The Problem of Evil: The Strongest Argument Against God

A tsunami takes a quarter of a million people in a morning. A child dies of cancer while her parents pray. If God is good and God is able — why? We state the argument at full strength, concede what it gets right, and then answer it.

Before we begin: this article has two kinds of readers, and they need different things. Some of you are here to think — you’ve met this argument in a video or a debate and you want it handled seriously. Others are here because you are living it — you are asking “why?” from inside a loss. This article is written for the first reader. If you’re the second, the philosophy below is true but it will feel thin, and that’s not a defect in you. The last section is for you, and so is our door — no argument required.

Now, the argument. It is ancient — versions of it are attributed to Epicurus more than two centuries before Christ — and it remains, by broad agreement on all sides, the most powerful objection to God ever formulated:

Modern philosophy sharpened this into two distinct weapons. The logical version (stated most famously by J.L. Mackie in 1955) claims a strict contradiction: God and evil cannot coexist, the way a square circle cannot exist. The evidential version (stated most famously by William Rowe in 1979) retreats to probability: perhaps God and some evil could coexist — but the sheer amount, intensity, and apparent pointlessness of actual suffering makes God’s existence unlikely. Rowe’s example has haunted the literature for fifty years: a fawn, burned in a forest fire sparked by lightning, dying slowly over days, seen by no one. No free will is involved. No lesson is learned. Who is that suffering for?

The logical version: the argument that lost

Here is something most people have never been told, on either side of this debate: the logical problem of evil — the claim that God and evil strictly cannot coexist — is widely regarded within professional philosophy as defeated. Not by a preacher. By an argument.

In the 1960s and 70s, Alvin Plantinga formalized what’s now called the free will defense. The core insight: for the atheist’s contradiction to hold, it must be true that an all-powerful God could create any world He likes. But “all-powerful” has never meant the ability to do the logically contradictory — God cannot make a married bachelor — and here is a candidate contradiction: creating free creatures who are guaranteed to choose only good. If genuine freedom means the real ability to choose otherwise, then a world with free creatures capable of love is a world where evil is genuinely possible — and it may be that no world containing the great good of freely chosen love could be created without that risk. Plantinga didn’t need to prove this is why God permits evil; he only needed to show it’s possible — because a single possibility breaks a claim of strict contradiction.

It worked. Mackie himself, in his final major book, acknowledged that the argument he had championed did not show believers to be irrational — the problem, he conceded, was not after all a straightforward contradiction. William Rowe — the architect of the evidential version, and no friend of theism — noted in the very paper that launched his own argument that the logical problem was generally conceded to have failed. When the two leading atheist philosophers of the problem of evil both set the logical version down, that is not apologetics spin. That is the scoreboard. This is why the serious debate moved to the evidential version — and why, when someone on the internet tells you “God and evil are a flat contradiction, checkmate,” they are running an argument professional atheism retired fifty years ago.

The evidential version: the argument that remains

Rowe’s version is humbler and harder. Fine, God and evil can coexist — but look at the fawn. Look at the sheer volume. Surely some of this suffering is pointless — serving no good that outweighs it. And a perfectly good God would prevent pointless suffering. The engine of the argument is an inference: I can see no justifying reason for this suffering; therefore, probably, there is none.

The best-known response, from Stephen Wykstra, asks a deceptively simple question: is that inference any good? “I can’t see it, so it probably isn’t there” works only when you’d expect to see the thing if it existed. No elephant in my garage? Fair conclusion — I’d see one. No carbon monoxide in my garage? My eyes prove nothing. So the question becomes: if an infinite God had reasons spanning all of history, every ripple of every event, the inner lives of every person affected — would a creature with our vantage point expect to see those reasons? The gap between us and God is not the gap between a smart human and a smarter one. It’s closer to the gap between an infant and a surgeon: the infant experiences the needle as pure, pointless cruelty, and is wrong — not because the pain isn’t real, but because the reasons live outside the infant’s horizon.

Honesty note: this response — “skeptical theism” — carries a real cost, and its critics press it hard: if we can’t judge when suffering is pointless, can we judge anything about God’s ways? We think the response survives (trusting a surgeon’s reasons is not the same as knowing nothing about surgeons), but you should know the debate is genuinely two-sided here. What Wykstra’s point secures is more modest and still decisive: the evidential argument’s engine — from “I see no reason” to “there is no reason” — cannot bear the weight placed on it. An argument for atheism that rests on the assumption that a finite mind should be able to audit an infinite one is quietly assuming the thing it needed to prove.

What a world without evil couldn’t contain

There’s a further question the argument rarely pauses on: what exactly are we asking God to have made instead? Press on the wish — “a world without suffering” — and it begins to dissolve. A world where no one can be harmed is a world where courage is impossible, because there is nothing to risk. Compassion is impossible, because no one suffers. Forgiveness is impossible, because no one wrongs you. Perseverance, sacrifice, faithfulness through the worst — every quality we admire most in human beings is logically parasitic on a world where things can go badly wrong. John Hick called this the soul-making insight: if God’s purpose were maximizing pleasant sensations, this world is inexplicable — but if the purpose is forming persons — free beings grown into courage, compassion, and love — then a safe, padded, frictionless world is precisely the world that could not do it.

Concession, again, because this is where glib versions go wrong: soul-making does not explain the distribution. Suffering is not portioned out as a curriculum — it crushes some and skips others, and some evils (the fawn again) form no visible soul at all. Soul-making explains why a good God’s world might contain the possibility of real harm; it does not pretend to itemize the ledger. It is a piece of the answer, not the answer.

