I
Consider the situation of our species. It is among the strangest situations imaginable, and the species has somehow contrived never to find it strange. We awoke inside a reality we did not make. We have no memory of its beginning and no testimony from anyone who was present at it. We can observe only a small region of it, so small that the word region flatters us, and we have been observing that region for an interval which would not register on any clock the universe might keep for itself: a few thousand years of writing and a few centuries of instruments, set against some fourteen billion years from which we were absent for all but the final instant. And almost at once, long before we had learned to cure a fever or to say why the sky is blue, we began to issue definitive accounts of the whole. We named its author, dated its creation, declared its purpose and foretold its end. A mayfly that hatched at dusk and devoted its only night to a treatise on winter would be a modest creature beside us. It would at least have felt the evening grow cold.
Every people has done this, and every people has looked upon the efforts of the others with pity. The Egyptians knew that the sun was carried each night by boat through the underworld. The Babylonians knew that the sky was half the carcass of a slaughtered goddess. The schoolmen knew the order of the heavenly spheres and which choir of angels turned each one. We smile at all of this, and our smile is the most instructive thing about us, because it is precisely the smile with which each of those peoples regarded the fables of the people before it. The modern man laughs at the cosmologies of antiquity while carrying about with him a cosmology of his own which he has never once examined, assembled out of dimly remembered popular science, the moral habits of his parents and the general mood of his decade, and which he does not call a cosmology at all, but simply the way things are. Every generation believes itself the first to have escaped the primitive errors of the generations before it. That belief is the most primitive error of all, and the only one no generation has ever escaped. It is this, and not any particular doctrine, that I mean by the arrogance of certainty.
II
I should say at once what I am not arguing, since the position I mean to defend is easily mistaken for a cheaper one. I am not arguing that nothing can be known. That thesis is nearly as old as philosophy and roughly as interesting as the schoolboy who has just discovered it. It refutes itself in the act of being stated, for the man who knows that nothing can be known already knows one thing too many; and in any case nobody holds it, since the sceptic who professes to doubt the existence of the world has never yet doubted it at the edge of a cliff. There is an enormous difference between saying “I know nothing” and saying “I refuse to claim more than I know.” The first is a posture, and usually a vanity. The second is a discipline, and a far harder one, because it obliges a man to keep an honest ledger of his beliefs, in which every entry is weighted by what he can actually show for it, and to go on keeping it when the totals disappoint him.
And we can show a great deal. It would be a strange humility that pretended otherwise. Within a few centuries a species of anxious primates, whose nervous systems were shaped for finding fruit and escaping leopards, has weighed the sun, read the chemistry of stars it will never visit, caught in its telescopes light that set out before the Earth existed, and found, coiled inside its own cells, the written instructions for making itself. Mathematics has given us truths that do not decay. Observation, disciplined by logic and made public by experiment, has given us knowledge that anyone willing to repeat the work may check. These are the greatest achievements of our kind, and I regard them with something close to reverence. But success is a poor teacher of modesty. Because our methods have answered so many questions inside reality, we have begun to assume that they have answered, or soon will answer, the questions about reality: why it exists at all, and why it has this character rather than some other. These are questions of a different order, and the confidence we earned in the first order does not pass automatically to the second, any more than a man who has learned every room of a house has thereby learned who built it, or why, or what lies beyond its walls. It is remarkable enough that a brain assembled for the savanna can do mathematics at all. It would be more remarkable still if that brain could be trusted, unsupervised, on the questions furthest from anything it was assembled to do.
Why is there something rather than nothing? Why these laws and not others, and why laws at all, rather than a chaos in which nothing could be predicted because nothing ever happened twice? Why does the world submit so obediently to mathematics, a discipline we appear to invent at our desks and then find already written into the orbits of the planets? Why is any of it accompanied by experience, so that there is something it is like to be a living creature, when all of the machinery might just as conceivably have run in the dark? Is physical reality the floor of things, or does it rest on something else? Could anything precede it, and would the word precede even mean anything there? Does existence have a purpose? Is there a creator? And if there is, by what right do we assume that it resembles in any particular the gods of human religion? Xenophanes observed two and a half thousand years ago that the Ethiopians gave their gods flat noses and dark skin, that the Thracians gave theirs blue eyes and red hair, and that if horses could draw, they would draw their gods as horses. Nobody has improved on the observation, and remarkably few have taken it to heart.
