The Logos Scripture · The Stoa · Dialectic
David Bai · 2026-08-17 · hangs on The Book of Popper 5.10
I
Charles Taylor occupies the most respectable apologetic seat in the contemporary academy; on the side of the ordinary reader, that seat has always belonged to C.S. Lewis.
He was a tutor at Oxford, a scholar of the first rank in medieval and Renaissance literature, and the author of fiction that has outlasted him; yet what carried his name across the world was, in the end, the most formidable popular defence of Christianity the twentieth century produced. Mere Christianity was made out of wartime broadcasts, and it uses none of the heavy, clouded vocabulary of modern sociology, but is written with the briskness and the common sense of a classical logician, so that millions of listeners under the bombing could follow it. He does not retreat into mystery, and he does not ask anyone to lay down his reason; he claimed to have been dragged into faith by the sheer weight of evidence, kicking and struggling the whole way.
When I wrote on Taylor I set out the three roads that defenders of the faith take when they meet the secular world. Coleridge tampered with the definitions, splitting Understanding from Reason so that the fork between truth and God could never appear. Dostoevsky built no wall and offered no sophistry, and simply would not pay; he told you the order set in his own heart and argued nothing further. Taylor moved from argument to diagnosis, and made the modern secular man a patient — closed, flat, wanting something.
Lewis is none of these. He had no use for Coleridge's sophistry, and he would have refused Dostoevsky's sentence outright, holding as he did that a Christianity which is not factually and historically true is a delusion of no value, not worth a man's afternoon. He meant to show that it is true.
So he took a fourth road, and it is harder to guard against than any of the three: he alters no definitions, substitutes no question, and diagnoses nobody. What he does is count the doors for you.
II
Begin where his book begins, since the whole building is raised from there.
Lewis's first brick is what he calls the objective moral law. Every man, he says, carries within him a natural law telling him that he ought to do the right thing, a law that does not change with the age; and where there is such a law there must stand behind it an overseer, a lawgiver.
The step is taken too quickly.
The moral law is not a decree that has hung there since the beginning. It is a way of staying alive, settled out over long ages among creatures that live in groups. We cannot find it in the simplest organisms, and we cannot say that every man keeps it now — the world is full of people who do not, and they go on living all the same. It is no more than a rule generally agreed upon, which under ordinary conditions allows people to survive together.
Nor is it ours alone. Rams have their rules and wolf packs have theirs, and the groups that kept no rules were cleared out of the record long ago. Altruism, cooperation, not slaughtering one's own kind — these are in the end strategies by which genes continue, not edicts handed down from above.
That the world has regularities is true enough; but regularity establishes law, not will. A law oversees nobody and demands nothing. To read the uniformity of nature as the handwriting of a god is neither logically compelled nor harmless, since it misleads precisely those readers who trust him.
III
On the identity of Jesus, Lewis offers the famous trilemma.
A man who was merely a man, he says, and who said the sort of things Jesus said, could not possibly be a great human teacher; he is either a lunatic — on a level with the man who says he is a poached egg — or else a devil out of hell, or else he is what he claimed to be. You must choose, and no middle course is allowed you.
The passage collapses in two places, and the two are independent of each other.
The first is the premise. So far as I can tell, the Jesus of the gospels claims at most to be the son of God, not God himself; and the sayings in which the claim is least ambiguous are gathered in John, the latest of the four by decades. Besides, son of God will bear more than one reading: if a man holds that all men are children of God, then saying it of himself does not strictly conflict with his also being a great moral teacher.
The second is the list. The trilemma omits a fourth possibility, and that fourth is the one most historians now take: legend. The gospels were written down decades after his death, in a language he did not habitually speak, by communities that already held doctrinal positions. A Jewish rabbi may perfectly well have been an excellent moral teacher whom his followers and their church raised, layer upon layer, into a god. The Buddha never called himself God, nor did Confucius, and later ages made sages of them regardless. The same thing, done the same way.
Hang that door on the wall and the trilemma has not an ounce of force left. So it is never hung.
IV
Let me say plainly why this is more dangerous than the other three.
