Abstract
It is widely accepted that pictures can only depict visible things. The paper criticises this ‘visibility constraint’ on the objects of depiction. The constraint is shown to imply that the range of visibilia is settled prior to an investigation of what can be seen in pictures. By contrast to this, I suggest that settling what can be seen in pictures is relevant to settling the range of visibilia. It is what we experience in pictures, and not the objects of depiction, that is subject to a visibility constraint, I propose. After criticising the widely accepted visibility constraint in the first half of the paper, I outline an account of what we experience in pictures in the second half of the paper.