Many people in Darwin's day naturally shared his dilemma about how to see the world as a whole realistically. As time went on, they increasingly felt that the supernatural element might be an unnecessary extra, a frill added to the natural world, a wish-fulfilment springing from human childishness and self-indulgence. The supernatural seemed, as Stephen Jay Gould lately put it, ‘the representation of raw hope gussied up as rationalized reality’.¹ The moral issue was then clear. Religion—meaning Christianity—was an addiction that honourable people must break. And if that left the world without meaning, then meaning must be somehow rebuilt or finally dispensed with.

In attempting this rebuilding, sages made much use of science. The beauty and order of the Newtonian universe seemed to offer a new home to exiles from the traditional Eden. It gradually became clear, however, that this beauty and order might not be any more secure than what they had replaced. Only in patches is the order and beauty of the world directly visible. To believe in it as a whole requires faith.

Up till now that faith—that conviction of a universal order—had been backed by and expressed in belief in God. Without that, what remains is just the conviction of scientists that the world must finally conform to science—that doubts and confusions will eventually give way, revealing underlying order. But might not that faith too be mere wish-fulfilment? Is there any real guarantee that the partial order we have seen so far is not a misleading varnish on hidden disorder? Or that, even if it has been real so far, it will not stop next week? Or that it covers matters we have not yet experienced? Induction makes us expect continuance, but what justifies induction? As Russell pointed out, ‘The man who has fed the chicken every day throughout its life at last wrings its neck instead, showing that more refined views as to the uniformity of nature would have been useful to the chicken.’²

Darwin's story illustrates the very selective way in which what we may call heroic Enlightenment puritanism works. Not all wishes get mortified, only certain chosen ones. Puritanism—when it is used as a term of criticism or abuse—means objecting to pleasant or easy things merely because they are pleasant or easy. This can lead to counting things as temptations which ought not to be so counted.

In the case of thought, puritanism calls on us sternly to avoid wish-fulfilment—to abstain from accepting any belief merely because we would prefer it to be true. But, as many people have pointed out, we do prefer to believe the world to be intelligible. Have we a right to indulge that preference?

There are two ideas here which we might like to find true—first, that the world is in fact ordered, and second, that that order is accessible to us, which is naturally taken to mean that it is an expression of a mind like our own, but much greater. The first idea has its own problems, as we shall soon find. But it is the second that really needs attention. Its critics often assume, as Gould does, that this is obviously just a piece of wish-fulfilment, a couch for the lazy-minded. In more hierarchical ages, that may well have been true. But moral and political taste and imagery have changed profoundly in the last three centuries, and it is no longer true today.

In our current individualistic climate, the idea of having an authority above us who always knows best is far from welcome. The virtues we are taught to revere most are no longer ones appropriate to subordinates, such as patience, loyalty and obedience, but ones fit for solitaries or rulers, like autonomy, independence and moral courage. (Of course these are officially now meant to fit us for life as equals, but that is something of a pious hope.) We find it hard enough to relate to directors set above us here on earth; it is far harder to know what to do with a heavenly superior, even a very remote and abstract one. That change of moral temper, and not any scientific discovery, seems to me the root cause of the modern estrangement from traditional religion.

Changes of moral tone like this are, of course, never complete; they are matters of emphasis, which is just as well. Both for political and private purposes, both these attitudes are always needed, and are always present. Complete individualism would be as unworkable as complete conformity. Yet changes of emphasis do have a marked effect, and the upvaluing of individual pride is evidently making trouble for us in accepting a subordinate cosmic status that did not trouble our less assertive ancestors at all. They simply did not feel the need to think of themselves as ranking first in the universe.

This is surely the change that has called for a general massacre of the supernatural, a purge of the wide and varied fauna that normally inhabits the human imagination. The root cause is not any shift in the nature of the explicit arguments. These remain as vast, difficult and confusing as ever, much hampered by the lack of an adequate language to express them. They are not what really determines people's faith. Neither, however, is the cause of change a moral reform of the kind usually suggested, namely a sterner mood of resolution among scientists and other sages, enabling them to cast off chains that earlier generations dared not shift.