Coleridge's wall can be seen if you look at it, and Taylor's substitution heard if you listen for it. But a list drawn up in advance looks exactly like rigorous inference: three doors, eliminate two, and what remains must be true — which is a legitimate form of reasoning. The whole fraud sits in one place, namely who drew up the list and whether it is complete; and that is the one place he never argues.
So the more a reader wishes to be reasonable, and the more faithfully he chooses among the options handed to him, the more securely he lands where he was meant to land. Lewis does not ask you to abandon reason; he asks only that you exercise it inside the boundary he has drawn. There is no cheaper snare in the world.
V
What is remarkable is that Lewis had already drawn the method as a picture.
The design of Mere Christianity is a hall standing above the denominations, with doors opening off it — one to Rome, one to Canterbury, one to the reformed churches. He brings you only as far as the hall, he says; which room you enter is your own affair.
Look at the shape of that hall. It has the same skeleton as the trilemma: the doors are given, their number is given, and whether the hall itself exists is never a question that comes up.
A hall that keeps off the wind and rain is a fine thing, but it has to exist first. A real room, or a real umbrella, is small and will keep the rain off you; a hall that cannot be shown to exist, however handsomely drawn, will not stop a single drop when the weather turns.
The labour of his life went into making that hall convincing. He never answered what holds it up.
VI
Lewis has a second famous instrument, the Argument from Desire.
Creatures, he says, are not born with a desire for which no satisfaction exists. An infant feels hunger, and there is food; a duckling wants to swim, and there is water; a man feels desire, and the thing desired is real. He then carries the biological pattern over to the human spirit: if I find in myself a desire which no experience in this world can satisfy, the most probable explanation is that I was made for another world.
Taylor looks at the disquiet in the modern mind and calls it a machine that has broken; Lewis looks at the same emptiness and calls it a homing signal. It is beautifully done, and it breaks in one place, and breaks cleanly.
It breaks here: between a universal human appetite and a highly specific set of doctrines there runs no logical pipe whatever.
Grant him the premise; grant that the unquenchable human longing for vastness, for mystery, for significance does point at something beyond this world. Grant all of it, and the hunger still does not match the menu he sets down. What a man feels beneath a night sky is an appetite for the infinite, for the sublime, for what does not perish; it is not, by any stretch of imagination, a craving for a virgin birth, for a deity requiring blood in payment, or for an empty tomb somewhere in Judea. No one has ever looked up at the bending of space and time and thought: ah, what my soul has been missing is a bodily resurrection, together with a strict code of ecclesiastical ethics.
He takes something enormous, universal, belonging to everyone — the human capacity for awe — and uses it as a wooden horse to wheel in a parcel of intensely provincial ancient doctrines; he treats an appetite for the cosmos as though it were a binding contract signed with a particular church.
VII
There is a graver matter still: his biological premise is overturned by his own example.
The duckling wants to swim and there is water — quite so, but the order runs the other way. The water was there first, and selection then built the wanting into the duck. The thing precedes the desire; the desire does not come first and oblige a thing to exist. Hunger is useful only because food already existed, and creatures evolving in a world containing food grew the apparatus of hunger to meet it.
Once that direction is clear, the law turns completely over. Any mind capable of inference, of combination, of extending an existing want past its object will manufacture a great many desires with nothing to answer them: men want to fly without machinery, men want never to die, men want to be loved by everyone, men want the dead to come back. Not one of those four has its object lying ready in the world, and every one of us has had them.
We want wings and were given none, and must build aeroplanes according to the laws instead. Most of what a man desires is not standing ready in nature; it has to be made, one thing at a time. As for God, science cannot make him either — science can at most make an artificial intelligence, and however near to a god that intelligence may come, it will not thereby prove that another god has been keeping company with us all along.
That a longing should find nothing to satisfy it is not strange in the least. It is exactly what a mind that reasons ought to produce.
VIII
A word here about the man himself.
Lewis called himself the most dejected and reluctant convert in all England, as though he had been backed into a corner by evidence and obliged to put his hands up. The line is quoted everywhere.
