Westphal, Jonathan - Review of Roy Sorenson's Seeing Dark Things

05/28/2010

Paper delivered for panel discussion

This is a review of the Roy Sorenson book Seeing Dark Things, which is a book that espouses the causal theory of perception and the nature of shadows and silhouettes. Author takes the opportunity to discuss Sorenson's main points, but also disagrees with them and elaborates his own position in the review.

The first example author discusses is the example of "Near" and "Far", two planets that completely occlude the sun at the exact same time (from the point of view of the Earth). What happens is that the far planet "Far" occludes the sun, and, simultaneously, a closer planet, "Near" also moves into the same exact position that Far occupies in our field of vision. The closest thing to us is the near side of Near. Nevertheless, Sorenson argues that when we see the big black blot in front of the sun, we are seeing the thing that actually causally stops the sun's rays, which is the far side of Far. Author agrees-- though he explains it not as bolstering the causal theory but instead as part of a theory of shade and in describing celestial bodies in 'purely astronomical terms'. Author also worries about what this kind of example might underwrite for the causal theory, that, somehow, we can 'see' the far (obscured) sides of things.

The next discussion can be brought out by the example of a bird flying (and landing) onto the near side of a barn, but coming to rest somewhere in the barn's shadow. The light source is on the other side of the barn and when the bird flies into the shadow it disappears. From this simple case it seems there is an inconsistent triad:

1) If X causes a shadow, then some light is falling directly on X
2) X cannot cast a shadow through an opaque object
3) All shadow is shadow of something

With the bird/barn case, it seems the barn is casting a shadow through the opaque object of the bird, a violation of 2. But then it is clear that the shadow is in the shape of the barn, upholding (sort of) 3. But if we say the barn isn't casting a shadow through the bird, then 1 saddles us with the absurdity that some light is hitting the bird in order to keep the barn-shape of the shadow.

Author makes a comment on the ungrammatical aspects of 3, but first goes about denying the sense of 2 that generates the inconsistency. Author argues for a distinction between the concept of shade and that of shadow. Shade is the lack of illumination, whether partial or complete. Shadow is shade with shape. Shade can be cast through opaque objects, since 'through' isn't being used as a physical-causal term. With this distinction, author examines 3 and concludes that confusing these concepts also contributes to the inconsistency: All shadow is shadow of something: true. All shade is shade of something: true. All shade is the shadow of something: false.

The next matter of consideration is that of silhouettes. Sorenson's example is of a church steeple that is at a distance, set high in the sky against a low sun or other light source. Here all you make out is the silhouette of the steeple. Here author examines Sorenson's argument is that when you see the silhouette, you see the steeple. Author's problem is one of equivocation. Surely the outline of the shape of the steeple is 'seeing' the steeple, but the dark interior is just the shade of the far side of the steeple, which you don't see. Sorenson would argue (given his Near/Far discussion), that you do see the far side of the steeple. Here author and he part ways and it leads to the next example, this time created by the author: the wall. Author argues that a wall that casts a shadow onto the near side of a road is revealing to you a shaded near side, not a causally-efficacious far side. If, for instance, you dropped something valuable on the near side, you would look there for it, not on the other side, where you might 'see' it!

The last part of the review discusses Sorenson's odd claim that shadows can only be dark, not colored. Sorenson creates the concept of "Filtows" to represent colored light. Author is just comfortable saying that shadows can be colored. (Sorenson is worried that things like cathedrals will end up being illuminated by shadow.)