In the philosophy of perception, objects are typically frontlit. But according to Roy Sorensen, backlit objects have surprising lessons to teach us about perception. In backlit conditions, ‘the principles of occlusion are reversed’, Sorensen (2008, p. 25) maintains. In particular, he claims we see the back surfaces of backlit objects. But as I argue in this paper, Sorensen’s arguments in support of those claims are flawed. After criticizing Sorensen’s arguments, I attempt to address a residual puzzle about backlit objects. The upshot of the paper is that cases of backlighting do have important philosophical lessons to teach us after all – lessons concerning the notion of perceptual occlusion.