You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
The problem of evil is widely regarded as the most formidable objection to belief in the God of classical theism. If God is omnipotent (all-powerful), omniscient (all-knowing) and omnibenevolent (all-loving), why is the world saturated with evil and suffering? The difficulty is that each of God's traditional attributes seems, on its own, to commit God to preventing evil: an all-loving God would want to, an all-powerful God would be able to, an all-knowing God would know how and that it occurs. Yet evil persists. This lesson frames the problem rather than resolving it. It sets out the two great forms the problem takes — the logical problem (Mackie's inconsistent triad) and the evidential problem (Rowe) — secured on the basic distinction between natural and moral evil, and then signposts (without developing in detail) the three responses the AQA 7062 specification names: the free will defence, Hick's soul-making theodicy, and process theodicy as presented by Griffin. Those responses are studied in depth in their own right; here the task is to understand, as sharply as possible, the problem they are trying to answer.
Before stating the problem we must distinguish two kinds of evil, because the distinction shapes every response to it.
| Type | Definition | Examples |
|---|---|---|
| Moral evil | Suffering brought about by the free, deliberate choices of moral agents | Murder, theft, torture, genocide, slavery, war |
| Natural evil | Suffering arising from the operation of the natural world, independent of human will | Earthquakes, tsunamis, disease, drought, congenital disability, animal predation |
Key term: Moral evil — evil that results from the morally significant free actions of agents (human, or in some accounts angelic).
Key term: Natural evil — evil that arises from the natural order — natural disasters, disease, decay — rather than from any agent's choice.
The distinction matters because the most popular response — the free will defence — has obvious purchase on moral evil (it locates the cause in misused freedom) but struggles with natural evil, which is not the product of any creature's choice. An earthquake that kills thousands cannot, at least not straightforwardly, be blamed on human wrongdoing. So when assessing any response, a key test is always: does it cover natural evil as well as moral evil? It is also worth noting a grey zone: some suffering is jointly produced — a famine caused by drought (natural) and worsened by human greed and mismanagement (moral) — which complicates any neat division.
Natural evil is the harder case, and it is worth seeing why. Three strategies are commonly tried to fold it into a free-will framework, and each is contestable. (1) Some (following Augustine's broad approach) attribute natural evil to a cosmic misuse of freedom — the fall of angels, or the disordering of nature consequent on the human Fall. The difficulty is that the geological and fossil record shows predation, disease and mass extinction long predating any human wrongdoing, so a historical human Fall cannot be the cause of natural evil. (2) Others argue that natural evil is the inevitable by-product of a law-governed world: a universe stable and regular enough to support free, rational agents — one in which fire warms and also burns, in which tectonic plates build continents and also cause earthquakes — must contain the possibility of natural harm. This is more promising but raises the question why an omnipotent God could not have fine-tuned the laws to spare the innocent. (3) Hick's soul-making approach (signposted below) treats natural evil as a necessary environment for moral and spiritual growth. The point for now is simply that natural evil cannot be waved away as "really" moral evil; it presses the problem in its own right, and any adequate response must address it directly.
The problem is also ancient and persistent, which is itself significant: it is not a modern sceptical invention but a question the great theistic traditions have always felt the force of from within. The biblical Book of Job is an extended wrestling with the suffering of a righteous man; the Psalms repeatedly cry "how long, O Lord?"; and philosophers from Epicurus to the present have pressed it. This matters for evaluation because it shows that the problem is not merely an external attack to be repelled but a genuine tension internal to belief in a good and powerful God — which is why theists have devoted such enormous effort to it.
The logical (or a priori, or deductive) form of the problem claims that theism is internally inconsistent: the very existence of evil is logically incompatible with the existence of the theistic God, so the believer holds a self-contradictory set of beliefs. It does not appeal to how much evil there is; even a single instance would, if the argument works, suffice. This is the strongest form of attack a critic can mount, because if it succeeds it shows belief in God to be not merely improbable but impossible — as impossible as belief in a round square. For the same reason it is the most vulnerable form, because the theist needs only to find one coherent way for God and evil to coexist to break the alleged contradiction. The history of the debate is, in large part, the history of this attack being pressed (Epicurus, Mackie) and then widely judged to have been parried (Plantinga), after which the action moves to the evidential battlefield.
The shape of the problem is ancient. It is traditionally traced to Epicurus (341–270 BCE) — the formulation is preserved and transmitted by later writers such as Lactantius — and is sometimes called the "Epicurean paradox" or trilemma. Its logic can be set out as a dilemma about God's will and power:
The power of the formulation is that it works by elimination: it takes each combination of "able/not able" and "willing/not willing" and shows that every option except the one the theist needs (able and willing) is consistent with the evil we observe — while the option the theist needs seems inconsistent with it. The theist must therefore explain how a God who is both able and willing to prevent evil can coexist with a world full of it. Everything turns on whether there is some further consideration — a reason God might have for permitting evil — that the bald trilemma overlooks. This is exactly the gap the responses try to fill.
