Counterfactual Empiricism

Counterfactual Empiricism

Classical empiricism is the doctrine that all knowledge is knowledge of impressions: you know what, and only what, you have perceived with your senses. This doctrine runs into trouble with things you know but have not perceived—concepts of the unperceived. It would appear to imply that you can’t have concepts of, and knowledge about, atoms and bacteria (among other things). What I am calling counterfactual empiricism seeks to avoid this problem while remaining true to empiricist principles. It says we can have knowledge about things we can imagine seeing (hearing, tasting, etc.). I can imagine seeing bacteria and atoms, since they are just smaller versions of what I actually see all the time (or so we can suppose). I know what it would be like to see an atom or bacteria. Hence, I have the concept and can grasp the meaning of the terms. The doctrine, then, is that all knowledge is of actual or possible impressions—what I would see if I could see. Knowledge is of what I can envisage seeing. This seems an improvement on the classical, more restrictive, doctrine. In what follows I will confine the discussion to seeing and ignore the other senses, though the same points apply mutatis mutandis. I will be particularly concerned with the concept of mind.

Counterfactual empiricism is not free of difficulty: for there are apparently many things that cannot be seen, no matter how counterfactually. A partial list would be the following: numbers, values, universals, logical forms, propositions, causation, necessity, infinity, selves, and minds. Do you know what it would be like to see any of these things? Can you envisage (imagine) seeing a number, an ethical value, a Platonic form, a logical form, causal or logical necessity, an abstract proposition, an infinite collection, a self, a mind? These don’t seem like the kinds of thing anyone could see, and hence imagine seeing. But perhaps we can partially imagine seeing them; some sort of sensory content is associated with them. Perhaps we can imagine seeing a number by imagining seeing a numeral (we actually see those). Perhaps we can have a mental image of the color red considered in itself. Perhaps a logical form looks remarkably like a written sentence. Perhaps necessity looks rather like a man pushing a cart. Perhaps a self looks like a glowing point. All these seem pretty farfetched, though not entirely devoid of intuitive appropriateness. Perhaps exceptionally imaginative people have quite vivid images of such things; they may say they know just what these things would look like if they could be seen. And wouldn’t this be a valuable form of knowledge—knowledge of how thing would look if they could be seen? Even if it is impossible that they could be seen, this is how they would look if seen; we can at least imagine that. Can’t God imagine what a number would look like if it were seen—even if he could never see it? It is possible to imagine impossibilities (Escher drawings). True, we cannot imagine nonsense—we can’t imagine seeing colorless green ideas sleeping furiously. But it isn’t nonsense to suppose that numbers can be seen—just metaphysically impossible. It isn’t contradictory. Nor is it like envisaging seeing slithy toves gimbling in the wabe. There can be difficult cases in which it is hard to decide whether something can be imagined being seen.

