There are 160,000 children who will not see tomorrow. Not because of war or human negligence alone — but because the planet we inhabit, the bodies we were born into, the bacterial world we share, operate with no observable preferential treatment for the innocent. This is not an emotional appeal. It is the datum. And it has a name: the problem of evil.
I. The Architecture of the Dilemma
Epicurus formulated it in the third century BC, and no one has honestly dismissed it since. His triad is brutal in its simplicity: if God is willing to prevent evil but unable, he is not omnipotent; if able but unwilling, he is malevolent; if both willing and able, why does evil exist? If neither willing nor able, why call him God at all?
The logical form of this argument held dominance for centuries. J.L. Mackie (1955) argued that the simultaneous truth of "God exists," "God is omnipotent," "God is wholly good," and "Evil exists" constitutes a formal logical contradiction — a set of propositions that cannot all be true. He was not writing polemic. He was doing philosophy with the precision of a logician. And the academic world treated it as such.
Alvin Plantinga's Free Will Defense — the most celebrated theistic response — argued that a world with genuinely free creatures capable of choosing good is more valuable than a world of moral automatons, and that God, in creating free beings, necessarily permitted the possibility of moral evil. It satisfied many. It satisfied Mackie's strictly logical objection. But it answered nothing about nature — about earthquakes, cancer, parasitic wasps, and the deaths of children who never made a moral choice.
II. The Evidential Turn — and Why It Changes Everything
Paul Draper's 1989 paper "Pain and Pleasure: An Evidential Problem for Theists" shifted the ground permanently. Draper was not asking whether God's existence is logically incompatible with suffering. He was asking a simpler, more devastating question: given what we actually observe about the distribution of pain and pleasure across biological life, which hypothesis makes the observed data more probable — theism or indifference?
His answer: indifference wins. The distribution of suffering in the natural world — the ratio of pain to pleasure in the animal kingdom, the randomness of who suffers and who does not, the fact that most creatures that have ever lived died in agony through predation, starvation, or disease — is precisely what you would expect if no mind were attending to the process. It is not what you would expect of a mind that is omnipotent, omniscient, and wholly benevolent.
This is the evidential problem. Not "why does evil exist" — but "why does the pattern of evil exist?" The pattern is what makes free will insufficient as an answer. A theodicy that explains individual instances of suffering through free will or soul-making does not explain why a parasitic wasp lays its eggs inside a living caterpillar so that its larvae can feed from the inside out, consuming non-vital organs first to keep the host alive as long as possible. No human freedom is at stake. No soul is being refined. The caterpillar has no theology.
III. Natural Evil and the Limits of Theodicy
Theodicy — the project of justifying God's ways in the face of evil — has been the most ambitious sustained effort in philosophical theology for two thousand years. Its greatest practitioners — Augustine, Aquinas, Leibniz, Plantinga, Swinburne — are formidable minds. Their arguments deserve respect. They do not deserve the credulity they have not earned.
Augustine proposed that all natural evil is either the consequence of the Fall or permission for greater goods we cannot perceive. Swinburne argues that natural suffering provides the epistemic distance necessary for genuine faith — that a world in which God's presence was self-evident would eliminate freedom of belief, and therefore freedom itself. These are serious positions. They are also, in their specific claims, unfalsifiable by design.
The critical concept is gratuitous evil — suffering that serves no greater purpose, that no justifying reason could account for. William Rowe (1979) argued that such evils exist — a fawn dying slowly over five days in a forest fire started by lightning, with no human witness, no soul to be refined, no freedom at stake — and that the probable existence of even one genuinely gratuitous evil is sufficient to make theism improbable.
The theistic response — called "skeptical theism" — holds that we cannot know whether any evil is genuinely gratuitous, because we cannot perceive the full causal-moral structure of the universe. This is logically available. But it purchases its victory at a cost: it concedes that our moral intuitions cannot be trusted to track reality. And if our moral intuitions cannot be trusted when we judge suffering to be gratuitous, it is not obvious why they should be trusted when we judge anything to be good — including the goodness of God.
IV. The Scale Is Not a Detail — It Is the Argument
It is tempting to treat the problem of evil as a puzzle about edge cases — about rare, extreme, or philosophically constructed scenarios. The fawn in the forest. The child with cancer. These are not edge cases. They are the norm. Let us be precise about the norm.
