part of the chapter argues that we should not think of the process of drawing from life as constituted by two stages, first seeing how the model looks, and then manually modifying a drawing surface to convey that look. Rather, the drawing process involves both visual and manual skills throughout, and sometimes, it is only through the process of drawing that the artist comes to see how the model looks. The second part of the chapter clarifies the nature of the skill exhibited by drawing from life by examining how artists typically train to do it. Finally, I present a way of thinking about pictorial realism in terms of the kind of visual skill that drawing from life requires. On this way of thinking, realistic depiction is not a matter of capturing some particular aspect of the visual world, but rather a matter of conveying our visual sensitivity to the contingent features of objects in our surroundings, whatever they may be.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View