The position is not that theists such as Newton and Galileo were too ignorant and childish to see what was obvious to T.H. Huxley and Bertrand Russell, nor that they were too dishonest to admit it. Neither is it that new facts, discovered since their day, have put their theistic attitudes out of date. (They would not have been among the people who were surprised when the first Soviet astronauts failed to find God in outer space.) It is true, of course, that their views were influenced by the moral and political climate of their times. But then, so are ours.

Contemporary climate always makes a great difference to what counts as a temptation. It may indeed be characteristic of small children to want a familiar system and to depend on having someone above them. It is also characteristic of adolescents to throw off that system and to protest against those guardians. Moving from the first state to the second may indeed be progress, but it is not yet maturity. Mature people are supposed to be relaxed about questions of rank and to look for such system as suits the general good, whether familiar or otherwise. A passionate conviction that there cannot possibly be a mind superior to one's own is certainly not something that can be drawn from science. It seems to me much more like what Gould briskly calls the representation of raw hope, gussied up as rationalized reality.

If, however, some hope is raw, might there be other kinds of hope that are cooked, processed, somehow tested and found reasonable? We do, as I pointed out, naturally hope that the world is orderly. We like it that way. As we have seen, this idea of a basically ordered world is even one which, today, may be very important to us emotionally, may seem an important aspect of our salvation. All of us, including those ignorant of science, find this idea sustaining. It controls confusion, it makes the world seem more intelligible. But suppose the world should happen in fact to be not very intelligible? Or suppose merely that we do not know it to be so? Might it not then be our duty to admit these distressing facts?

This is a real difficulty. We are all children of the Enlightenment, whatever other forebears we may acknowledge. It has been a cardinal principle of our upbringing that we must never believe things simply because we want them to be true. But how are we to apply that principle to cases where our wanting-them-to-be-true is essentially a matter of the satisfaction of reason?

Rationalist thinkers—ones who trust our natural leaning towards an intelligible world—have always argued that the real world must indeed be perfectly rational. This assumption, however, has gradually become discredited because the ideas of various rationalists about what actually was rational conflicted and led to rather irrational disputing. Kant, for instance, thought that Euclidean geometry was a necessary form dictated by the nature of rationality, a shape to which reality must conform. Hegel, notoriously, proved on rational principles that there could only be seven planets… And so forth.

This simple, confident rationalism, in fact, is officially dead. But that does not stop much of science—especially some kinds of science, especially theoretical physics—from going on very much as if it were true. When cosmological theory advances, and rules that something must have been the case about the Big Bang, it is generally assumed that this is not just a move in a highbrow game played in physics labs, but the discovery of a fact about the real world. It is not an empirically discovered fact, and there need not be any direct physical evidence for it. It rests on the logical coherence of theory.

If we ask a physicist why this reliance on human reason is all right, when Kant's and Hegel's and indeed Marx's reliance on their reason led them so wrong, he is likely to say that there is nothing objectionable about relying on reason as such; these people just reasoned wrongly. We are then (more generally) justified in assuming regularity in nature. Main current theories can be assumed to be true directly of the world.

This cheerful assumption of harmony between our faculties and reality—known as ‘naive realism’—is the attitude which scientists normally hold, and perhaps must hold, about the work they are doing while they are doing it, just as the rest of us do about whatever we are dealing with. Most of the time this faith is perfectly satisfactory, but occasionally it begins to matter how ambitious the claims involved are.