The problem is not merely that children die. The problem is that this is statistically predictable. It is not the result of anomalous cruelty or random catastrophe. It is the structural output of a biological and ecological system operating exactly as designed — or exactly as undesigned. In either case, we are left with an enormous inferential task.
If a good God designed the system, then the suffering of innocents was anticipated, chosen, and permitted on a daily basis by an omnipotent agent. If no God designed the system, the suffering is terrible but not a theological problem. The question is which of these two explanations better fits the data. That is not an emotional question. It is an evidential one.
The philosopher Marilyn McCord Adams has argued for a "defeat" model of theodicy — that God does not prevent all horrors but works within them, defeating their negative meaning through ultimate redemption in the eschaton. This is a sophisticated and theologically mature position. It requires, however, a level of confidence about life after death, about the nature of ultimate reconciliation, and about the moral economy of the universe that the observable evidence simply does not supply. It asks us to trust a God who has, in the observable record, remained silent during the deaths of a hundred trillion conscious creatures.
V. Where the Argument Stands — and What It Demands of You
The problem of evil does not prove God's non-existence. It does not need to. What it does is establish that the prior probability of theism, given the full evidential picture of biological life, is significantly lower than it would be if suffering were distributed as a benevolent designer's product should be. That is a precise, limited claim. It is also an honest one.
The honest epistemic position here is not comfortable. The theodicies are serious. The problem is serious. Neither side can claim decisive victory. But the person who believes in a good God and looks squarely at the fossil record, at the distribution of childhood mortality, at the behavior of the parasitic world — that person is carrying a burden of explanation that cannot be discharged by sentiment, by tradition, or by the observation that the question is hard.
Hard questions demand honest answers. The argument from evil is not an attack on God. It is an argument about what the evidence says — and what it does not say. Truth is knowable. But it demands the courage to look at the evidence without flinching. We are the ones who choose whether to look.
The following sources constitute the primary intellectual foundation for this argument. Each is cited in its strongest form — the version that most challenges the opposing position. Read the strongest versions of both sides before forming a conclusion.
- Draper, P. (1989). "Pain and Pleasure: An Evidential Problem for Theists." Noûs 23(3), 331–350. The seminal paper establishing the evidential form of the problem. Draper introduces the Hypothesis of Indifference and argues it explains the pain/pleasure distribution better than theism. The cleanest formulation of the evidential argument. Search this source ↗
- Rowe, W.L. (1979). "The Problem of Evil and Some Varieties of Atheism." American Philosophical Quarterly 16(4), 335–341. Introduces gratuitous evil via the fawn thought experiment. Argues that the probable existence of even one genuinely gratuitous evil is sufficient to make theism improbable. Rowe later moderated his position — the evolution of his thinking is itself instructive. Search this source ↗
- Plantinga, A. (1974). The Nature of Necessity. Oxford University Press. The definitive statement of the Free Will Defense. Required reading for anyone who wishes to understand why Mackie's logical formulation did not end the debate. The theistic counter-move that forced the evidential turn. Search this source ↗
- Mackie, J.L. (1955). "Evil and Omnipotence." Mind 64(254), 200–212. The classic logical problem of evil. Read this before reading Plantinga. The argument Mackie thought was decisive, and the response that demonstrated he had underestimated the options. Search this source ↗
- Swinburne, R. (1998). Providence and the Problem of Evil. Oxford University Press. Argues that natural evil provides the epistemic distance necessary for genuine faith. The most systematic theistic treatment of the problem. Disagree with it thoughtfully — don't dismiss it easily. Search this source ↗
- Adams, M.M. (1999). Horrendous Evils and the Goodness of God. Cornell University Press. Proposes the "defeat" model — God does not prevent all horrors but defeats their meaning through eschatological redemption. The most sophisticated contemporary theodicy. Engage it seriously. Search this source ↗
- UNICEF. (2024). The State of the World's Children. United Nations Children's Fund. Current empirical data on child mortality. The statistical dimension of suffering as a matter of public record — not as polemics, but as the datum that any theodicy must account for. Read source ↗
Where Does This Argument Lead You?
Select the conclusion that most honestly reflects your assessment of this evidence.
The Problem of Evil is the most personally felt objection to theism — and the hardest to answer from a distance. But the evidence arc doesn't leave you here. The NOW side shows that the machinery of life — including suffering — required intentional architecture. And the historical record names the Architect who entered the suffering himself.