Minds on Monday: Molyneux's Question and My Question

Imagine that two electrodes are placed on your skin, perhaps on the back of your hand or on your forearm. Now, one is electrified weakly, so it produces a faint tingling sensation. Or maybe it's heated up so you feel a hot spot pressing against your skin. You see two electrodes, but feel only one tingle or hot spot. Can you point at or visually describe (e.g., “the one on the right”, “the red one”) the live electrode? Let me call this My Question. (You, Respected Reader, can call it “Mohan’s Question”.)

My Question is interestingly related to William Molyneux's famous Question.

Suppose a man born blind were able to distinguish between a cube and a sphere by touch. Suppose that these objects were placed on a table and the man made to see. Quaere: Whether by sight, before he touched them, he could tell which was which?

A number of positive responses to Molyneux’s Question assume a positive response to My Question (asked concerning the newly sighted man). H. P. Grice imagined the man tracing the outline of the sphere with his finger in the way he remembered doing before being able to see. (This involves kinaesthesia, but we don’t have to worry about this, since he used kinaesthesia in his pre-sighted days.) He can see that his finger cuts inside the corners of the cube, Grice says. So he knows that the visible cube is not the tactual sphere. But note: only if the seen position of the finger can be identified with its tactual/kinaesthetic position.

Let’s get rid of the complication of the newly sighted man. Suppose that the answer to My Question is ‘Yes’. You can visually identify the live electrode. How do you do it? An empiricist might say: you can only do it by correlating tactile space and visual space. But another answer is possible: you can only do it by identifying a single point in space in two different ways. For one might reasonably insist that there is only one space, and only one point occupied by the tip of the live electrode. But there are two sensory representations of this single spatial point.

Gareth Evans takes a One Space approach to the Molyneux question. He imagines a philosopher named 'V' (for Evans?) who insists that there is only one concept of sphere—not separate tactual and visual concepts. (I think Descartes would have held this view: think of what he says about the chiliagon. I'll return to this in a moment.) Evans thinks that the One Space interpretation implied a Yes answer to Molyneux’s Question.

I don’t think Evans was right. One could have a single definition of the concept sphere: a solid such that there exists an interior point from which all boundary points are equidistant. (This is the kind of definition that Descartes liked.) One would need to learn how to tactually identify things that satisfy this definition (which, in conformity with Descartes' demands, contains no tactual element). And learning how to visually identify things that satisfy it might be a separate process. If so, the answer to Molyneux’s question would be No.

But are points of space really identified separately by the different senses? Consider the ventriloquist illusion: you see the ventriloquist dummy’s mouth move while the ventriloquist himself speaks without moving his mouth. Irresistibly, you hear the sound coming from the dummy. Why? Because your perceptual system resolves the conflict between visual evidence and auditory evidence in favour of vision. Vision dominates, at least with regard to location. But this suggests that there is a common representational system: it’s not that vision represents the moving mouth at visual place VP and audition represents the voice as at auditory place AP. Put that way, there is no more conflict here than in the normal case. Rather vision represents the moving mouth at place P and audition represents the voice at some other place P’—and the conflict is resolved in favour of vision.

Evans acknowledged that there is a single representation of space that underlies all of the different senses—this representation is in terms of bodily actions, he thought. Berkeley thinks about the tactual space in terms of bodily actions: Evans thinks that Berkeley goes wrong because he thinks that the other senses do not represent space in bodily terms, as touch does.  Evans’s philosopher V doesn’t in fact think of a sphere in the way Descartes does, i.e., in terms of the geometrical definition given above. He thinks of it as a certain configuration in bodily action space.

There is a certain tension in Evans's position (or rather in V's). He thinks that there is only one concept of sphere, not one for vision and one for touch. Despite this, he thinks that each sense-modality has a separate representation of space: "I assume that the apparent direction of a sound is a phenomenal property of the perception of the sound." But he softens this position by proposing that these separate representations are commensurable because all are in terms of bodily actions.

I am unpersuaded. I prefer to think of things in the Cartesian way. When you point at the live electrode, your action does not constitute the location of the electrode; it is guided by the cross-modal (or perhaps amodal) representation of its location. Arguing for my position is not something I can do here. But let me put it on the table.