Berkeley: The Central Arguments 
by A.C. Grayling.
Duckworth, 218 pp., £19.50, January 1986, 0 7156 2065 7
Show More
Essays on Berkeley: A Tercentennial Celebration 
edited by John Foster and Howard Robinson.
Oxford, 264 pp., £22.50, October 1986, 0 19 824734 6
Show More
Show More

As a child I collected butterflies, and I remember being impressed by a comic cartoon which showed another collector, older and more experienced than myself, who had accidentally swallowed a specimen he had been chasing. Later I felt the same sense of incongruity when I read Berkeley’s claim that everything he perceived was really in his mind. Surely he was overdoing it. True, his was only a case of mental ingestion, and anyway the butterfly would not be taken in by a single act: first, the blue of its wings, then with more difficulty their shape and size, and finally even the grainy arrangement of their scales which would only show up under a microscope. But how could he do it? His portrait does not show him with his hand over his mouth and an expression of dismay on his face like the man in the cartoon. On the contrary, he looks like someone who is happy to have made his point.

The tercentenary of his birth is a good moment to review his achievement. His claim is that nothing exists when it is not being perceived and that, when it is being perceived, it exists in the mind of the perceiver. This suggests that the butterfly clicked into existence when the collector first spotted it. But if this is why we find Berkeley’s claim incredible, we must think again. He also tells us that, when none of us is looking at it, it is sustained in the mind of God, like the tree in the quad and everything else.

There is a conversation at the beginning of Forster’s novel The Longest Journey which shows how important it is to add ‘and everything else’. If we forget this addition, it will look as if the butterfly would be in trouble without God’s sustaining mind, unable to cover the hundred yards between one collector’s visual field and another’s. But if the butterfly would be in trouble, so too would the hundred yards. Berkeley’s idealism is total. He is not telling us that we are located at different points in physical space, where God sends each of us the perceptions appropriate to that particular seat in the theatre. Everything is ingested, including space itself.

Perhaps this explains Boswell’s verdict: ‘Though we are satisfied his doctrine is not true, it is impossible to refute it.’ Maybe it is because the doctrine is total that it seems to be irrefutable. Any consideration that can be brought against it, or any piece of apparently contrary evidence, will be accepted by Berkeley only after it has been pre-digested by his theory. If he can make anything assimilable by the mind, why not everything? On the other hand, the truth of the matter can hardly depend on which side carries the onus of proof. Is it then a stand-off?

It is certainly a disappointment to those who expect a philosophical controversy to be settled like a scientific one. It also makes them feel that philosophy is not a serious subject. If Berkeley’s rejection of matter does not mean that we can walk through physical objects, how else can it be meant seriously? To take another example which gives practical people the same impression of frivolity, Zeno tells us that an arrow cannot traverse an infinitely divisible distance in a finite time, because it must first cover half the distance, then half the remainder, and so on. Are we then to suppose, like the character in Stoppard’s play, that Saint Sebastian died of fright?

What the two cases have in common is that the philosophy cannot change the ordinary beliefs. However, Zeno presumably was not trying to change our belief in motion, but only to challenge us to reconcile it with the infinite divisibility of space. Berkeley was not trying to change our belief in the world. He did not even regard his doctrine as in any way paradoxical. On the contrary, he thought that it agreed perfectly with common sense, and it is only the unconverted who find it paradoxical. Is this because they misunderstand it?

If so, the opportunities for misunderstanding are reduced by two tercentennial publications which present his system with admirable clarity and explain how he worked his way into it so completely. One is a comprehensive account of his philosophy written in order to demonstrate that his main claims are more defensible than is commonly allowed. The other is a collection of new articles which assess his arguments and interpret his conclusions with the sympathy and care which they deserve but have not always received.

There are philosophies which can be divided and accepted in part, but Berkeley’s is not one of them. It is extraordinarily unified and he could almost claim that a sympathiser who had not gone all the way with him had not gone any of it. If we swallow colours but strain at spatial properties, or swallow spatial properties but draw the line at microscopic ones, we cannot be doing this for his reasons, which do not admit fellow-travellers. His idealism is a philosophy of total ingestion and any refusal amounts to a repudiation of it.

This is because it is perceptual idealism rather than conceptual idealism, evolved from his views about sense-perception rather than his views about thought. A conceptual idealist, in the Kantian tradition, would argue that our patterns of thought make an essential contribution to the world as we know it, and, though it is possible for this philosophy to be taken all the way to the extreme conclusion – that the mind contributes everything – this is by no means a necessary consummation. Kant himself did not go so far. But Berkeley based his idealism on the consideration of sense-perception, and if he had stopped half-way, with Galileo or Locke, and left some of the properties of things outside the mind, he would not have been a perceptual idealist.

