|
Introductory
The claim that ordinary acts of looking and seeing are rooted in prejudice runs counter to the intuitions or most of us. But that is the claim I want to make. Intuitively, sight is straightforward and unmysterious. The world is there, and we see it with our eyes. How could there be any problems? How could there be any doubts? Our eyes are the very things we use to settle problems, to dispel doubt. If you don’t know, look and see. Our eyes, at least, tell us the truth where all else is guesswork: like the camera, they never lie, or fabricate. ‘I saw it with my own eyes.’ ‘Seeing is believing.’ ‘See for yourself.’ These sayings shew the extent to which we place our trust in our powers or sight. Looking to see is in fact the securest way known to us or establishing truth. But this is certainly not to say that it is infallible. On the contrary: I shall argue that seeing is far more unreliable and speculative than we usually realise. We see as much by guesswork as we do by actually receiving information about what is ‘out there’ before our eyes. And quite apart from this, seeing is a mechanical process in our brains which can go wrong like any other mechanical process, and has its own peculiar limitations like any other mechanical process. Far from accepting that our eyes always tell us the truth, we should rather ask whether they ever tell us the truth at all. First of all, the physiological process of seeing, by which our brains receive information through the eyes about the outside world, is extremely complicated. Consider what goes on in a very simple act or seeing. Figure 1 [to come] illustrates the case of a man looking at an illuminated arrow. To begin with – an obvious point but one easily forgotten – we don’t in a sense see objects themselves at all, but only the light they emit or reflect. In the case of a candle-flame, say, the light is being emitted by the object we are viewing, so what we see is an essential property of the object: you can’t have dark candle-flames, normally speaking. But in the case of a more ordinary scene, such as a plate of egg-and-chips, we are dependent on another light source, whether it is the sun or an electric light-bulb, or whatever, for seeing what is before us at all. When in normal parlance we say we can see a plate of egg-and-chips, in another sense we see the pattern of light reflected from a light source by a plate of egg--and-chips. It’s nonsense to ask what a plate of egg-and-chips looks like when it isn’t reflecting light, because things only look like anything when they are reflecting light – but the fact remains that nothing more meets our eyes than reflected or emitted patterns of light, which are one stage removed from the ‘real things’ that compose the world. I don’t of course mean to advise that we never say again ‘There’s a plate of egg-and-chips’, or the like – for obviously whatever it is that happens when we see things, it is this which, as a matter of fact, we do call ‘seeing things’. So that when I say ‘Vie don’t really see things, but only patterns of light’, this is just a striking way of pointing to the perhaps unexpected processes involved in what we call seeing. I am aware of, and have no time for, the objections of the philosopher who says ‘Anyone who says we don’t really see things must by definition be wrong.’ He has a point, but it is not germane to mine. A second point about the seeing process, evident in the same picture, is that the image cast on the back of the eye is upside- down. This is a well-known fact which doesn’t worry people particularly nowadays: but there have been those, not least Leonardo da Vinci, who thought that this fact posed a serious problem. If the image cast on the back of the eye is upside-down, how is it that we see everything the right way up? Attempts have even been made to find some mechanism in the head which turns the image the right way up again: but apart from being doomed to failure, these attempts weren’t really necessary, because it isn’t as if a sort of little man in our heads is watching the image on the retina. It doesn’t matter which way up the image is, or what other distortions it undergoes, so long as it preserves the necessary information about the object. Why this is so will become clearer when we consider what happens later in the seeing process. At any rate, let us now pass to the next stage in the process, now that we have reached the retina (from the Latin rete, a net), so called because it is made up of a network or mosaic of tiny cells sensitive to light. When an image of an object is cast on to the retina by the lens, some of the tiny cells which make up the retina are stimulated by the light, and start sending electrical pulses to the next cells in the chain which joins the eye to the part of the brain which is used for seeing. The cells which form the first layer of the retina are generally referred to as ‘receptors’, for it is they that receive the message before relaying it on towards the brain. Like all the nerve cells of the human body, they consist basically of a cell body and one or more tails, or axons, which pass a message to the next cell in the chain. Figure 2 [to come] is a schematic illustration of a nerve cell. The message which starts in the receptors in the retina gets passed on through all sorts of extremely complex networks and interconnections until eventually it reaches the back of the brain, which is the part we use for seeing. We know that this is the part of the brain which enables us to see, because if it is cut out, we can no longer see, And if it is artificially stimulated into action, we suffer visual hallucinations. At this stage in the process of seeing, when the message has travelled from the retina to the back of the brain, the miraculous ‘translation’ of the message into conscious terms occurs. No one understands how this happens, or what it involves, nor perhaps are they ever likely to, but the fact remains that when the back of the brain takes on a certain pattern of activity in response to a pattern of light playing on the retinas, then we have the experience of seeing the object or scene which is responsible for that light-pattern. It is the fact that the experience of seeing is thus apparently identical with various states of our brains