But a man who already wishes to go through a door will assemble his reasons, and assemble them with complete sincerity. Sincerity is not evidence. A person raised inside a faith may be wholly without pretence, may be altered through and through, may be indistinguishable from the most devout — and none of that bears on whether the claims are true. That people make themselves into what they wish to be is ordinary and human; but having become it is not the same as the thing being so.
IX
Now the word cold, since this is where apologists press hardest.
The objective universe, they say, is cold and without feeling, and only a god who cares for you can give warmth.
I grant half of it. The laws of physics hold no tenderness towards us: they do not smile because you have discovered one, and they take no offence when you break one. You work out a regularity and it does not know.
But the other half must be said. For that very reason they are not cold either. Cold and warm are sensations in a body, matters of the nerves; the laws of the universe are neither cold nor warm, they merely run as they are. Embrace a block of ice and the ice is cold — but the cold is spoken by your skin, not by the ice.
And real warmth has always been something between people: an embrace, someone looking after you, someone who stays in the room while you are ill.
From which it is plain why that god had to come down off the altar. A pure lawgiver consoles nobody; the one in the older books, who could be angered and could regret, has at least the shape of a person about him; and when men are asked to hand over their whole lives, nothing will do but a god who can be nailed to wood, who can hurt, who can die. The Trinity grows out of exactly this place.
It is not a structure that fell from heaven. It was measured out and built to the customer's requirements, inch by inch.
X
Were the piece to end here it would be the same as the other three: set a man up, name his method, and stop. But there is one more thing about Lewis that must be taken separately.
In the third chapter of Miracles he set out to prove naturalism self-refuting. His case runs: if human thought is produced entirely by the physical and chemical processes of the brain, and those processes are irrational, then we have no reason to trust that our thoughts see the truth; and since naturalism is itself a conclusion reached by such thoughts, it overturns itself.
The argument breaks at the outset, and it breaks on a single word.
Physical processes are not irrational. At most they are non-rational — they perform no inference, and that is all. These are not the same thing. A calculator performs no inference and computes correctly; a clock does not understand time and keeps it accurately. Not reasoning is not the same as not reliable. Lewis needs the second meaning and has only the first, and the argument falls at its first step.
There is a further turn. A man can reason at all precisely because nature is lawful. If nothing held steady, if today's cause did not bring tomorrow's effect, there would be no inference to be made and no inferring creature would ever have grown. The thing he treats as the threat is the premise he is standing on.
XI
On the second of February 1948, at the Socratic Club in Oxford, the twenty-eight-year-old Elizabeth Anscombe took that chapter apart in public.
Her method was to separate physical cause and effect from logical ground and consequent: how a thing came to be produced and what entitles it to stand are two different questions.
Her example was a poor one — the ground is wet, therefore it rained. That step will not hold: a burst hydrant or a water cart will wet the ground just as well. Inference of that kind is at best probable, and it is what Hume called custom.
But the distinction does not collapse with the example. All men are mortal, Socrates is a man, therefore Socrates is mortal — that step is not a habit, and whether it holds has nothing to do with what anyone has experienced.
So the position is this: her example failed, and Lewis's argument was wrong regardless. His own step had broken already, whether or not anybody came forward to say so.
XII
He went back and changed it. In the second edition of 1960 the chapter is rewritten.
But look at what he changed. He no longer argues about whether thoughts are valid or invalid; instead he argues that naturalism cannot give a sufficient account of the relation between physical cause-and-effect and logical ground-and-consequent, and therefore that for reason to exist at all a rational framework must be woven into the fabric of reality.
Not one word of the conclusion moved. What moved was the road to it.
This is the very thing I set down in writing about Marx. Patching is no crime in itself; any theory may be wrong, and a wrong theory ought to be corrected. There is only one question to ask: once the patch is on, what does it say?
This patch answers a question about how a physical brain reasons by supplying a divine framework. But how does that framework get into the brain, and by what means does it turn a sequence of chemical events into a valid inference? On this he says nothing, and can say nothing. He has not untied the knot. He has moved it somewhere it cannot be seen.
