The problem of Alvin Plantinga's modal defense of God

Alvin Plantinga’s defense of God remains one of the most sophisticated contemporary defenses of theism. However, Alvin Plantinga’s sophisticated framework creates an unexpected vulnerability: the same logical structure which Plantinga uses to prove God’s existence can also prove the necessary existence of anti-God—an omnipotent, omniscient, and omnimalevolent being—creating a logical inconsistency which renders theism implausible and atheism plausible. 

For Plantinga, believing in God is accepting a proposition that there exists a personal being who is omnipotent, omniscient, and omnibenevolent (Plantinga, 2000, pp. 3, 468). Plantinga shows in his modal argument in The Nature of Necessity, that, if it is possible that God exists, then God must necessarily exist: 

P1. It is possible that a being with maximal greatness (a being that is omniscient, omnipotent, and omnibenevolent in every possible world) exists. P2. “But for any property P, if P is possibly exemplified, then there is a world W and an essence E such that E is exemplified in W, and E entails ‘has P in W’.”

C1. Therefore, there is some possible world in which a being with the property P—maximal greatness—exists.

By definition P1, a being with maximal greatness exists in all possible worlds.

C2. Therefore, a maximally great being (i.e. God) exists necessarily in the actual world (Plantinga, 2010, pp. 214-216).

However, “for Plantinga's argument for a maximally excellent being, if valid, would also work for a maximally evil being” (Hick, 1989, pp. 78) . The following reductio ad absurdum argument has been crafted based on John Hick’s similar argument in his book, An Interpretation of Religion:

P1. God necessarily exists (C2 above).

P2. God is maximally great.

P3. Moral good does not imply metaphysical superiority over moral evil. 

C1. (From P3) Therefore, there is no metaphysical superiority of an omnibenevolent being as compared to an omnimalevolent one.

C2. An anti-God (an omnipotent, omniscient, and omnimalevolent being) possibly exists.

P4. Anti-God is the opposite of God; it must be omnimalevolent, omnipotent, and omniscient in all possible worlds (maximally evil).

C3. Because anti-God possibly exists, and anti-God is maximally evil, then, according to Plantinga’s logic (P1 to C2 above), anti-God necessarily exists. 

C4. Both God and anti-God necessarily exist.

P5. Anti-God cannot coherently exist at the same time as God; anti-God is not logically compatible with God. 

C5. Because premises P1 to P4 lead to a logical inconsistency (C4), at least one premise must be false.

P6. P2 and P4 are definitions.

C6. Either P1 or P3 is false, or both are false

P7. P3 is true.

C7. Therefore, P1 is false; God does not exist.

This argument demonstrates how modal arguments could become problematic when utilized to ‘prove too much’, and in this case, an anti-God. But this argument relies on the assumption of the truth of P3, that moral good does not imply metaphysical superiority, and P5, the logical incoherence of God and anti-God existing simultaneously. 

Premise 3 establishes that moral good does not retain any metaphysical or existential superiority than moral evil. Historically, many theists have erroneously attributed good as a property which is metaphysically perfect. Because God is defined as a perfect being, moral perfection has been accepted as part of God’s being. But perfection in the metaphysical sense does not equal perfection in the moral sense because they are essentially separate qualities—metaphysical perfection does not imply moral perfection. 

Many philosophers have argued in support of this claim. Immanuel Kant distinguishes between moral and physical properties: moral properties are deduced from reason and autonomy, while metaphysical properties are essential attributes to a being. (Kant, 2010, pp. 7-19). Kant’s categorical imperative demonstrates that moral worth comes from acting out of duty, not the consequences or metaphysical properties of the action itself. Thus, to be metaphysically great does not require moral goodness, and there is no metaphysical superiority in being good than being evil. Jean-Paul Sartre also argues that existence precedes essence and that moral value comes from authentic choice and responsibility in his book, Existentialism is a Humanism, with his famous quote, “man is nothing but what he makes of himself,” showing that moral identity rises from free will and not predetermined ontological status (Sartre, 2007, pp. 26). 