that is perhaps best calculated to undermine what confidence we have in the reliability of our sight. If seeing is a mere brain process then a whole host of things could go wrong, for the brain is so delicate. Besides, many things could go differently, for the brain is only one among many possible mechanisms, some of which actually occur in other species. But even if we are not concerned about the reliability of the process, and only wish to know how it works, there are still many problems about how the brain states which are produced in us manage to give us the experience they do. Psychology might be defined as the study of states of mind, or the study of our experiences: so the psychology of perception is the study of the experiences of vision. And the main problem to which the psychology of perception is addressed is this: how is it that the presentation of patterns of light in our retinas leads to our seeing things in the way that we do? How is it that what starts off as a mosaic of messages sent from the cells in the retina ends up as our perception of an object or scene? Though perhaps we shall never be able to provide a satisfactory answer about the ‘transformation’ of brain activity into conscious mental states, at least we can deduce some of the rules or principles by which this transformation is achieved. And the most fundamental principle about seeing things, to which I shall refer constantly, is this: whenever we see anything, we are using plenty of other information besides that which is actually present at that moment on our retinas. If the brain had to decide what was going on out there purely on the strength of the pattern of light on the retinas, it would have an impossible task, as we shall see. The brain needs help in interpreting the retinal image, and it gets this help not from the eyes. In fact the pattern of light cast on the back of our eyes often provides only a small part of the evidence used by the brain to produce our perceptions (by ‘perception’ I mean an experience of seeing something). As a simple example of this compare in your imagination two people waking up on a dark morning in the same room, which has its walls covered with patterned wallpaper. One person has seen the room in daylight before: the other has never seen it before. Of course the one who knows what the pattern on the wallpaper is will see the pattern more clearly, despite the inadequate light, than the person who has no previous experience of the wallpaper. This second person may be able to make nothing at all of the vague blurs that he sees. Obviously the first man is supplementing the information on his retina from his memory. You may say that this is a special case because the viewing conditions are unsatisfactory: but in fact any situation, however good the light, suffers from deficiencies of a similar, though less severe kind. We always supplement what our eyes tell us, because if we took the image on the retina at face value, we’d either not understand it at all, or we’d be misled. To appreciate this, you only have to consider the differences between the image and what we see. The image is tiny, but we see objects with their proper size. A whole mountain panorama is represented on precisely the same area of retina as a matchbox seen close to. The image, as we have seen, is upside-down: but we see things the right way up. The image is distorted and disproportioned because of asymmetries in the design of the eye, but what we see is undistorted and correctly proportioned. The image is on the retina, but what we see is in surrounding space, outside the eyes. All these discrepancies are present even in the case of a single image fixed on the retina, even if whenever we saw a given object, there was always the same pattern of light in the same position on the retina. But normally of course the image is changing all the time, fluctuating wildly, so that the problem of how we see a stable world becomes more acute. But let’s start with a relatively unmoving image. Suppose I’m in a room talking to some people: consider the image on my retinas as I look at them. The light reflected from them forms an image on my retinas and a few tiny fractions of a second later this image gives rise to a pattern of activity in my brain, whereupon I have the experience of seeing the people. Suppose that one day – and is this not in theory perfectly possible? – we discover what the pattern of activity in my brain was when I was having the experience of seeing such a group of people. And suppose that we used this knowledge to create this pattern of activity in my brain artificially, by using electrodes, thin pieces of wire transmitting tiny electric currents. If this artificial activity was created accurately enough I would think that I was really seeing the people, though of course there wouldn’t in fact be anybody there to see. There might be no intrinsic difference between the natural and induced experiences, but there certainly would be a difference between the states of affairs that had given rise to them. Something similar to this has in fact already happened. Patients with chronic brain conditions such as depression or epilepsy have sometimes undergone brain surgery in the hope of improving their prospects. Such surgery is often preceded by exploratory stimulation of different parts of the brain to see what effects are produced, and so, what parts can safely be removed. And when this exploration enters the visual part of the brain, the patients experience hallucinations –sometimes simple ones, such as bright flashes of light, and sometimes more complicated ones, such as whole scenes stirred from memory, with every stick of furniture in place. It all depends which part of the brain you stimulate with your electrode. It is interesting to speculate how we know that we are not all permanent victims of such an experiment. I suppose we couldn’t all be, but each one of us can imagine to himself the possibility that all his experiences are an illusion produced in his brain by some malevolent brain surgeon. I’ll come back to this problem later. At the moment I’m posing problems rather than providing solutions. Another problem set by the example of my experience of seeing a group of people is the problem of the third dimension. My