As to whether naturalism can give the account — it can. The brain is an evolved organ and a tool for staying alive. A species whose internal causation did not roughly map the causation outside it, which misjudged the line a predator was running or the depth of a river, would be gone within a few generations. What we call reason is that apparatus, tuned by selection until it draws reality tolerably straight.
And suppose it could not. Lewis's account cannot either, and is far more helpless. To answer a hard question in biology with an unknowable theological mystery has the shape of an answer and delivers nothing whatever.
Anscombe did not accept his revised conclusion. A few philosophers still work this line, but it has never become an obstacle that mainstream philosophy has had to steer around.
XIII
So how much is left of the honesty and seriousness that everyone repeats?
In July 1960 his wife died of cancer. The following year he made a book out of the notebooks he had kept while mourning, and put a false name on it: N.W. Clerk. Only in 1963, after he was dead himself, was it reissued under his own.
It is an extraordinary book, and he does not round the corners. The first line: no one ever told me that grief felt so like fear.
Stop on that sentence.
If a man truly believes in heaven, and his wife goes first, what he should feel is sorrow, and missing her, and the emptiness of the house. Not fear. Fear is what uncertainty produces — you are afraid because you do not know where she has gone, afraid that on the other side there is nothing at all. By the time he wrote that line he was no longer sure.
Further in it gets harder. What reason have we, he asks, apart from our own desperate wishes, for believing that this being is good by any standard a man can conceive? And he entertains the thought that it may be a sadist working on a cosmic scale.
To that point the book is genuine. A man setting down his worst hour exactly as it came, without dressing it — not one of the other three ever did that.
But the book does not stop there.
It goes on, and comes to rest on this: that the picture of God he carried was not a divine picture, that it has to be shattered again and again, and that the one who shatters it is God himself — he is the great iconoclast.
Look at what that step accomplishes.
God did not come. The prayers were not answered. His wife died all the same. Not one of those facts has altered. What has altered is what they are called: the one who did not come is now called present, and the silence is now called a lesson.
The conclusion did not move by a word.
XIV
So we see the same act performed twice.
In 1948 his argument was refuted, and twelve years later he rewrote the chapter — the conclusion fixed, the road to it changed.
In 1960 his faith gave way beneath him, and a year later he wrote the book — the conclusion fixed, and what changed was the name given to the collapse.
Whatever happened, the destination survived. The argument was demolished and it survived; God watched his wife die without a word and it survived that too.
So let the question be put: what would have had to happen for him to say that he was wrong?
I asked the same question in writing about Marx. The answer here is the same answer. Nothing. Not one thing.
As to why he did it, I do not guess. Some say reputation, half a lifetime of books, interests he could not cut himself free of. I cannot verify any of that, and I do not intend to put into an essay what I cannot verify.
I set down only the two things that can be checked: the book of 1961 carried a false name, and his own name went on it in 1963, after he was dead. The rest the reader may see for himself.
XV
So the four roads run as follows. Coleridge tampered with the instruments of reason themselves. Dostoevsky put the instruments down and admitted his preference outright. Taylor exchanged argument for diagnosis and pronounced a healthy man ill. Lewis touches no instrument, substitutes no question, and diagnoses no one; he merely counts the doors and tells you that is all the doors there are. It is the cheapest of the four, the most like reasoning, and the hardest to catch in the act.
And the distance between him and Dostoevsky is far smaller than people suppose. Dostoevsky said it in the open: if the truth were outside Christ, I would remain with Christ. Lewis never says any such thing — but he did it twice, once in his study and once at his wife's bedside.
The only difference is that one of them said so.
As for where we should be going, it is not that hall of his drawn on paper. It is a house that knows the laws it lives under, where we grant that the universe is indifferent and are not made desolate by it — since understanding the laws will protect you, and what people do for one another is warmth enough for this world. There are a great many people who have never known that god, and they live in health and with dignity all the same.
What I have pointed at is the list. As for the man: he wrote down his worst hour, which none of the other three ever did; and having written it, he went back to where he had started. Both sentences have to be said.
The Stoa →The Stoa is not scripture. Everything here is signed, may be wrong, may be refuted, and may never be edited into the scripture. (cf. The Rule 7.3)