Contemporary philosophers of perception are unlikely to go to Berkeley’s extreme. They look back at him in awe because they see him as an ancestor of heroic stature who would have had to be invented if he had not existed. However, he did exist, and his history sets a limit to mythology. It is a limit we often fail to notice today. We adopt him and unconsciously project onto him our own tendencies to perceptual idealism, instead of giving an accurate account of the way in which he actually achieved it. Grayling and several of the contributors to the anthology explain how this happens. When modern philosophy cuts off our view into the world, it usually does so by arguing that we do not see the butterfly’s wing because we are only directly aware of our own visual sense-data, and similarly with the other four senses. Russell takes Berkeley in this way and criticises him for confusing the act of perception with its object. But this is not an accurate interpretation of Berkeley’s position, because it credits him with a view of perception which may be ours but certainly is not his.

His view was one that could hardly have been formulated in this century. He began by arguing that the perceptible qualities which the materialist attributes to physical objects are all in the mind of the perceiver. According to him, they have to be in the perceiver’s mind, because, if they were not, there would be no perception. So far, this is more or less what Russell says about colour-vision. The divergence comes at the next step: Berkeley did not go on to say that perceptible qualities belong to things in the mind, like Russell’s sense-data. If ‘belong’ means ‘qualify’, his view was that they do not belong to anything. If, on the other hand, ‘belong’ means ‘depend on’, then, because they are not independent, they must belong to something. Now experience shows us that they depend on, and so belong to, the perceiving subject. Therefore, what is essential to their existence must be being perceived: esse est percipi. Consequently, when the materialist protests that the colour blue also belongs to the butterfly’s wing, Berkeley treats this as an absurdity, not just because it takes the colour twice over, but because the colour could not possibly belong to the butterfly’s wing. For the only belonging of which we have any experience in this case, and so the only belonging which we understand, is belonging to the subject who perceives it, and that kind of belonging is not a possible relation between the colour and the wing.

It is instructive to compare this argument, which took Berkeley all the way to perceptual idealism, with the argument which in this century has taken some people some of the way with him. Berkeley refuses to allow that there are any mental particulars to be qualified by perceptible qualities. This gives his argument an apparent advantage over the modern argument, because it makes it more catastrophic. It is a short argument relying on a swiftly adminstered shock. Even if the shock is not instantly persuasive, it creates a situation in which it is likely to become persuasive without anything more being done. This effortless self-authentication is exactly expressed by Berkeley’s irresistibly lucid prose: it seems as if the decisive move was made before he put pen to paper. John Austin borrowed a phrase from Aristotle’s Poetics to describe this effect: ‘events outside the tragedy’. If we want to get off this stage, how are we to do it? The ladder seems to have been kicked away.

The parallel argument current today hustles us back into the mind less precipitately and less irrevocably. The suggestion that perceptible qualities qualify sense-data in the mind allows more room for manoeuvre. At least we can agree without committing ourselves to the dogmatic thesis that the only possible kind of belonging is a kind that could not relate a property to a material object. True, we are going to be in trouble if we admit that the only objects of which we are ever directly aware are sense-data, but still it will not be contradictory for us to suppose that perceptible qualities, or some of them, might also qualify things outside the mind. The door of the mental prison will not be double-locked.

Although the second lock strikes Berkeley as an advantage, it may be one that he cannot take and keep. If he had not tried to take it, could he still have made his case? This question is more philosophical than historical, but it is none the worse for that. If we want to know whether Boswell’s verdict was the right one, we need to see how Berkeley would fare on a more open field of controversy rather than entrenched in a dogmatic position. This does not require much historical imagination, because in the 18th century Hume saw the problem in much the same way that we see it today.

One way of resisting perceptual idealism is to reflect on the implications of Berkeley’s refusal to locate us at different points in physical space, where God sends us perceptions appropriate to each point. That had been an idea of Malebranche’s, which Berkeley refused to accept. But could he really manage without physical space? If Wittgenstein’s critique of sense-datum language is correct, it provides a framework which is indispensable.

Wittgenstein puts pressure on the theory of the detached phenomenal world at two points. First, if the detachment is complete in my case, how can I even get the idea that I am one finite mind among others? If I turn my attention inwards, I find no trace of a perceiving subject, and I am not in a position to argue that my body must be the one which is located at the place from which what I see would by the laws of optics be seen. The point is not that I am a body rather than a mind, but that, if no use is made of my body, I have no criterion of identity. Berkeley would protest that I can identify my body in his system, because it has been mentalised as a particular set of perceptible properties. But that is not good enough. What needs to be explained is not what my part of Berkeley’s system is like after my mind has swallowed my body, but how my part of his system can possibly have been set up originally without any use being made of my physical location in space.

