Dissolving moral criticism may make divine goodness empty

One influential objection to theism begins with a tension about the meaning of moral language. Theists often describe God as perfectly good, just, loving, and opposed to cruelty. Yet when difficult cases arise - especially suffering, apparently pointless evil, or divine commands that seem morally troubling - a familiar reply is that God's nature and purposes are so far beyond human understanding that ordinary moral judgments cannot simply be applied to God.

That reply may be reasonable in some contexts. Human beings are limited, and it does not follow from our inability to see a sufficient reason for an event that no such reason exists. But it also creates a serious philosophical danger. If concepts such as goodness, justice, and love become too unlike their ordinary meanings when applied to God, then the claim that God is perfectly good may cease to say anything definite. If those concepts retain enough of their ordinary content to make the claim meaningful, however, then moral criticism of apparently cruel or unjust actions remains a legitimate challenge.

The strongest form of the objection

The argument is not that God must be morally identical to an ordinary human agent. Nor does it assume that human beings possess complete knowledge of every morally relevant fact. Its stronger form is a conditional challenge about the conditions under which moral predication has content:

  • If goodness and justice apply to God in a stable and relevant sense, then we may ask whether particular divine actions are consistent with those properties. The existence of apparently gratuitous suffering, or the command of apparently cruel acts, cannot be dismissed merely by saying that God is God.
  • If goodness and justice have no recognisable relation to the standards those words ordinarily convey, then calling God "perfectly good" risks becoming little more than an honorific. It may express devotion or approval without describing a property that distinguishes a good God from a morally indifferent, deceptive, or cruel one.

This is not an exhaustive choice between human morality and total unintelligibility. A theist may appeal to analogy, arguing that divine goodness has a different mode of application while preserving enough structure for the claim to be informative. The challenge is then to identify what content is preserved and why that content supports both praise of God and criticism of apparent cruelty.

Critics sometimes call this a "trap," but the label should not obscure the argument's limited target. The dilemma is not a proof against every form of theism; it challenges accounts that appeal to ordinary moral language when praising God and then withdraw all intelligible standards when suffering is raised as an objection. A theist can reject both horns by defending an analogical account of goodness, but must then explain which features of goodness remain stable and why they make divine goodness more than a label immune to criticism.

The concern has roots in several philosophical debates. J. L. Mackie famously defended a logical problem of evil, arguing that traditional claims about omnipotence, perfect goodness, and evil generate an inconsistency unless one of those claims is qualified. Many later philosophers have judged that strict logical argument less successful, especially in light of free-will and other theodicies. William Rowe developed an influential evidential argument from apparently pointless suffering, while Paul Draper compared the distribution of pain and pleasure under the hypothesis of an indifferent natural order with what might be expected under theism. These arguments do not assume that humans understand every divine purpose. They claim instead that some suffering appears to lack any morally sufficient justification, and that this appearance can have evidential significance without amounting to certainty.

The semantic point also resembles a concern raised in debates about divine command theory, although the theories should not be caricatured. Divine command theories typically concern the source of moral obligation, not the claim that every feature of goodness is created by a command. A crude version in which an act becomes good solely because God commands it faces the familiar arbitrariness objection: the command would explain why the act is required, but not why God's will is worthy of trust. Modified versions answer that God's commands are grounded in a loving and necessarily good nature. They avoid simple arbitrariness, but they also make an independent account of divine goodness essential. A similar problem arises if divine goodness is defined so flexibly that every possible divine action is automatically compatible with it.

Why the critique has real force

The argument exposes a genuine asymmetry in some popular apologetic responses. A theist may use ordinary moral language when praising God's mercy, justice, or love, but then retreat to radical mystery when confronted with suffering. If the same words change their practical meaning whenever criticism begins, the description risks becoming unfalsifiable.

There is also a difference between saying that our moral knowledge is incomplete and saying that our moral concepts cannot be applied to God at all. The first is modest and plausible. Parents, governments, and medical professionals can have morally sufficient reasons for allowing harms that appear bad in isolation. Human observers often lack relevant information. But this does not make "cruelty" meaningless. It means that deciding whether an act is cruel requires attention to intention, circumstances, consequences, and alternatives.

The second claim is much stronger. If humans cannot understand any intelligible criteria by which God could count as just or loving, then those words lose their contrastive and justificatory force. They could still express the speaker's allegiance, but they would no longer explain why God is better than a morally neutral force, or why one divine action is preferable to another. An entity that causes suffering and an entity that prevents it would be equally compatible with "perfect goodness" only in the thin sense that the term had been insulated from all possible evidence. The point is therefore about the content and use of the claim, not about a proof that no transcendent moral reality is possible.