At the modest extreme, our ‘realism’ might only mean that we believe the thing we are talking about is really there. At the more ambitious pole, it might mean that we believe it is exactly as we describe it. This ambitious claim gets less and less convincing as one conceptual scheme after another has to be changed. Our ancestors were, we think, seriously mistaken in classing whales as fish or oxygen as dephlogisticated air. But of course they were not wholly wrong; they had grasped some part of the truth. The whales were indeed not birds nor the oxygen a form of brickdust.

Radically sceptical suggestions that all our knowledge is just a social construction, not shaped at all by anything outside us, do not make much sense. We need a workable compromise. We need somehow to insist that the world really is there, and that we are not wholly mistaken about its nature, while admitting that our acquaintance with it is slight and patchy, blurred by all sorts of cognitive weaknesses. ‘Realism’ is not the name of one among two rival alternatives. It is not a football team for which one can sign up. It is one pole of two between which we must somehow find a place.

How far, then, can scientists be realists? In some sense, as just mentioned, they must be naive realists when they are actually working. When, however, they feel sceptical doubts, they can put on a different hat and use a more cautious attitude—some chosen form of positivism or instrumentalism. They can say that what they seem to be saying about the world is not meant literally, but is only an explanatory construction summing up past observations so as to predict future ones. As Rom Harré puts it:

This is, as he points out, quite an old view. Copernicus's posthumous book De Revolutionibus contains a preface, added by its nervous editor, to explain that its alarming doctrines need not be taken as literal truth. Similarly, Bishop Berkeley wrote that asking whether the earth moves is really only asking

Scientists need not, then, believe that anything outside the minds of observers exists at all. When they seem to be speaking about the movement of remote stars, or about the behaviour of imperceptible particles, they are not really referring to those external things. They are just describing the hypothetical experiences that human observers would have in certain circumstances. ‘Particles’, ‘galaxies’ and the like are only the names of concepts convenient to operate with, not the names of real objects at all.

This kind of view posed no great problem for Berkeley, who held that all reality was spiritual anyway. For him, the unobserved outside world really did not exist except as a set of ideas in the mind of God. Berkeley was an ‘idealist’, in the full metaphysical sense that involves (not high moral ideals, but) the rejection of matter. He had paid his metaphysical entrance-money for his remarkable view about science. J.-L. Borges, who is fascinated by Berkeley, has written a story about a world of this kind, where people really do not believe in objects which are not being perceived:

For Borges's people, the idea that unobserved objects might continue to exist is an absurd, inconceivable paradox…

There is plenty of point in such a story, but this state of things surely demands unquestioning belief in a spiritual world of a peculiarly solid sort. This is not just a distinct culture; it is (as Borges says) a different world, with a kind of imagination that may indeed not be human at all. There would have to be a strong supernatural background to supply the continuity that is lost from the world if moons, tables and human bodies really only exist at times when they are observed.

Berkeley's ideas, however, have been taken up by theorists who did not always notice what drastic changes they called for. Nineteenth-century thinkers, following Ernst Mach, adopted this notion as a general explanation of scientific claims about the external world, often without paying the extra and accepting metaphysical idealism. Instrumentalism was transplanted without the soil in which it had originally grown. Their successors, less and less interested in metaphysics, have become even less aware that they were doing this.

Without God as a receptacle for the vast unobserved background, Berkeley's suggestion looks much odder. If it is then to become something which can actually be believed, rather than just being recited in order to get out of a difficulty, it needs some different kind of metaphysical support, which no one seems yet to have supplied. Since, however, these larger implications are often not noticed, instrumentalism is seen as metaphysically economical, as a modest, parsimonious way of avoiding saying anything about unobserved entities. It has been an important source of the modest, minimalist image of science which we noticed in chapter 1.

Like other modest, minimalist claims, instrumentalism looks good when you want to keep out of trouble, but it is impossible to live up to it in normal life. As just mentioned, most scientists simply forget it most of the time. They assume that they are talking about the real world directly. Without this belief, they, like the rest of us, would be liable to find their imaginations totally disoriented, so they quietly use it. For the more reflective among them, however, this double life is uncomfortable.