The decisive modern statement is J.L. Mackie's in his paper "Evil and Omnipotence" (1955). Mackie argues that the theist is committed to three propositions that form an inconsistent triad — any two can be held together, but not all three:
The propositions are not explicitly contradictory, Mackie concedes; the contradiction emerges once we add what he calls "quasi-logical" bridging principles that the theist also accepts: that a good thing always eliminates evil as far as it can, and that there are no limits to what an omnipotent being can do. Granting these, a wholly good, omnipotent God would eliminate evil entirely — so if such a God existed, evil would not. Since evil plainly does exist, Mackie concludes that at least one of the three propositions must be abandoned: either God is not omnipotent, or not wholly good, or does not exist. The believer who affirms all three is, he charges, being positively irrational.
Key term: Inconsistent triad — Mackie's claim that the three propositions "God is omnipotent," "God is wholly good," and "evil exists" cannot all be true together (given the bridging principles), so theism is logically incoherent.
Key term: Theodicy — from the Greek theos (God) and dikē (justice): an attempt to justify the goodness and power of God in the face of evil, typically by giving a reason why God permits it. (A "defence," by contrast, aims only to show theism is not inconsistent, without claiming to give God's actual reasons.)
Mackie also anticipates and rejects several "fallacious solutions" — escape routes he thinks fail. To the suggestion that evil is necessary as a counterpart to good (we could not recognise good without evil, just as we could not see "great" without "small"), he replies that this limits God's power: an omnipotent God should be able to make good recognisable without real suffering, and in any case the argument would only justify a little contrasting evil, not the appalling quantities we actually find. To the claim that evil is necessary as a means to good, he objects that this concedes God is bound by causal laws like a limited craftsman, contradicting omnipotence. To the grandest reply — that the universe is better with "second-order goods" (courage, compassion, forgiveness) that logically require first-order evils (danger, suffering, wrongdoing) to exist — he responds that this generates second-order evils too (cowardice, cruelty, malice), and the theist has no account of why an omnipotent God permits those.
His sharpest challenge, however, is reserved for the free will reply. Grant, says Mackie, that freedom is a great good worth some risk of evil. Still: if it is logically possible for a free being to choose the good on any given occasion, then it is logically possible for a free being to choose the good on every occasion — there is no contradiction in the idea of a person who is genuinely free and yet always, freely, does right. An omnipotent God could therefore have created only such people: beings with free will who never in fact misuse it. Since God did not, either God could not (so God is not omnipotent) or God would not (so God is not wholly good). The free will defence, Mackie concludes, fails to explain moral evil. (It is precisely this challenge that Plantinga's free will defence is designed to answer — see the signpost below — by denying that it was within God's power to actualise a world of free creatures who always freely do right.)
It is worth dwelling on the structure of Mackie's case, because the strongest answers grasp why it is a logical problem. Mackie is not appealing to how much evil exists, nor to any particular horror; his claim is that the bare coexistence of God and any evil is contradictory, once the bridging principles are granted. That is why a single successful counter-instance — one logically possible scenario in which God and evil coexist — would suffice to defeat it. This is exactly what the free will defence (as opposed to a full theodicy) sets out to provide: not God's actual reason for permitting evil, but merely a possible one, which is all that is needed to break a charge of inconsistency.
The evidential (or inductive, or probabilistic) form of the problem is more modest in its logic but, many think, harder to answer. It concedes — as most philosophers now do, largely because of Plantinga's free will defence — that the logical problem can be met: it is not strictly contradictory to hold that God and evil coexist. But conceding consistency is not conceding much. The evidential problem argues that the amount, intensity, distribution and apparent pointlessness of suffering in the world make the existence of God highly improbable, even if not logically impossible. Evil is not a knock-down disproof; it is powerful evidence against theism, in the way that a thousand failed experiments are evidence against a hypothesis without strictly refuting it.
The shift from logical to evidential is best understood as a strategic retreat-and-regroup by the critic. The logical problem aimed for a proof that theism is incoherent; the free will defence blocked it by supplying a possible reason for evil. The evidential critic responds: very well, it is possible that every evil serves a greater good — but is it plausible? Look at the kinds and quantities of suffering: the millennia of animal agony before humans existed, the routine devastation of natural disasters, the torture of children. Is a world like this what we would expect if it were governed by an omnipotent, omniscient, perfectly loving God? The evidential problem says: no — such a world is far more probable on the hypothesis of no (good, all-powerful) God, and that probabilistic asymmetry is real evidence against theism. The burden is thus thrown back on the theist not merely to show consistency but to make the evidence unsurprising.
The clearest version is William Rowe's (1931–2015). Rowe focuses on gratuitous (pointless) suffering — suffering that serves no greater good, that is not logically necessary for any outweighing benefit, and that even an omnipotent being could have prevented without thereby losing some greater good or permitting some equally bad evil. His argument runs:
Key term: Gratuitous evil — suffering that brings about no outweighing good and is not necessary for any greater good; pointless suffering that an omnipotent being could have prevented at no cost. Its existence is the crux of the evidential problem.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.