I do not ask these questions in the manner of a man who has privately decided that the answer to each of them is nothing and is merely waiting for his audience to catch up. That is a common trick and a dishonest one. I ask them because I do not know the answers, and because, having looked with some care at what has been offered in reply, I do not believe that anybody else knows them either. The theologian who tells me the name and the intentions of the creator claims far more than his evidence will bear. But the confident materialist who assures me that there is certainly nothing beneath or beyond the physical world, and that consciousness is certainly nothing but the electrical traffic of the brain, commits the same offence in the opposite direction. The absence of evidence for a particular god is not evidence for his particular metaphysics. He has only exchanged one closed box for another and mistaken the exchange for an escape. On some of these questions I have leanings, and on the question of purpose they are not warm ones. But a leaning is not a proof, and I will not dress the one in the clothes of the other. The questions stand open because, when I examine them honestly, that is where I find them.
III
If the evidence is so thin and the confidence so thick, then the confidence must be coming from somewhere other than the evidence, and it is worth asking where. I think the answer is that “I do not know” is a very uncomfortable place to live. It gives a man nothing to stand on. It does not tell him what his suffering is for, or where his dead have gone, or how he ought to behave, or whom he ought to blame, or what will become of him when his own turn comes. A certainty, whatever its merits as a description of the world, is an excellent piece of furniture. Religion supplies a purpose, a morality, a community, and an account of death that can stop a child crying at his mother’s grave. A political doctrine supplies a complete explanation of society, in which every misfortune has an author and every author has a name and an address. Even the man who prides himself on believing only what science tells him will very often, if one listens long enough, turn out to have built an entire metaphysics out of results that support nothing of the kind, and to defend it with a heat that no result has ever warranted. Nature, we used to say, abhors a vacuum. The mind abhors it far more, and will fill an empty box with anything at all, provided only that the box is no longer empty.
From this follows a method, and it is the sharpest instrument I know. When we examine a belief we ordinarily ask only whether it is true. That question must always be asked, and always asked first. But there is a second question, which we owe to Nietzsche more than to anyone, and which is often the more revealing of the two: what kind of creature needed this belief to be true? What fear does it quiet, what humiliation does it repair, what appetite does it license, whose authority does it protect? To answer these questions is not to refute the belief. The origin of an idea is not its verdict, and a doctrine born of terror may, by some accident, be correct. But when a belief proves to be exactly the belief its holders would have had to invent if it were false, we are entitled to hold it with a good deal less confidence than they do, and to notice that their confidence is evidence about them rather than about the world.
It would be very pleasant to stop here, with the author and his reader standing together on a height and looking down with a certain sympathy on the believers below. That is exactly why the instrument must now be turned around. Disbelief consoles too. There is a real and not especially noble pleasure in being the one who has seen through things, and a man can grow as attached to his disillusionment as any saint to his visions. Nihilism can become an identity, worn like a black coat. Contrarianism offers, at almost no cost, the sensation of standing apart from the herd, and the herd of contrarians is by now one of the largest there is. Even rationality can become a costume, a vocabulary and a manner by which one signals membership in the right tribe while none of the work that rationality actually demands is done. The man who congratulates himself on having escaped dogma has very often found only a more sophisticated dogma, whose chief attraction is that it flatters his intelligence instead of his hopes.
I do not exempt myself from any of this. The genealogical question applies to me with full force, and it is not a comfortable one to put. What kind of creature needs the universe to be indifferent? Perhaps one who wishes to feel brave, or who would rather be tragic than ordinary, or who has come to take a certain pride in bearing what others cannot. I am unable to rule any of these out, and so I am obliged to hold my darkest convictions to the same standard I apply to the brightest convictions of other people. No belief is granted immunity because it is mine. None is exempt from revision because it was difficult to reach, or because I have suffered for it, or because abandoning it would embarrass me. If I am serious about following the truth wherever it goes, I must accept that it may go somewhere I dislike, and that this includes the possibility of its leading me back toward positions I once took great pleasure in abandoning.
