Our understanding of vision took a long time to develop. Many obstacles had to be overcome: fundamental misunderstandings about the nature of light (physics), prohibitions against dissecting human bodies (anatomy), erroneous preconceptions about the eye's function (physiology) and errors in understanding perceptual, or interpretive, aspects of vision (psychology). To appreciate the struggle to achieve even our current, imperfect level of understanding, consider a few of the main historical landmarks.
A theory of vision must explain how information crosses the space between the perceiver and the object he/she is looking at. Early theories of sight offered three major opinions on this matter. One opinion held that the eye sends out rays to objects and that these rays give the viewer information about color and shape. Among the best known proponents of this view were the Pythagoreans, adherents to the religious and scientific philosophy developed by Pythagoras (he of the theorem). A second opinion held that sight depended upon an interaction between images that were ejected from the eye and the perceiver's own spirit, or soul. Socrates and Plato were among the best known of this theory's supporters. Finally, another opinion held that when people see, they actually make contact with the objects they see, or with replicas of those objects.
In the fifth century B.C., the Greek philospher Democritos believed that every object contained numerous replicas of itself. These replicas were called eidola (singular is eidolon). Every object continuously emits eidola. If someone looks at some object, its replicas fly toward the viewer and enter his/her eyes. According to this scheme, you are able to see some object only because one of the object's eidola has winged its way into your eye. Several observations were taken as support of the eidolon theory. For instance, the fact that someone could see his/her reflection in water or in a mirror proved that the viewer is emitting eidola.
Eidolon theory may strike you as preposterous. However, try to refute eidolon theory without using modern knowledge about physics and light. It's very hard, but for centuries those were the handicaps that faced anyone who doubted the theory. Take a moment to come up with logical arguments against eidolon theory that don't depend on modern physics or modern instruments. No one argument alone was sufficiently powerful to deliver a knockout blow to eidolon theory, but the accumulated weight of argument laid upon argument ultimately prevailed.
Here are two of the arguments that contributed to the theory's ultimate rejection (I've excerpted these from V Ronchi's book, The Nature of Light: An Historical Survey, Cambridge: Harvard University Press, 1970).
Suppose that you are watching a parade. The eidolon theory suggests that you see the marchers because each marcher's eidolon reaches your eyes. What, though, governs the relative positions of all the eidola? How do the eidola maintain themselves in a formation that corresponds to the parade formation?
According to eidolon theory, you can see your reflection because one of your own eidola struck the mirror and returned to your eye. But if each eidolon was flying away from you, it should fly with its "back" toward you; how does the eidolon contort itself so not only do you see its front but also so that in the mirror your right ear appears toward your right and your left ear toward your left?
As I said before, an alternative theory of visual perception suggested that sight occurs because rays emanate from the perceiver's eyes. After leaving the eyes, these rays explored the world, and upon encountering objects, the rays returned, carrying information about those objects. One of the foremost proponents of this theory was Euclid, whose contributions to geometry you probably know about; Aristotle and Plato also subscribed to this theory.
Euclid (he of geometry fame) offered many arguments on behalf of rays emanating from the eyes and against the idea that objects emitted eidola. One of his arguments is particularly interesting because its depends neither on physics, nor logic, nor geometry; instead, it's an argument from purely perceptual grounds. Arguing against the idea that objects emitted eidola, Euclid wrote:
"If the act of seeing were caused by the emission of images --and from all bodies images were being emitted perpetually, which would excite our sense of vision-- how is that he who searches for a needle, or peruses a page of a book, does not see all of a sudden the needle or all the letters? is it perhaps because he is not concentrating? No, since while someone searching carefully may not find a thing at once, often others, while talking, and therefore not concentrating on the task at hand, find the object more quickly." (translated in Ronchi, 1970; p. 16; edited lightly by Sekuler)
So for ten centuries versions of these two theories competed with one another: one claiming that objects emitted copies of themselves and the other claiming that vision depended upon rays that emanated from the eyes themselves. During those ten centuries, philosophy and science steadily declined in the western world, while new centers of excellence were developing in the Islamic world. Though little known today in the West, one of the greatest scientists of medieval time was Abu Ali Mohammed Ibn Al Hasn Ibn Al Haytham, whom we know today as Alhazen. During the waning years of the tenth century and the first years of the eleventh century, Alhazen worked out most major details of our contemporary theories of light and vision.
Most surprising, Alhazen used experimental methods not much different from those in use today. Instead of relying on received wisdom --ideas passed down from authorities such as Aristotle-- Alhazen insisted on checking things out for himself. This may sound straightforward and commonsensical, but in Alhazen's time this approach was virtually unimaginable. Among other topics, he studied afterimages that resulted from looking at bright objects. He also demonstrated that objects that are visible under some lighting conditions can become invisible under others. The best example of this latter phenomenon is the invisibility of stars during the day (before Alhazen, people believed that stars couldn't be seen in the daytime because the stars had actually been extinguished). Alhough Alhazen made understandable mistakes about the eye's anatomy and optics, those mistakes are trivial when weighed against his powerful insights.