HistoricalChristian.Faith

The logical problem of evil and the free-will defense — does it answer Mackie?

Type: supporting argument (off-timeline) · Assessment: sound — as an answer to the logical problem the free-will defense holds up, and holds up by near-consensus: Plantinga need show only that it is possible a good, omnipotent God could not actualize a world with moral good but no moral evil (the “transworld depravity” possibility), and that bare possibility is enough to break Mackie’s claim that the theist’s beliefs are self-contradictory. The critics conceded the point — Mackie himself in 1982, and Rowe explicitly — and the argument then moved to the evidential ground the defense never claimed to hold. Be precise about the scope: this answers only the logical form, not the evidential problem of apparently pointless suffering; and its extension to natural evil rests on a far more speculative move (that natural evil may be the free work of fallen non-human spirits) which is adequate for a defense — it need only be possible — but carries no independent probability. · advanced by Alvin Plantinga, with roots in Augustine and the Greek fathers

Related arguments: the modern analytic problem of evil — the evidential problem of evil · skeptical theism

See also: the modern-debate survey that sets Mackie, Plantinga, Rowe, and skeptical theism side by side; the ancient free-will answer of Irenaeus and Basil, and Augustine’s deficient-cause account of the bad will, of which this is the modern modal form. The extension to natural evil connects to the divine council reading of a rebel cosmos.

The claim

The logical problem of evil, in J. L. Mackie’s 1955 statement, is not that evil makes God unlikely but that the theist’s own beliefs are logically inconsistent: “God is omnipotent; God is wholly good; and yet evil exists,” and “if any two of them were true the third would be false.”¹ The charge is the hardest possible one: not that the theist is probably wrong, but that no possible reason could reconcile an omnipotent, wholly good God with any evil at all, so belief is incoherent from the start.

Alvin Plantinga’s free-will defense (God, Freedom, and Evil, 1974) replies that the triad is consistent, because a further proposition is at least possibly true and, together with the three, entailment-free of contradiction:

A world containing creatures who are significantly free (and freely perform more good than evil actions) is more valuable, all else being equal, than a world containing no free creatures at all… To create creatures capable of moral good, therefore, He must create creatures capable of moral evil; and He can’t give these creatures the freedom to perform evil and at the same time prevent them from doing so.

— Alvin Plantinga, God, Freedom, and Evil (1974).² Two features are decisive and easily missed. First, it is a defense, not a theodicy: Plantinga does not claim to give God’s actual reason for permitting evil, only to exhibit a possible one — and possibility is all that answering an inconsistency charge requires (to show that “p and q” are consistent it suffices to find a possible r that is consistent with p and, with it, entails q; the r need not be true or even plausible, only possible). Second, the load-bearing possibility is transworld depravity: it may be that every creaturely essence is such that, in any world in which that creature is significantly free, it would go wrong at least once. If that is so much as possible, then it is possible that God, though omnipotent, could not have actualized any world containing moral good but no moral evil — and Mackie’s alleged contradiction is broken.³

Who advances it

The reading is old; the modern form is Plantinga’s. The free-will answer to evil is the earliest sustained Christian reply — Irenaeus lodges moral evil in a creature “made free from the beginning,” Basil makes each soul “the first author of his own vice,” and Augustine traces the bad will to a deficient, not an efficient, cause, so that God is in no sense evil’s author while the creature remains justly answerable. What Plantinga added is rigour: recast in the modal logic of possible worlds, the ancient claim “God could not make a free creature that could not sin” becomes a precise counter to a precise charge. The formal machinery — counterfactuals of creaturely freedom, feasible versus merely possible worlds, transworld depravity — is the same apparatus that drives the middle-knowledge debate, and Plantinga is a Molinist about it. The defense is now the standard reference reply in analytic philosophy of religion, expounded in every textbook and encyclopedia treatment of the logical problem.

The evidence for

The critics conceded it. This is the argument’s real strength, and it is unusual: the philosopher who set the logical problem in its hardest twentieth-century form later granted that the defense answers it. Mackie, in The Miracle of Theism (1982), wrote that “since this defence is formally [that is, logically] possible… we can concede that the problem of evil does not, after all, show that the central doctrines of theism are logically inconsistent with one another.”⁴ William Rowe, opening the case for the rival evidential argument in 1979, said the same from the other side: “Some philosophers have contended that the existence of evil is logically inconsistent with the existence of the theistic God. No one, I think, has succeeded in establishing such an extravagant claim.”⁵ Surveys of the field record the settled verdict bluntly: “all parties admit that Plantinga’s Free Will Defense successfully rebuts the logical problem of evil as it was formulated by atheists during the mid-twentieth century.”⁶

The logic is airtight for what it attempts. An inconsistency charge is defeated by a single consistent model. Grant — as even the atheist must, since it is a claim only about possibility — that libertarian freedom is a great good, that a genuinely free creature is one God does not causally determine, and that it is possible every such creature would misuse its freedom at least once. Then the four propositions (God is omnipotent, God is wholly good, God creates significantly free creatures, every such creature would freely go wrong at least once) are jointly consistent and jointly entail that evil exists. So the original triad cannot be self-contradictory, since a consistent set entailing it exists. Nothing here needs to be actually true; the whole weight rests on possibility, which is exactly the modest thing a defense claims.

