That Deceptive Line: Plato, Linear Perspective, Visual Perception, and Tragedy

In The Renaissance Rediscovery of Linear Perspective, one of Samuel Edgerton’s claims is that Filippo Brunelleschi and his contemporaries did not develop a three-dimensional style of representing the world in painting as much as they reappropriated a way to depict the natural world in painting that most mirrored the human perception of it. Edgerton claims, and convincingly supports, the idea that many earlier cultures had utilized linear perspective in their visual arts, and he points to one of the earliest extant works on architecture, Vitruvius’s De architectura, to support that claim. In second-century Roman architect Vitruvius’s account of theatrical architecture, he cites an example, supposedly from the Attic Greek era, of a primitive form of “scenography.” According to Vitruvius, during the staging of one of Aeschylus’s plays, Agatharcus of Athens worked as a “stage manager” and wrote an account of how he created a painted “backdrop” that hung behind the players on the skene. The account describes a geometric system of portrayal on this backdrop that assumes “a fixed center is taken for the outward glance of the eyes and the projection of the radii, we must follow these lines in accordance with a natural law . . . what is figured upon vertical and plane surfaces can seem to recede in one part and project in another.”¹ Though there is some controversy surrounding the proper interpretation of this statement, it would seem that Vitruvius is describing the depiction of a three-dimensional image on a flat surface, so that the horizon line of the stage is seemingly extended into the painted backdrop.

Though there are no surviving examples of Greek wall painting or scenography, there is some evidence to suggest that the Greeks understood and utilized the visual principles of linear perspective in their painting. Images preserved at Pompeii, which seem to draw inspiration from much earlier Greek art, depict three-dimensional scenes that illustrate that this technique was understood and performed in Greek wall painting.² If Vitruvius’s account accurately describes what Edgerton asserts, it is important to note that one of the earliest references to linear perspective in Western culture springs from theatrical production.

Even Plato seems to speak about linear perspective. One of the problems that he has with visual mimesis (representation) in painting appears in Book X of the Republic. He writes that one of the problems with a painting of a bed versus a bed that exists in the real world is that the painting “touches only a small part of each thing” that it depicts.³ In other words, if one is gazing at an actual bed, one can walk around the bed and look at it from multiple angles. However, if one is looking at the bed depicted in a painting, that viewer can only gaze at the bed from the painter’s point of view. This statement seems to make the most sense if one assumes that the Greeks had utilized linear perspective in their painting.

In this essay I will make the assumption that the Greeks utilized linear perspective in painting, and I will tease out a Platonic response to this technique in the visual arts. I will begin by offering reasons to think that one of Plato’s concerns with painting is that it relies on a singular line of sight that is inherently problematic. This embedded objection appears in Republic X, and it may be an objection present in other Platonic discourses on art. Next, I will show good reasons to think that this objection to painting might also be an objection to the performance of theatrical tragedy. Finally, I will offer an alternative approach to linear perspective’s relationship to tragedy, incorporating Stephen’s Halliwell’s interpretation of Plato as presented in The Aesthetics of Mimesis. We are often taught that Plato is an opponent of tragedy, and that Aristotle’s Poetics is a response to Plato’s condemnation of the form. However, often it is unclear exactly why Plato opposes the tragic form. In this essay I intend to offer at least one Platonic objection...