That is probably why some of them have lately turned to another way out, saying that the universe itself is in some strange sense the product of our minds—that we produce it by observing it, that it is simply a mass of information, validated by and continuous with the mass that we have in our own thoughts. (Information, we may note, is the modern substitute for spirit.) Thus a curious, ill-understood kind of idealism has lately been creeping back, suggesting that the universe is in some bizarre way our creation. No doubt this is calculated to make it seem less surprising that it should conform to the laws of our thought. We will be looking further at this strange neo-Berkeleyan view later.⁶

Altogether, then, the difficulty about justifying scientific confidence is a real one. If this confidence is all right, what makes it so? As I have tried to show, this is not just a question about the nature of the physical world, nor about the possibility of knowing that world. These questions about knowledge cannot be properly approached without raising central questions about what James called our ‘passional nature’, about what matters to us, about the general aim of our lives.

Knowledge is not an isolated phenomenon. It is made possible by trust, and we do have a choice about what we will trust. The immense achievements of modern science have grown directly from a revolutionary decision to trust the physical world—to assume that it had an underlying order. This involved a corresponding decision to trust those faculties in ourselves that serve to seek such an order. At the Enlightenment, we—Western intellectuals—took a resolution to trust those faculties above all, rather than other faculties which had previously seemed people's surest guides, faculties centring on their moral being and their own self-knowledge. That, I think, is what is centrally meant by calling the age that has followed a scientific one.

What, however, are the faculties in which such a scientific age puts its trust? They are often described as being sceptical, critical and methodical. They are taken to centre on negation—on discipline, on the control of thought, on resisting natural errors. But this overlooks the immense, primary, positive act of trust directed to the physical universe that was necessary before these critical skills could be used. Highly disciplined, sceptical thinking was not itself a novelty, certainly not something peculiar to the modern age. There were very sophisticated sceptics in ancient Greece, and also among late medieval philosophers. That scepticism, indeed, was what built up the impression of futile cleverness which finally discredited scholasticism. Descartes's distinctive move, which made room for Galileo and for modern physics, was not his scepticism, but his finding a way out of that scepticism by a radical act of trust in scientific reason.

That act of trust was explicitly a religious one. Descartes argued that we can trust the Creation because we can trust God the Creator. He defended that position by complex arguments, carefully related to the methods of science. But arguments of this kind turned out ill-suited to this work. The kind of God they described was inevitably something more like a natural force than like the God that people experienced in their lives. As time went on, this ‘God of the philosophers’ became more and more of a distant abstraction, and it became less clear why there was any need to add him to other scientific concepts. Thus, finally, Laplace told Napoleon that ‘he had no need of this hypothesis’.

As far as the internal success of science was concerned, this was of course understandable. At that time, Newtonian and mechanistic thinking looked complete and self-justifying. But that fact is no help towards the question ‘what, fundamentally do you put your trust in?’ Such trust is not just a matter of procuring information. It is a matter of profound reliance, of what one believes lies under the surface of life, what will endure when that is shaken.

If all religious and contemplative regard for the world and its creator are removed from the scientific attitude, we may be left answering simply, ‘Negation remains. You can always trust your critical faculties.’ The trouble with this is not just that it is depressing, but that it is false. Critical faculties, left to themselves with nothing positive to criticize, cannot function at all. They have no standard to work by. They rapidly eat themselves away and disappear up their own orifices, as they have always done in purely sceptical thought.

Moreover, the idea that we should rely solely on them—the sense that somehow we ought to—is itself a moral judgment, and a very remarkable one, which those critical faculties themselves cannot validate. Taking this line would not release us from the need to treat this whole issue as what it is—a moral one, a question about how we should aim to live. This is what is still not sufficiently noticed about the change to a ‘scientific age’. It has been above all a moral change. A change in what we trust has inevitably involved a change in what we admire and honour, therefore in the direction we give to our lives.