Additionally, both thought experiments and real-life examples demonstrate that metaphysical perfection is different from moral perfection. Imagine a perfectly efficient torture device, metaphysically perfect in performing its function, but morally evil. This device, despite being ‘perfect’ in the metaphysical sense, is used for morally evil purposes. In real life, power and knowledge can be utilized for evil purposes; metaphysical knowledge like mathematical equations work equally well for nuclear bombs and medical devices, a surgeon’s skills can both heal a patient and assist in crime, and weapons could be utilized for protection but also murder. 

The philosophical analysis from Kant and Sartre, combined with thought experiments and real life examples, logically leads to the conclusion that moral good does not imply metaphysical superiority, thus validating premise P3.  

Premise 5 assumes that God is incompatible with anti-God. This assumption can be proven in multiple ways. Firstly, two omnipotent beings cannot exist simultaneously, as the omnipotence of at least one being must be undermined. Secondly, if God acts with maximum goodness, and anti-God acts with maximum evilness, then, for any action or state (good or evil), one must affirm the action, and the other must deny the action. And because of their omnipotence, God and anti-God will create a situation where a single action both must occur and must not occur, creating a logical impossibility. Thirdly, the existence of God and anti-God simultaneously undermines the purpose of believing in God. Both the divine command theory and divine sovereignty is undermined, because neither can God be the commander of morality due to anti-God, nor can he assume sovereignty over the universe and creation themselves. There is no reason to believe in a God who cannot define morality or be sovereign. 

The defense of P3 and P5 above shows that the argument is philosophically and logically grounded. Several potential counterarguments, however, still require addressing, including the privation theory of good, the claim that belief in anti-God cannot be properly basic, and the argument that anti-God and maximal evilness is incoherent. 

Critics might argue that, because evil is the privation of good, evil does not have as much metaphysical reality as good, as it simply describes the lack of a property, and not an opposite property (Augustine, n.d., pp. 44). Nevertheless, evil is not necessarily the lack of good. In philosophy, good is a value statement, meaning that a good action brings about more value (Schroeder, 2021). A being which never commits a good action does not necessarily commit evil. Consider a disabled person having trouble getting up stairs. Helping him would be ‘good’, not helping him would be ‘neutral’, and purposely kicking him downstairs would be ‘evil’. In this case, a lack of ‘good’ (not helping) cannot be defined as evil, since this lack of action is neither causing an increase or decrease in overall value. Evil, in contrast, is an action which produces negative value, such as purposely kicking the disabled person downstairs. Moreover, as William Rowe points out, the privation theory does not account for natural and pointless evils, such as lightning striking a tree, causing a forest fire and pointless suffering of animals, creating negative consequences instead of simply lacking good consequences (Rowe, 1979, pp. 338). J.L Mackie provides another critique of the privation theory by pointing out that the privation theory inconsistently denies evil’s reality while simultaneously relying on it to explain why God permits suffering. This contradiction undermines the theory’s ability to genuinely address the moral problem of evil (Mackie, 1955, pp. 201–202). Thus, the privation theory of evil fails due to its failure to explain the nature of evil, and its incoherent use even in a theist system. 

Plantinga might argue that belief in God is properly basic (giving it possibility), while belief in anti-God is not. A properly basic belief is a rationally justified belief without being inferred from other beliefs or supported by arguments and evidence, such as the belief that other people have minds (Plantinga, 2000, pp. 287). Plantinga states that a properly basic belief requires a warrant, and “a belief has warrant for a person S only if that belief is produced in S by cognitive faculties functioning properly (subject to no dysfunction) in a cognitive environment that is appropriate for S’s kind of cognitive faculties, according to a design plan that is successfully aimed at truth” (Plantinga, 2000, pp. 180).  Plantinga argues, under the A/C (Aquinas/Calvin) model, that God created in us a cognitive faculty—the sensus divinitatis—to sense God. We use this cognitive function (as it is functioning properly), under an appropriate cognitive environment and God’s design plan aimed at truth, to form our properly basic belief in God. 