Second, if the detachment of my phenomenal world is complete, how am I to identify the various perceptible qualities of my sense-data? In life as it is lived, I first acquire language at the interface between myself (mind and body) and my surroundings. Berkeley’s theoretical withdrawal into the phenomenal world would leave me without the necessary criteria of identity for discriminating perceptible qualities. Anything that seemed to me to be thus and so would simply be thus and so. I would have nothing firm to push against in my first efforts to acquire a language. Again, it is no good Berkeley protesting that in his system the action at the interface between me and my surroundings can be described in mentalised terms. What needs to be explained is how this part of his system can possibly have been set up originally without any use being made of my independent physical environment.

These are modern arguments, but a philosopher’s place in history does not give him diplomatic immunity in our century. They are arguments about language, but if language is subject to these constraints, so too is thought. They are arguments designed to show that Berkeley worked his way into his system without the concepts that he needs when he is in it. They also show us how the conceptual loss actually occurred. It occurred because he hustled us too quickly off the middle ground of daily life, where we do not find any difficulty in our dealings with the two worlds, physical and phenomenal. He should have taken the transition more slowly, and then he would have observed the gradual depletion of his system. This is the crucial point, and it can be used to exhibit the connection between modern and 18th-century criticisms of perceptual idealism.

Anyone who reads Berkeley’s account of the mentalisation of apparently physical qualities will be struck by his assumption that structural properties do not give him any special difficulty. He starts with the colour of the butterfly’s wing, and it seems fairly easy for him to transfer it to the mind. Of course, he is using the dogmatic argument that it is impossible to attribute any perceptible qualities to physical objects. However, it is important that our acquiescence is largely based on the fact that we know that the physical description of the cause of our seeing the wing as blue need not involve any attribution of the colour blue to it. But what about structural properties, like spatial distance or temporal interval? Do we also believe that the movement of the butterfly in physical space can be described in a similarly impoverished way? Or the speed of its movement, which is also a function of lapse of time? The trouble is that in these cases the impoverishments would reduce the physical assets of the flight of the butterfly to zero. Now Berkeley would not be perturbed by this, because he is using his dogmatic argument to show that no perceptible properties – not even structural ones – can possibly be attributed to physical objects. But at this point we are likely to lose the motive for our acquiescence. How can structural properties be dragged into his system in this way? Perhaps what is happening is that he is only pretending to have brought them in by continuing to use their names.

There is another way of keeping our grip on the difference between structural properties and what may be called pointilliste properties, which is something that is worth doing if we want to appreciate the strength of Galileo’s type of scientific realism. Many fast-moving animals push out ahead of themselves the cone of a long distance sense. This gives them an array of points separated by distances which we call ‘angular’, because that is how the creature uses them when it moves ahead in one direction rather than another. In our case, the sense is sight, but in the case of a bat it is sonar. Now there could be a fairy-tale in which a bat and a philosopher compared notes. They would soon agree that the pointilliste properties of their arrays need not be attributed to the physical objects that caused them. In fact, the bat would go further and say, in the spirit of Berkeley, that no meaning could be attached to the attribution of an echo-interval to an object. But they would be impressed by the similarity between the structures of their arrays and nothing would induce them to treat them in the same way as the pointilliste properties. Of course, Berkeley would not be refuted, because he could still rely on his dogmatic argument. But is that so important?

The suggestion that Berkeley’s system failed to do justice to structural properties was developed in the 18th century by Kant, but not in this simple Galilean way. Kant was a limited conceptual idealist, and it was from that point of view that he criticised Berkeley’s perceptual idealism. Still, his criticism was that it is a system too impoverished to achieve the objectivity needed to make it agree with common sense and science. The anthology contains a very good article by Ralph Walker explaining this development. Time, space and causation all suffer under Berkeley’s phenomenological pointillism. Maybe this would not have satisfied Boswell, but there are philosophical arguments which are best countered not by head-on confrontation but by an outflanking movement.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 8 No. 14 · 7 August 1986

SIR: David Pears (LRB, 5 June) identifies Berkeley’s philosophy as a form of ‘perceptual idealism’. This, Pears states, leaves Berkeley quite unable to give any account of himself as a ‘finite mind among others’, and the problem is then traced to Berkeley’s attitude about physical space. The underlying dispute between Pears and anyone who takes Berkeley seriously is about what must be counted. Pears insists on objects and physical space; Berkeley’s analysis requires only reliable predictions. To ask therefore how the flight of the butterfly may be traced does not demand a preliminary account of the structural properties of butterflies, unless the question is about the structural properties of butterflies. Like every other observer, Berkeley’s starting-point must here be the colours seen. Their several relationships can then figure in a descriptive system determined by the questions which follow. Berkeley’s ideas remain the sine quibus non of any observation claims whatever.

Désirée Park
Wolfson College, Oxford

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences