Ontological Argument

In 1078, the Benedictine monk Anselm of Canterbury proposed, in his Proslogion, an argument he believed could prove God's existence from a single premise, without appeal to the created world at all. He defined God as 'that than which nothing greater can be conceived' (id quo maius cogitari nequit) and argued that even someone who denies God's existence — the 'fool' of Psalm 14 — must understand this concept when it is presented to them, and so this being exists at least in the understanding.

Anselm's crucial move was to argue that if this greatest conceivable being existed only in the understanding and not in reality, we could conceive of a still greater being — one that had all the same properties but also existed in reality. But this would contradict the very definition of the being as 'that than which nothing greater can be conceived'. Therefore, Anselm concluded, the greatest conceivable being must exist in reality as well as in the mind.

Scholars continue to debate whether Anselm offered one argument or two distinct arguments in the Proslogion — one from degrees of greatness (Chapter 2) and one from necessary existence (Chapter 3), the latter emphasized by Norman Malcolm in the twentieth century. The distinction matters because the necessary-existence version more closely anticipates the modal arguments developed by Plantinga eight centuries later.

René Descartes, in the Fifth Meditation, offered a related but distinct version of the ontological argument. Rather than reasoning from comparative greatness, Descartes argued that existence is simply one of the perfections that belongs to the essence of a supremely perfect being, just as having three angles summing to 180 degrees belongs to the essence of a triangle. To conceive of God — a being possessing all perfections — while denying that God exists would be as incoherent as conceiving of a triangle whose angles do not sum to two right angles.

Descartes was aware this move seemed suspicious, and he defended it by insisting existence is not being arbitrarily added to the concept of God in the way one might imagine wings on a horse; rather, it is a necessary component of the very idea of a supremely perfect being, discoverable by clear and distinct rational reflection alone.

This version drew immediate criticism from Descartes's contemporaries, most famously in the objections collected alongside the Meditations. Pierre Gassendi and others pressed the worry that Descartes had simply built existence into his definition by fiat, and that no amount of careful reasoning about a concept can tell us whether anything in reality corresponds to it — a worry that Kant would later develop into a full systematic critique.

Anselm's argument received a challenge within his own lifetime from a fellow monk, Gaunilo of Marmoutier, who wrote a 'Reply on Behalf of the Fool'. Gaunilo argued that if Anselm's reasoning were valid, one could use exactly the same logical form to prove the existence of the greatest conceivable island — an island more excellent than any other conceivable island, which therefore, by parallel reasoning, must exist in reality, since an island existing only in the mind would not be the greatest conceivable island.

Since it is absurd to suppose such reasoning could actually conjure an island into existence merely from its definition, Gaunilo concluded something must be wrong with Anselm's argument form itself, regardless of the specific content about God.

Anselm's classical response, echoed by later defenders, is that islands (and similar concrete objects) are composed of properties without intrinsic upper limits — one can always conceive of an island with more trees, better soil, or a more pleasant climate — whereas God's greatness is composed of perfections like knowledge, power, and goodness that do admit of an intrinsic maximum. If this disanalogy holds, the Perfect Island parody targets a disanalogous case and leaves Anselm's core argument about a uniquely maximal being untouched, though critics continue to construct more sophisticated parodies attempting to avoid this response.

Immanuel Kant's Critique of Pure Reason (1781) delivered what many regard as the most devastating classical objection to the ontological argument, targeting both Anselm's and Descartes's versions. Kant argued that 'existence' (or 'being') is not a real predicate — that is, it is not a property that adds content to the concept of a thing. To say 'God exists' is not to attribute a further characteristic to the concept of God, but merely to assert that the concept of God has an instance in reality.

Kant's famous illustration involves imagining a hundred real thalers (a coin) versus a hundred merely possible thalers: the concept of a hundred thalers is exactly the same whether or not any actually exist in one's pocket; what differs is not the concept's content but whether reality contains an instance of it. If existence added nothing conceptually, then no amount of analyzing the concept of a supremely perfect or maximally great being could ever tell us whether it is instantiated in reality — that is always, for Kant, a further, synthetic question that concepts alone cannot answer.

This objection, though enormously influential, has not gone unchallenged. Contemporary logicians using predicate logic with quantifiers (following Frege and Russell) have offered more precise ways of stating Kant's point — existence functions as a second-order property of concepts (whether they are instantiated) rather than a first-order property of objects. This technical vindication of Kant's intuition is part of why later ontological arguments, especially Plantinga's, deliberately avoid treating existence as an ordinary predicate at all.

In the 1970s, Alvin Plantinga reformulated the ontological argument using the tools of modal logic and possible-world semantics, explicitly designed to sidestep Kant's objection. Plantinga defines a being as 'maximally excellent' if it has omniscience, omnipotence, and moral perfection, and 'maximally great' if it has maximal excellence in every possible world. The argument's key premise is simply that it is possible that a maximally great being exists — i.e., there is at least one possible world in which such a being exists.

Given the modal logic system S5 — the system most philosophers accept for reasoning about metaphysical necessity — if a maximally great being possibly exists, then since maximal greatness entails existing (with maximal excellence) in every possible world, it follows that such a being exists necessarily, and hence exists in the actual world too. The argument is formally valid; the entire weight of the debate rests on whether the possibility premise is true.

Plantinga himself was notably modest about what his argument achieves, writing that it does not prove theism true in a way that would compel every rational person's assent, since a rational person could instead accept the parallel premise that a maximally great being's existence is impossible. His conclusion was that the argument shows theistic belief to be rationally acceptable, not that it settles the God question with mathematical certainty.

A powerful line of criticism against Plantinga's version notes that the argument is perfectly symmetrical: just as a theist might find it plausible that a maximally great being possibly exists, an atheist could with equal apparent plausibility assert that a maximally great being's non-existence is possible (i.e., that no such being exists in any possible world). Since S5 entails that both cannot be coherently true together, the argument alone gives us no non-question-begging way to prefer one possibility premise over the other.

This symmetry has led many philosophers, including some sympathetic to theism, to treat the modal ontological argument as dialectically inert on its own — valid but not independently persuasive — better understood as clarifying what would need to be true (the coherent possibility of maximal greatness) for theism to be necessarily true, rather than as a freestanding proof.

A further worry, raised by philosophers skeptical of modal intuitions generally, is that our sense of what is 'possible' regarding exotic, maximally great beings may not be a reliable guide to genuine metaphysical possibility, in the way our modal intuitions about ordinary objects (like whether a given table could have been made of different wood) plausibly are.

Unlike cosmological, teleological, or moral arguments, the ontological argument requires no empirical premises about the created world; its entire force (if any) derives from conceptual and logical analysis. This makes it simultaneously the most elegant and the most contested argument in natural theology — elegant because, if sound, it would establish God's existence with the certainty of mathematics, and contested because every step invites deep and unresolved metaphysical questions about concepts, possibility, and existence.

Most contemporary philosophers of religion, including many theists, regard the ontological argument as, at best, one component of a broader cumulative case for theism rather than a standalone proof. Its lasting philosophical value lies less in settling the God question decisively and more in sharpening our understanding of modality, existence, and the logical structure of perfection — questions that remain central to metaphysics quite apart from their theological application.