But hypothetically, anti-God might have created a similar cognitive capacity—the sensus malignitatis—to sense anti-God; if we use this function, as it functions properly, under an appropriate cognitive environment and anti-God’s design plan aimed at truth in the belief of an anti-God, then belief in anti-God could similarly be properly basic. One might claim that only a good being—God—can design faculties aimed at truth, while an evil being—anti-God—aims to deceive. Yet this makes the assumption that truth in the metaphysical sense is intrinsically good. Plantinga’s condition of a design plan aimed at truth, refers to cognitive reliability, not morality. Truths could serve evil just as well as it serves good. For example, an anti-God might want the entire world to believe in him to cause more suffering, contrary to God wanting the world to believe him to bring about good. Therefore, the properly basic belief of anti-God is warranted by Plantinga’s model, and the logical incoherence still exists. 

A theist might also deny the incoherence of an anti-God, due to it merging metaphysical greatness with moral evilness; in other words maximum greatness includes moral goodness, while omnimalevolence is an imperfection. However, this response is question-begging and assumes that metaphysical greatness includes moral greatness—the very thing being argued. Describing omnimalevolence as an imperfection, similarly, cannot hold true, since metaphysical perfection does not imply moral perfection. Furthermore, there is nothing ‘incoherent’ in the concept of anti-God itself. The premise of Plantinga’s modal argument relies on the possibility of God’s existence (P3). This possibility does not rely on empirical evidence, but rather the logical coherence of the belief in God. Because they rely on the same logical structure of properly basic beliefs (shown above), if we can coherently believe in a maximally great being, we can also coherently believe in a maximally evil being. To deny the coherence of anti-God is to deny the coherence of God as well. 

A theist could go further to question the coherence of maximal evilness, claiming that it is unclear what such a being with this property would value and whether it would self-destruct. Because anti-God is the polar opposite of God, and, as shown earlier, evil is not ‘lacking’ but a ‘negative’, anti-God can coherently exist, valuing the opposite of what God values. For example, if God values well-being, then anti-God values suffering. Anti-God would also not self-destruct as it needs to exist in order to cause more evil, just like how God needs to exist in order to achieve good. Additionally, because maximum greatness has no metaphysical superiority over maximum evilness, denying the coherence of maximum evilness denies the coherence of maximum greatness as well.

Ultimately, this reductio argument reveals a fatal flaw in Plantinga’s argument for God’s necessary existence. Plantinga’s sophisticated modal reasoning, despite being one of the best, undermines itself in proving the logical contradiction that both God and anti-God necessarily exist. When even the best modal arguments fail, atheism remains the more rational position. Rather than requiring faith without evidence, atheism represents the logical conclusions of rigorous philosophical analysis. Therefore, atheism is not implausible, but rather a reasonable conclusion to the logical failures of the most sophisticated theist defenses.

References

Augustine. (n.d.). Confessions and Enchiridion (A. C. Outler, Trans.). Christian Classics Ethereal Library. https://www.ccel.org/ccel/a/augustine/confessions/cache/confessions.pdf

Draper, P. (2022, March 22). Atheism and Agnosticism. Stanford Encyclopedia of Philosophy. Retrieved June 30, 2025, from https://plato.stanford.edu/entries/atheism-agnosticism/#DefiAgno

Hick, J. (1989). An Interpretation of Religion : Human Responses to the Transcendent. Palgrave Macmillan UK.

Kant, I. (2010). Groundwork of the metaphysics of morals (M. J. Gregor, Trans.). Cambridge University Press.

Mackie, J. L. (1955). Evil and Omnipotence. Mind, 64(254), 200-212. https://courses.edx.org/c4x/MITx/24.00x/asset/MackieonEvil.pdf

Plantinga, A. (2000). Warranted Christian belief. Oxford University Press.

Plantinga, A. (2010). The nature of necessity. Clarendon Press.

Rowe, W. L. (1979). The Problem of Evil and Some Varieties of Atheism. American Philosophical Quarterly, 16(4), 335?341. http://www.jstor.org/stable/20009775

Sartre, J.-P. (2007). Existentialism is a humanism : including, A commentary on the stranger (C. Macomber & A. Elkai?m-Sartre, Trans.). Yale University Press.

Schroeder, M. (2021, March 4). Value Theory. Stanford Encyclopedia of Philosophy. Retrieved June 30, 2025, from https://plato.stanford.edu/entries/value-theory/