retinal images are two-dimensional and slightly curved and yet I see some of the people as being further off than others, and I can judge the relative distances of all the objects in the room. How is this possible? The answer to this problem is more widely known than the answers to other problems in this field. We use several sources of information, but the most important one can be illustrated by a simple experiment. Hold your two index fingers up in front of you, one at full arm’s length, and the other at half arm’s length, in a straight line between the first finger and your nose. Now shut one eye at a time, alternately, and notice the difference in the relative positions of the two fingers, depending on which eye is open. Depending on how far away something is, its positions in the two retinal images will be more or less different, and the brain uses this variable disparity to compute the distance of what we see. We shall have more to say about this later. So our principal method of seeing distance is quite straight- forward. Even so it is perhaps strange that the third dimension is not something we consciously deduce. It really looks as if it’s there, without our thinking about it. There’s a special kind of appearance a thing has when it’s a certain distance away, despite the fact that the image it casts on the retinas is as flat as a pancake. A difference between two flat images is translated into perceived distance. This is one of the more remarkable feats performed by our perceptual mechanism. There are some other more general problems about seeing which I’ll mention now; all of them are relevant to my contention that we always go beyond the information provided on the retinas in order to see things. First of all we manage to recognise things, and indeed things continue to look much the same to us, even though they probably never give us quite the same retinal image on two separate occasions. Take someone you know well. Each time you see him the size of the image he casts on your retinas is probably different, because you see him at different distances. As you know, the size of the retinal image varies inversely with the distance of the object casting it. And yet your friend doesn’t look a different size each time you see him. And you recognise him without any difficulty, even though the retinal image on each occasion may be quite unparalleled in your experience. Consider a coin: as you tilt it in your fingers the retinal image changes from a circle to an ellipse, and eventually into just a line or bar: then the process is reversed. But the coin doesn’t look as if it’s changing shape. Consider a piece of white paper. In the sunlight it reflects about a hundred times as much light as it does in an electrically lighted room – and yet it doesn’t seem to be of a different intrinsic brightness in the two situations, though of course we can notice the difference in the degree to which it is illuminated. Once again we take into account more than just the retinal image – otherwise the paper in the sunlight would look whiter than it does in the room. Consider movement. Look at an object in front of you and move your eyes from side to side. The image of what you are looking at is now travelling to and fro across your retinas: but the object looks as if it’s staying still, not as if it’s moving. Now pick the object up and move it from side to side in front of you, keeping your eyes still. Perhaps exactly the same motions as before are gone through by the image – but this time the object looks as if it’s moving. Why exactly didn’t the stationary object look as if it was moving when we moved our eyes in front of it? A clue to the answer to this question is to shut one eye and move the other eye-ball with your finger. This is not altogether easy to do. Suppose you have shut your left eye. Now press gently with your index finger on the right hand side of your right eye-lid, keeping the eye open, until the scene before you appears to shift substantially to the right. Then keep your right eye in its new position, and with your left hand reach out quickly and try and touch some object in front of you – a book, say – at a particular point on its surface. Did you touch the point you aimed at? Another problem that shows how we supplement subconsciously the information in the retinal image is the problem of ambiguous figures. These figures illustrate clearly the fact that even though the retinal image may stay precisely the same we can still see it in more than one way. This happens all the time but with special figures we can illustrate it particularly dramatically. These pictures fall into three classes: ambiguous pictures; mystery pictures; and photographs of familiar objects from unfamiliar angles or distances that ‘suddenly make sense’. Here are some examples of all three kinds of pictures: [to come]
What all these pictures do is as it were to catch the brain in the act of supplementing the retinal information. Normally this supplementation happens so fast and so automatically that we hardly believe it has happened at all. But in these special cases we can watch it in action, and so it is easier to believe in in general. Consider too printed words. These cast the same retinal images in the eyes of a child as they do in the eyes of an adult. And yet the child makes nothing of them; hardly even notices them, perhaps. But to the adult they are significant. Of course you can say that the difference is that the adult knows how to read, but that doesn’t explain what the adult does with his retinal stimulation that the child doesn’t do with his. To summarise what has been said so far, we could say that when we see something we take into account, subconsciously of course, the retinal image and other things besides. (These other things are very various indeed, but at this preliminary stage it is convenient to lump them together.) If we didn’t take these extra things into account, we couldn’t see the third dimension; we couldn’t recognise things unless they were always seen at the same distance, at the same angle, and in the same illumination; a stationary object would look as if it was moving as we moved past it; and there would only be one way of seeing ambiguous or trick figures. We must now go into this in more detail. |