Theism lacks a clear symmetry-breaker
The argument examined here is not the deductive claim that the existence of suffering is formally inconsistent with God's existence. That is the logical problem of evil. Plantinga's work made it difficult to defend a simple contradiction from the standard divine attributes and the existence of evil, although the status of all versions of the logical argument remains debated. The argument developed here is evidential: the amount, distribution, and apparent pointlessness of suffering seem less expected if an all-powerful, all-knowing, perfectly good God exists than they would be on at least some naturalist hypotheses.
The proposed reply is often called skeptical theism. It says that human beings are badly placed to judge whether God would have sufficient reasons for permitting particular evils. Our knowledge is limited, our perspective is local, and the reasons available to an omniscient being could involve consequences far beyond our understanding. Therefore, the fact that we cannot see a justifying reason for suffering does not strongly support the conclusion that no such reason exists.
The symmetry objection accepts this reply for the sake of argument, but asks whether it can be used consistently. If our inability to see God's reasons prevents us from inferring that suffering counts against a good God, why should the same inability prevent someone from saying that suffering is compatible with naturalism, or even with a malevolent creator? If unknown reasons can preserve theism from disconfirmation, what independently identifies the good-God hypothesis as the more probable explanation?
The strongest case for the argument
The argument is strongest when framed as a challenge to comparative probability, not as a proof that theism is impossible.
Consider three broad hypotheses:
- Naturalism: reality contains no personal creator or cosmic moral governor. Suffering occurs through impersonal physical, biological, and social processes.
- Good-God theism: reality is governed by an omnipotent, omniscient, and perfectly good God, who permits suffering only for morally sufficient reasons.
- Evil-God theism: a deliberately constructed hypothesis on which an omnipotent, omniscient, and perfectly evil personal being permits or produces good only for sufficiently bad reasons. It is not a standard religious worldview, and whether a perfectly evil being is coherent depends partly on how moral predicates are understood; its role here is comparative and diagnostic, not as an independently established rival.
The point is not that these hypotheses are equally plausible or that naturalism has intentions. It is that a general appeal to inscrutable reasons cannot by itself do all the argumentative work for good-God theism. When suffering seems unexpected on good-God theism, the theist says that we should not expect to know all of God's reasons. An evil-God theorist could say that apparent goods serve hidden destructive purposes. A naturalist need not posit hidden purposes at all; the naturalist instead asks whether the observed pattern follows from impersonal laws, evolutionary history, and social conditions, while recognizing that the ultimate background may remain uncertain. Thus the hypotheses are not perfectly symmetrical, but the mere possibility of unknown purposes does not show that good-God theism is favored.
This is not a claim that naturalism literally has intentions or reasons. It is a claim about explanatory underdetermination. If the evidence is judged too limited to discriminate between visible suffering and the purposes of a good deity, then it may also be too limited to discriminate between visible suffering and the absence of deity. The same epistemic modesty that blocks a negative inference may also block a positive one.
There is a Bayesian version of the point. Evidence confirms a hypothesis when the evidence is more expected on that hypothesis than on relevant alternatives. Merely showing that suffering is compatible with good-God theism is not enough, because almost any event can be made compatible with a hypothesis by adding sufficiently unspecified background conditions. The relevant comparison concerns the likelihood of the observed amount, distribution, and apparent pointlessness of suffering given the competing hypotheses and their background assumptions. Priors are not magical numbers: they may reflect considerations such as simplicity and scope, as well as arguments supplied elsewhere in the cumulative case. Skeptical theism therefore does not by itself supply either a high prior probability or a favorable likelihood for good-God theism.
Philosophers such as William Rowe and Paul Draper have developed influential versions of the evidential problem of evil, while Stephen Law has used the idea of an evil God to challenge asymmetrical treatments of good and evil. The serious literature does not generally regard the evil-God hypothesis as an established rival worldview. Its value is diagnostic: it tests whether a principle is being applied neutrally. If "we cannot see the reasons" is sufficient to protect one hypothesis from the evidence, why does the same principle not protect a structurally contrary hypothesis?
What the objection gets right
It correctly exposes a weakness in a popular apologetic argument. Saying that God may have an unknown reason for every instance of suffering does not by itself make good-God theism probable. It can function as a logical defence: the theist may be entitled to say that no contradiction has been demonstrated. But avoiding contradiction is weaker than explaining the evidence.
This distinction matters. Alvin Plantinga's free-will defence, for example, was designed primarily as a response to the logical problem of evil. In its modal form, it argues that it is possible for God to have a morally sufficient reason to create significantly free creatures even if every feasible world containing such creatures includes some moral wrongdoing. That is an important achievement against the claim that God and evil are strictly inconsistent. It is not a probabilistic explanation of the actual quantity or distribution of suffering, and it does not by itself address natural evil, animal suffering, or apparently gratuitous harms.