IV
Most reflections on the limits of human knowledge end at about this point, with a gentle bow. We are small and the universe is large, and we must learn to live with not knowing. I find this ending unsatisfying and a little idle, because it treats the admission of ignorance as the conclusion of an inquiry when it is only the beginning of one. To admit that one does not know is a necessary act of honesty. To stop there, and to settle into one’s ignorance as though it were a position, is merely another way of closing the question, and of closing it with a better conscience than the dogmatist enjoys, which makes it all the harder to reopen.
If the honest answer to a question is that I do not know, then something has been disclosed that the dogmatist never discovers, namely that there is work to be done. The unknown is a territory, and territories can be entered. We can measure the thing, and then measure it again with better instruments. We can test it. We can construct a model of it and watch the model fail, and learn more from the particular manner of its failure than any success would have taught us. We can read, with patience, what was found by the many who stood at the same edge before us. We can build an instrument that sees what the eye cannot, or write a program to perform a calculation that no human lifetime would be long enough to finish by hand, or arrange an experiment so that nature has no choice but to answer yes or no. And we can discover, which is often the greatest advance of all, that the question was badly put, and exchange it for a better one. Understood in this way, uncertainty ceases to be a mood and becomes a set of instructions. It gives us nothing to believe and everything to do.
This is why I have never been able to regard thinking and making as separate occupations, still less as rival ones. The philosopher has long looked down on the engineer as a man who manipulates what he does not understand, and the engineer has returned the compliment by regarding the philosopher as a man who claims to understand what he could never manipulate. Both are mistaken, and for the same reason, which is that neither has seen that construction is itself a way of knowing, and in certain respects the most severe. Vico held that we can fully know only what we have made. I would put it more modestly and more harshly: we do not really know whether we understand a thing until we have tried to make it.
An idea can survive indefinitely in abstraction. It can be refined, admired, defended in argument, taught to the young and handed down for centuries without once being required to touch the world. Aristotle taught that heavy bodies fall faster than light ones, and for the better part of two thousand years the doctrine was repeated by learned men, any one of whom could have tested it in an afternoon and very few of whom did. In 1586 the Flemish engineer Simon Stevin and a friend carried two balls of lead up a church tower in Delft, one ten times the weight of the other, let them fall together, and heard them strike the board below with a single sound. It is fitting that he was an engineer, a man whose mills and sluices had actually to work. The moment one tries to build something according to an idea, the idea is driven out of the sheltered air of discussion and made to bear weight. The bridge stands or falls, the engine turns or seizes, the program runs or halts, and no quantity of eloquence will alter the result by a hair. Reality, which cannot be bribed and takes no interest in our arguments, is given a vote, and its vote is final. A theory can flatter its author. A working thing cannot.
Seen in this light, the questions we have not answered are no embarrassment. They are the most interesting feature of our situation. It is a common complaint, and a melancholy one, that we have arrived too late, that the great discoveries have all been made and the maps completed. Swift laughed at the geographers who filled the gaps in their maps of Africa with pictures, and set elephants over uninhabitable downs for want of towns. We have not stopped drawing elephants. We have only become better draughtsmen, and our maps look complete for much the same reason theirs did. Almost nothing is finished. Our knowledge is not finished; its foundations are still in dispute, and its largest questions have not so much been answered as postponed. Our understanding of intelligence is not finished; we cannot yet say with any precision what a mind is, which has not prevented us from feeling quite certain that ours is the only kind that matters. Technology is not finished, and civilization is very far from finished. Nor is it at all obvious that the human being is finished, or that the particular bundle of instincts and limitations which evolution happened to leave us with, for reasons that had nothing to do with truth, is the last word on what a knowing creature can become.
V
Certainty closes questions. I would rather leave them open, not because ignorance is a virtue, but because an unanswered question is still capable of becoming an answer, while a question that has been given a false one may wait two thousand years for somebody to climb a tower.
There are things I believe strongly. There are things the evidence makes almost impossible to doubt. And there are questions, far more of them than I would like, to which the only answer I can honestly defend is that I do not know.
I intend to reduce that number.