The hidden premise: evil cuts both ways

Now the move the argument never sees coming. Every version of the problem of evil borrows a premise it cannot pay for: that some things are objectively evil — not merely disliked, not merely disadvantageous to our species, but genuinely, really wrong. The fawn’s suffering is supposed to matter. The child’s death is supposed to be an outrage against how things ought to be. But “ought” according to what? C.S. Lewis, who came to faith through this very door, recalled that as an atheist his case against God was the universe’s cruelty and injustice — until he asked himself where he had gotten the standard:

“But how had I got this idea of just and unjust?”

— C.S. Lewis, Mere Christianity

A line is only crooked against a straight one. If the universe is, at bottom, matter in motion — no design, no purpose, no standard above the fray — then the fawn’s death is not evil; it is merely chemistry, and our horror at it is an evolutionary reflex with no more authority than a taste for sugar. The atheist who feels — as nearly all do, to their credit — that childhood cancer is not just unpleasant but wrong, cosmically wrong, is standing on a moral floor his own worldview never poured. This does not solve the problem of evil for the Christian. But it reveals something remarkable: the problem of evil is only a problem — rather than just a description — in a universe where good and evil are real. The argument’s ammunition is on loan from the worldview it’s firing at.

The answer Christianity actually gives

And now the honest pivot. After all the philosophy — the defeated logical version, the broken inference, the soul-making frame, the borrowed standard — Christianity’s deepest response to suffering is not an argument at all. It is an event. Every other picture of God leaves Him above the fray: serene, distant, untouched. The Christian claim is that the God in question came down into it — was born poor, was betrayed by a friend, was tortured by the state, and died young, unjustly, in agony, crying the very psalm of abandonment this article opened with: My God, my God, why have you forsaken me? Whatever the full reasons for suffering are, Christianity says God’s answer was not to explain them from a safe distance but to come and bear the worst of them personally. The theologian John Stott, after a lifetime with this question, put it plainly:

“I could never myself believe in God, if it were not for the cross.”

— John Stott, The Cross of Christ

Notice that this is the same shape as the ending of Job — the Bible’s own book-length treatment of the problem. Job demands an explanation for thirty-seven chapters. God shows up, and famously gives him none. What Job gets instead is presence — and, astonishingly, it is enough for him. That is not evasion; it is the claim that at the bottom of this question lives a person, not a proposition. And the story does not end at presence. The same faith stakes everything on a promise about where the whole thing is going:

If that promise is true, then no suffering is the end of any story — and eternity is not a consolation prize bolted onto the argument, but the context that reframes every ledger the argument tries to total from inside time.

Where this leaves us

Stated carefully: the logical problem of evil is defeated, by the concession of its own champions. The evidential problem runs on an inference — “I see no reason, so there is none” — that cannot survive the gap between a finite auditor and an infinite subject. The world the objection wishes for, scrubbed of all possibility of harm, turns out to be a world scrubbed of courage, compassion, and love. And the outrage that gives the argument its force — the certainty that suffering is genuinely wrong — is itself a signpost pointing at the very moral reality atheism must deny. Evil is a real problem for the Christian. It is a worse one for the atheist, who must feel its wrongness while explaining that the wrongness is an illusion.

But we promised honesty to the end, so: none of this makes the problem vanish, and no honest Christian claims to hold the complete ledger. What Christianity holds instead is a God with scars — proof that He is not indifferent — and a promise that the story is not over. Between those two, the believer stands in the same place Job did: without the explanation, in the presence of the One who has it.

And if you came to this page as the second kind of reader — not debating, but bleeding — then forget the syllogisms for now. The Psalms give you permission to say the angriest thing you’re thinking directly to God; that’s what they’re for, and He has heard worse from people He loved. You don’t have to win an argument to bring Him your grief. Bring it here if you need somewhere to start.

Sources

• Epicurus, as formulated in Lactantius, De Ira Dei (On the Anger of God), ch. 13 — the classical statement.
• J.L. Mackie, “Evil and Omnipotence,” Mind 64 (1955) — the modern logical version; see also his The Miracle of Theism (Oxford, 1982), where he concedes the argument shows no outright contradiction.
• Alvin Plantinga, God, Freedom, and Evil (Eerdmans, 1974) — the free will defense.
• William L. Rowe, “The Problem of Evil and Some Varieties of Atheism,” American Philosophical Quarterly 16 (1979) — the evidential version and the fawn; includes his acknowledgment regarding the logical version.
• Stephen J. Wykstra, “The Humean Obstacle to Evidential Arguments from Suffering,” International Journal for Philosophy of Religion 16 (1984) — the critique of the “no-see-um” inference.
• John Hick, Evil and the God of Love (1966; reissued Palgrave, 2010) — the soul-making theodicy.
• C.S. Lewis, Mere Christianity (1952) and The Problem of Pain (1940); for the grief-side of Lewis, A Grief Observed (1961).
• John Stott, The Cross of Christ (IVP, 1986) — the suffering of God as Christianity’s distinctive answer.
• Eleonore Stump, Wandering in Darkness: Narrative and the Problem of Suffering (Oxford, 2010) — the major recent academic treatment, including Job.
• Fyodor Dostoevsky, The Brothers Karamazov (1880), Book V — Ivan’s rebellion, the most powerful literary statement of the objection, presented by a Christian author.
• Scripture quotations taken from The Holy Bible, New International Version® NIV®. Copyright © 1973, 1978, 1984, 2011 by Biblica, Inc.™ Used by permission. All rights reserved worldwide.