This point matters in the problem of evil, which remains one of the strongest objections to theism even though it is not generally regarded as a knockdown proof. The argument does not require the simple principle that "a good being prevents every harm it can." Many contemporary philosophers, including critics of theism, reject that principle, since permitting some harms may be justified by goods such as freedom, responsibility, knowledge, or the prevention of greater harms. The strict logical problem defended by Mackie has therefore lost influence, while evidential arguments remain central. They ask whether the scale, distribution, and apparent pointlessness of suffering are better explained by a perfectly good and powerful personal God or by impersonal natural processes. The debate is comparative and probabilistic, not a proof based on a simplistic definition of goodness, and a naturalistic explanation's difficulties would not by themselves establish theism.

The strongest theistic reply

The strongest response does not say that moral language becomes meaningless when applied to God. It says that the language is meaningful but not univocal: "good" does not have exactly the same mode of application to God as it does to human beings.

Many classical theists describe God as the necessary ground or fullness of goodness rather than as one moral agent among others. On this view, God is not judged by an external moral law, as though a higher authority stood above God, and divine goodness is not an arbitrary product of commands. It is identified with a necessarily excellent nature or perfection of being. This claim should not be confused with every version of divine command theory: some natural-law accounts derive creaturely norms from the order of human flourishing, while command-based accounts make obligation depend more directly on divine will. Both kinds of view must still explain why the nature or will in question is appropriately called good and why that goodness excludes cruelty or deception.

Contemporary philosophers have proposed related but distinct versions of this idea. Robert Merrihew Adams defended a modified divine command theory in which the commands of a loving God are central to moral obligation, while God's goodness is not reduced to arbitrary commands. William Alston argued that a divine-command account need not be arbitrary merely because obligations originate in God's commands; its plausibility depends on substantive claims about God's goodness and authority, not on bare power. Richard Swinburne has defended the intelligibility of attributing moral goodness to God while also stressing the limitations of human knowledge concerning divine purposes. None of these positions makes the problem disappear: each must still explain how divine goodness has enough content to guide moral judgment.

The reply can also appeal to analogy. We call a human judge just when the judge gives people what they deserve within a legal and social order. We call God just not because God occupies the same institutional role, but because God's actions are ordered toward the right treatment of persons in a more fundamental sense. Human concepts may be extended rather than abandoned. The fact that God's goodness is greater, more comprehensive, or differently grounded does not by itself make the term empty.

This response has genuine strength. Moral evaluation does not require that God be a human being with human limitations. Nor must a moral standard be external to God in order for divine goodness to be non-arbitrary. A theist can consistently maintain that God is the source of moral reality while also holding that some actions are objectively incompatible with divine love and justice. On this account, saying that God has reasons beyond our understanding is not the same as saying that anything whatsoever could count as good.

Where the reply remains vulnerable

The difficulty is that analogy must preserve enough structure to support moral judgment. If divine goodness means that God acts from a stable, loving, non-arbitrary concern for creatures, then we have some basis for asking whether particular events fit that description. We may be mistaken in a given case, but the question remains intelligible.

Conversely, if "goodness" means only that God's actions express God's nature, while God's nature is described as good merely because it is God's nature, the explanation becomes circular. It no longer tells us why goodness excludes cruelty, deception, or indifference. The theist needs a substantive account of divine goodness, not merely a declaration that whatever God is like is good by definition.

One possible answer is that divine love is a necessary and intelligible feature of perfection: a perfectly good being necessarily wills the genuine flourishing of conscious creatures and never treats them as mere instruments. But this makes moral concepts relevant again. The atheist can then ask whether horrendous and apparently unnecessary suffering is consistent with such love, and whether proposed justifications are plausible. The theist may answer that the suffering is connected to goods or purposes inaccessible to us. That answer may sometimes be reasonable, but it cannot automatically settle the issue.

There is an important difference between epistemic humility and moral exemption. Epistemic humility says, "We should not confidently infer that an event is unjustified merely because we cannot see its justification." Moral exemption says, "No possible event could count against divine goodness because divine goodness has no standards we can understand." The first is a legitimate caution. The second risks emptying the claim of content.

What the argument establishes - and what it does not

The argument poses a serious consistency and meaning challenge for certain forms of theism; it does not establish that every form of theism is incoherent. A theist who relies on ordinary moral praise of God while making divine goodness immune to every intelligible moral criticism risks changing the subject. If "perfectly good" has content, that content must be sufficiently stable for the theist to explain why goodness rules out some possible divine actions and permits others. If the theist instead defends analogy or a distinctive account of divine perfection, the argument becomes a demand for clarification and justification rather than a conclusive refutation.

It does not establish atheism by itself. A successful theistic account of analogical moral language could avoid the charge of emptiness. Nor does the argument show that every instance of suffering disproves God, since moral evaluation depends on facts humans may not know. At most, the argument places a burden on theism to provide a non-circular, substantive account of divine goodness and to explain why the world's suffering is compatible with it.

The critique is therefore strongest when directed at evasive slogans, not at the most careful philosophical theology. It shows that appealing to God's transcendence cannot function as a universal escape from moral reasoning. If divine goodness is meaningful, it remains open to evidence and argument. If it is not, then the phrase may express allegiance, but it cannot do the explanatory or evaluative work that theism normally asks it to do.