Seeing Colors Without Seeing Light
Philosophy and Phenomenological Research July 9, 2026 Peer reviewed DOI: 10.1111/phpr.70145 via OpenAlex
Summary
The article argues that light itself is not seen, but is instead the medium through which we see colors. This view, tracing back to Aristotle and Alhazen, holds that colors are independent of light, which merely enables their visibility. The author proposes that color constancy—our ability to see stable colors under changing illumination—is not achieved by the visual system discounting the illuminant's color, as traditionally thought, but by experiencing chromatic variations as perspectival presentations of a single objective color property, analogous to how spatial variations reveal shape.
Study at a glance
| Design | theoretical or philosophical paper |
|---|---|
| Key finding | Color constancy is better explained by understanding chromatic variations as perspectival presentations of a single color property rather than by discounting the illuminant. |
Abstract
ABSTRACT This article defends a provocative claim: Light itself is not seen. Proposed by Aristotle, Alhazen, and recently Hilbert and Kalderon, this view regards light as the medium of sight rather than its object. Unlike reflectance physicalism or color relationalism, this medium theory maintains that colors are entirely independent of light, which merely provides conditions for their visibility. I argue that light's invisibility offers a novel solution to problems raised by color constancy. The traditional approach, originating with Helmholtz, explains constancy through “discounting the illuminant”—the visual system estimates illumination and compensates for it to recover objects’ “true” colors. But if light is colorless and invisible, it cannot contain the variable chromatic component that supposedly explains color variations and, thus, cannot be discounted as contemporary models suppose. Drawing on Husserl's perspectivist framework, I propose that chromatic variations under different illuminations are perspectival presentations of a single color property, analogous to how spatial variations present aspects of shape. We perceive stable colors not by discounting a chromatic component attributed to light, but by grasping variations as perspectives on the same objective color. I contrast this approach with color relationalism, reflectance physicalism, and enactivism, showing how it avoids difficulties they face while better capturing chromatic experience.