The symmetry objection also identifies a danger in unrestricted skeptical theism. If our cognitive limitations are invoked whenever evidence appears unfavorable, the position can begin to weaken ordinary reasoning about divine intentions. We may no longer be entitled to infer that a good God would oppose cruelty, protect the vulnerable, or prefer less suffering, because any apparent counterexample could be assigned an unknown reason. Critics of skeptical theism have therefore argued that some formulations threaten moral reasoning or produce a kind of moral skepticism. Defenders reply that the view need not reject all moral knowledge; it may deny only that we can infer God's reasons from particular cases. The details are disputed, but the concern is serious: a skepticism broad enough to neutralize the problem of evil may also threaten confidence in inferences about what God would do.
Most importantly, the objection establishes a real burden of proof. If the theist uses skeptical theism to argue that suffering should not substantially lower confidence in good-God theism, the theist needs more than the bare possibility of hidden reasons. They need some independent symmetry-breaker: a reason why the good-God hypothesis has greater prior plausibility, greater explanatory scope, or better fit with other evidence than naturalism and the evil-God hypothesis. This need not be a deductive proof or a single decisive argument; it may be a cumulative case from contingency, consciousness, reason, morality, religious experience, or other evidence. But those arguments must be assessed independently rather than smuggled in through the appeal to hidden reasons.
The strongest reply
The strongest theistic response is that the hypotheses are not actually symmetrical. Goodness and evil are not interchangeable labels attached to otherwise identical theories. A perfectly good being might permit suffering for the sake of freedom, stable natural laws, the development of virtue, the prevention of greater harms, or goods that require cooperation and risk. An evil being, by contrast, would have reason to prevent or corrupt goods wherever possible, though it might also permit goods instrumentally in order to intensify later suffering or corruption. If the world contains substantial and intelligible goods - including love, knowledge, moral concern, beauty, and the capacity to resist suffering - perhaps it fits good-God theism better than evil-God theism. But this is only a proposed asymmetry, not a conclusion from the bare existence of goods. The theist must still explain why the relevant goods require, or are evidentially compatible with, this amount, distribution, and intensity of suffering.
A theist might also argue that naturalism is not a complete explanatory competitor. Naturalism may describe how suffering arises through evolutionary, physical, psychological, and social processes, but it may not, on its own, answer why there is a law-governed universe, why conscious subjects exist, or whether objective moral obligations are real. Arguments from contingency, consciousness, reason, and moral realism could therefore provide independent evidence for theism. These are disputed philosophical questions, however, and an explanatory gap is not automatically evidence for God. Each argument must be assessed on its own merits and then compared with naturalist accounts and their costs. On this approach, suffering is not considered in isolation: it is one piece of a cumulative case, and its force may be outweighed by other considerations only if those other considerations genuinely support theism.
There is also an important difference between an evil-God hypothesis and naturalism. Evil-God theism adds a cosmic agent with a specific moral character and therefore owes an account of why its world contains the kinds of goods it supposedly opposes. Naturalism does not posit suffering as something intentionally permitted; it can seek causal explanations in evolution, physiology, psychology, and social processes, while treating some ultimate facts as brute or unsettled. This does not make naturalism complete or automatically probable, but it means that naturalism and evil-God theism are not mirror images. The theist must compare their explanatory costs rather than assume that the absence of divine purpose is itself evidence for a malicious purpose.
These are genuine replies, not evasions. A theory should be assessed against its total evidence, and the hypotheses are not automatically equal merely because each can be made logically compatible with suffering. The symmetry objection is strongest against arguments that treat "possible hidden reasons" as the whole case for theism. It is weaker against a theist who offers independent arguments for God's existence and then argues that suffering is not sufficiently improbable given the larger theory.
How the replies fare
The appeal to goods creates some asymmetry, but probably not a decisive one on its own. A good God could be expected to value love, freedom, and moral development, yet an all-powerful God might seemingly achieve many of these without permitting horrors on the scale observed. An evil God could also permit goods instrumentally, in order to make later suffering possible or to corrupt otherwise good agents. The mere presence of good and evil therefore does not straightforwardly identify the world's moral governor.
Naturalism also cannot be dismissed as simply "having no explanation." It can offer causal explanations for many forms of suffering, even if it does not provide a final metaphysical account of why there is a universe or why moral experience exists. Those are separate questions. Nor is an unexplained fact automatically evidence for theism. The existence of consciousness or objective morality may support theism for some philosophers, but both claims remain contested in philosophy and science.
Consequently, the symmetry objection does not prove naturalism, and it does not show that evil-God theism is as plausible as good-God theism. It establishes something narrower and important: suffering alone, combined with an unrestricted appeal to unknown divine reasons, does not establish good-God theism as the best explanation. The theist must supply an independent reason to prefer that hypothesis, while a critic of theism need not claim that the problem of evil is a deductive refutation; its force may be probabilistic, comparative, and cumulative.
The fair conclusion is therefore one of underdetermination, not evidential neutrality. Skeptical theism may successfully block an overconfident inference from "I see no reason for this suffering" to "there is no sufficient reason." But unless it is supplemented by independent evidence, it does not convert compatibility into probability. The debate remains open, with the argument from evil retaining substantial force; whether that force is outweighed depends on the strength of the independent arguments offered for theism and on how they compare with naturalist alternatives. Theism remains responsible for showing that its proposed symmetry-breaker is more than a mere possibility.