Could the mind ever be seen, and could we imagine this? It can be introspected, but could it be seen by the visual sense? If so, what would it look like? Consider this question: could sounds be seen, and if so, what would they look like? Well, we could clearly hook up sound waves in the atmosphere to our visual cortex and produce visual impressions as a result of auditory stimuli. The auditory inputs simply recruit the optic nerve and project to the visual cortex: the sound waves thus cause visual sensations. But could these sensations have a content that correctly represents the nature of sound waves? I don’t see why not: the sensations are as of waves in a medium, somewhat like seeing ocean waves. Phenomenologically, the sensations are not like ordinary auditory sensations, but they are veridical (as auditory sensations are not), since sound waves arewaves. Sound waves can be experienced with the ears and they can be experienced with the eyes; it is possible to see them, and possible to imagine seeing them. You might get hooked up tomorrow. We know what it would be like to see sound waves, even though we never do. The question I want to ask is whether the mind is like that: we never in fact see the mind, but could we? Could we be hooked up that way—and if we were, what would the mind look like? Someone might say there is no problem, since the mind is the brain and we can see the brain, even if we never do in fact. We could hook the brain up to the visual system in such a way as to produce visual impressions of the brain (or taste and smell impressions come to that). But that is not interesting and presupposes materialism. So, what would the mind look like under immaterialism? It might be said that it would look like a river, because consciousness is like a river; but this too is uninteresting and still too materialistic. Would it look like a cloud? Closer, but still wrong. Would it look like a fireworks display? Would it look like the Sun, or a starry night, or a kaleidoscope? Hmmm. Perhaps we can be more specific: what would a belief look like if it could be seen? Here we might envisage a kind of bridge between one thing and another, corresponding to a believer and a proposition, with the proposition reaching out to a distant landscape.[1] That doesn’t sound too far off; it gives one hope. What about Martians with superb imaginations that can fill in the details and produce an impressive visual impression of a belief? What if we really worked on it, recruiting top brain scientists? It doesn’t sound out of the question. After all, the mind is an empirical entity with definite characteristics, so all we need to do is attach a sensor to it that feeds into a visual system—and (voila!) visual impressions of a mind. We would see thoughts, pains, reasoning, emotions, etc. Yet we are very far from doing this—very far indeed. This is a kind of knowledge of the mind that we are nowhere near possessing, though apparently possessable. The mind does look a certain way if you have eyes that can see it. In fact, it now has a certain look—how it would look if you could see it (which in principle you could). It is disposed to look a certain way, where this disposition is grounded in its intrinsic nature. It has, as it were, a certain shape or form, a certain mode of perceptual presentation. Right now, it is in possession of such a mode of presentation, in addition to its introspective mode of presentation. We can’t imagine how it would look now, except very rudimentarily, but there is presumably a fact of the matter. There might even be a preferred way it should be seen to maximize veridicality.

Now I want to connect these reflections to the mind-body problem: would we be nearer to solving that problem if we knew what the mind looks like? We would have a visual representation of it, which we signally lack now. This might be brought together with our visual representation of the brain, so that they can be placed side by side. As things stand, we have no such opportunity; but if we had conceptions of mind and brain of like nature, we might be in a better position to solve the mind-body problem. Indeed, we might already have solved the problem. We might say, “Oh, now that I know what the mind looks like, I can see how it relates to the brain!” When we had no vision-based conception of the mind, we were at a loss to see the connection, but now we can compare like with like and see (literally) how it all fits together. It would be like comparing macro conceptions of matter with micro conceptions, both of which are vision-based, and seeing no problem of emergence. The problem is that we don’t have a vision-based conception of the mind; hence, the sense of mystery. The mystery might disappear if we had the right conception of mind, i.e., one continuous with our conception of the brain. Or at least we might have a sound basis to reason from, do experiments, etc. If this is so, we now know what it would take to solve the mind-body problem—a visual representation of the mind. And wouldn’t this solve the other minds problem at the same stroke, given that we could actually see the other’s mind not just infer it from behavior? If counterfactual empiricism were true of the mind, then we would be within reach of solving these problems. Of course, we are very far from being able to see minds or even to imagine what it would be like to see them. Classical empiricism always had problems accounting for our mental concepts (“ideas of reflection”) because we don’t sense our minds with our five senses; it would have done better to move towards counterfactual empiricism, accepting that our concept of mind is a work in progress waiting upon advances in brain science and a cure for mind-blindness.[2]

[1] Another example would be mental modularity: would the modular mind look compartmentalized if we could see it? Would it look a bit like a Lego set or a car engine? Would we see the divisions and distinctions in the modules where now they are abstractly understood? If we saw intentionality, would it look like a wire or a laser beam? Would pain look like a jagged edge or an erupting volcano? Or would these things look like nothing we have ever seen before but still look like something?

[2] I am well aware how far out on a limb I have gone in this paper—seeing the mind indeed! But in philosophy (and sometimes science) out on a limb is where you want to be—that’s where the best fruit is. In some possible worlds it might seem extraordinary to people, given their senses, that the mind is not perceptible to some unfortunate souls. In this world children draw pictures of the mind and there are photos of it in books. For these perceptive people we are lamentably mind-blind.