A post on Molyneux’s question that I wrote for this blog back in 2011 has since blossomed into a full-blown article on the subject, which was published in the most recent issue of Analysis (for an ungated copy, see here). Here is the abstract of the article:
Do spatial features appear the same whether they are perceived through vision or touch? This question is at stake in the puzzle that William Molyneux posed to John Locke, concerning whether a man born blind whose sight was restored would be able immediately to identify the shapes of the things he saw. A recent study purports to answer the question negatively, but I argue here that the subjects of the study likely could not see well enough for the result to have been meaningful. I then propose a way to improve the study, by including cues from object motion.
Along with a more careful discussion of the logic of Molyneux’s puzzle and the form a study would have to take in order to answer it, plus considerable attention to how the puzzle relates to various empirical studies of perceptual multimodality, one of the main things that is new in this article is the point I indicate toward the end of my abstract and in the title of this post (reference here), namely that motion plays a role in visual shape perception that is not dissimilar to the way that turning an object around in one’s hands (or moving one’s hands around it) enables one to perceive its shape through touch. (I am indebted to a conversation with Mohan Matthen for helping me to get clear on the importance of this point — though evidently I also could have paid more attention to Merleau-Ponty.) As I write, concerning the fact that the study in question presented motionless stimuli and permitted subjects only “to adjust their distance or viewpoint while remaining seated in front of the presentation table”:
In everyday life, the visual perception of complex three-dimensional shapes is often quite unlike this: just as you perceive a shape through touch by running your hands all around it, so visual shape perception frequently involves more than just looking at an object from a single angle. Despite the ‘simultaneity’ of the spatial information available to vision, watching an object as it moves with respect to one – or as one moves with respect to it – can bring into view features of it that would otherwise have been hidden, and makes available more information concerning how its surfaces are oriented in depth. (Indeed, in this respect visual shape perception is actually less simultaneous than is the perception of shape through touch: you may see something from only one side at a time, but touch an object from several angles at once.)
This is, of course, an observation that complicates things quite a bit, since any study that can resolve Molyneux’s puzzle must deal with naive perceivers, whereas the ability to integrate cues from object motion may have to be learned. But there is some experimental evidence, which I cite in my paper, that challenges this assumption. In any event, I think it is pretty clear that any adequate study of the visual capacities of the newly sighted will have to take object-motion into account, if only to show that the puzzle can never be anything more than a philosopher’s thought-experiment, since the only way a person can be “made to see” is by exploring the world actively through touch as well as sight.