It answers Mackie’s specific counter. Mackie’s sharpest move was the compatibilist one: if a man can freely choose the good on one occasion, “there cannot be a logical impossibility in his freely choosing the good on every occasion,” so God could have made “beings who would act freely but always go right,” and His failure to do so is inconsistent with omnipotence and goodness (Antony Flew argued the same in the same 1955 volume).⁷ Plantinga’s reply is that this quietly assumes compatibilism — that a free act can be causally determined by God — and the theist need not grant it. On libertarian freedom, which the defense presupposes, God cannot both leave a creature free with respect to an action and guarantee its outcome, because a determined choice is not a free one. And whether God could actualize a world of always-right free creatures is not up to Him alone: it depends on the counterfactuals of freedom, which may (transworld depravity) be uncooperative for every possible creature. Mackie’s “obviously better possibility” is therefore not obviously available — and need not be, for the theist to escape the contradiction.

Adversarial assessment

What holds. Against the logical problem the defense is decisive, and its opponents say so. It shows what it set out to show — consistency — and it does so with a genuinely modest premise (mere possibility), so the usual complaint against theodicies (that their proposed reason is implausible) does not touch it: a defense is not answerable for the plausibility of its r, only its possibility. The compatibilist rejoinder does not defeat it, because the theist may hold libertarian freedom without contradiction, and on that view the always-right world is not something omnipotence can simply mint.

What is narrower than it sounds. The victory is real but tightly scoped, and it is easy to overstate into “the free-will defense solves the problem of evil.” It does not. Three limits bound it:

• Only the logical, not the evidential, problem. Rowe, Mackie, and the rest granted consistency and immediately relocated the dispute to the evidential ground: not “is theism self-contradictory?” but “does the sheer scale and apparent pointlessness of suffering make God’s non-existence probable?” The free-will defense, which trades only in possibility, says nothing to a probabilistic argument, and Plantinga never claimed it did. A verdict of “sound” here is a verdict about the logical problem alone.
• Natural evil needs a separate, weaker move. The free-will defense covers moral evil (the wrong that wills do); it says nothing directly about natural evil (disease, disaster, animal suffering), which no human freedom produces. Plantinga extends it by a further possibility: that natural evil is the free work of non-human spirits — Satan and his cohorts — whose sin God permits on the same free-will terms.⁸ This is logically adequate for a defense — it need only be possible to block the natural-evil form of the inconsistency charge — and it is continuous with a genuine strand of Christian tradition (the fallen-angel reading this wiki weighs on the divine council timeline). But it carries no independent probability, and Plantinga does not offer it as probable; treated as anything more than a bare possibility it is far weaker than the moral-evil core, and critics reasonably find it the defense’s most strained extension. Keep the two ratings distinct: the moral-evil answer is sound against the logical problem; the natural-evil extension is merely a possibility, sufficient for consistency and nothing more.
• Consistency is not comfort, and not a theodicy. Showing that the propositions can all be true leaves wholly open why they are — the positive reason a theodicy would supply. The Fathers’ constructive answers (greater good, soul-making, privation, providence) and the modern evidential debate are where that work is done; the defense clears the logical ground and then hands off. Confusing the two is the commonest overreach, and a proponent who says the defense “answers the problem of evil” full stop should be corrected on exactly this point.

A residual objection worth naming. Some (e.g. from a broadly compatibilist or theological-determinist camp) still deny the libertarian premise the whole defense rests on, holding that God can determine a free act — in which case Mackie’s always-right world is back on the table and the defense fails. This is a real dividing line, but it does not lower the rating of the interpretation the page defends: the free-will defense is explicitly an argument from libertarian freedom, and against the logical problem it succeeds given its own premise, which is all a defense needs. Whether libertarian freedom is the correct account is a separate question, argued on the predestination timeline, not a defect in this argument.

Assessment

Sound. As an answer to the logical problem of evil the free-will defense holds up, and by something close to consensus: it needs only the possibility that God could not actualize moral good without moral evil, that possibility suffices to defeat an inconsistency charge, and the very philosophers who pressed the charge — Mackie in 1982, Rowe in 1979 — conceded that it does. The compatibilist counter (God could make free creatures who always choose the good) does not overturn it, because the theist may hold libertarian freedom, on which a guaranteed free choice is a contradiction in terms. What the rating does not license is the slide to “solves the problem of evil”: the defense answers only the logical form, is silent before the evidential argument from apparently pointless suffering, and reaches natural evil only through the much weaker (though, for a defense, logically sufficient) conjecture of fallen non-human agency. Rate the moral-evil core sound against the logical problem; keep the natural-evil extension marked as a bare possibility; and flag any framing that presents the defense as a complete theodicy as an overstatement of a narrow, genuine success.

Sources