The Ontological Argument (Anselm, Descartes, Kant, Plantinga)

An argument is a line of reasoning that gives reasons to accept a conclusion. This one is called ontological because it examines what God must be to be truly God. It starts from an idea and tries to see what follows from it. This starting point is exactly what sets it apart from other arguments for God.

Usually, we begin with something we encounter in the world. We observe what exists, what changes, or the difference between good and evil. The moral argument starts from this last question. Then we look for an explanation. The ontological argument starts elsewhere: it asks what we understand when we speak of God.

Its starting point is therefore a concept. A concept is an idea that helps us understand what we are talking about. The concept of a mountain is not the same as a real mountain. Likewise, understanding a description of God does not yet mean we have shown that God exists.

Anselm nevertheless proposes a special case. He asks us to think of a being whose greatness could not be surpassed. Here, greatness does not mean size or weight. It means what makes a being more fully perfect: its knowledge, power or goodness.

We can always think of a taller mountain or a more knowledgeable scholar. So we are not simply talking about the best being encountered so far. We are talking about a being to which no improvement could be added. A perfection here means a quality that is not weakened by any flaw.

The question then becomes very precise: could an absolutely perfect being fail to exist? For Anselm, this would mean admitting that it had a crucial lack. A real being would be greater than an idea present only in our minds. But this move from an idea to reality is exactly what his critics challenge.

To follow the discussion, we need two words. A contingent being exists but could have failed to exist. An island belongs to this category: we can understand what it is without having to say that it exists. A necessary being, by contrast, could not fail to exist. The argument from contingency starts from these things that could have failed to exist.

So the word necessary does not have its usual meaning here. An object may be necessary for my work without having to exist absolutely. In this debate, necessary means that the absence of this being would be impossible. This is a much stronger claim than saying it is useful or exists at present.

Finally, saying that a being is possible means here that it could really exist. Simply being able to say its description is not enough. A contradiction means affirming and denying the same thing in the same sense. The discussion will therefore ask whether the idea of God contains such a contradiction.

Saint Anselm of Canterbury is the first author of this argument. He sets it out in 1077-1078 in the Proslogion, a work written as a prayer. He seeks to understand better the God in whom he already believes. His intention is not to set a trap for those who do not believe.

“Lord, you who give faith understanding, grant me to understand, as much as you know to be good for me, that you are as we believe, and that you are what we believe.”

Saint Anselm, Proslogion, chapter II.^[1]

This approach combines trust and reflection. Anselm does not think that believing removes the need to understand. He seeks a line of reasoning that can stand on its own. To examine it, we can therefore follow its steps without assuming in advance that we accept its conclusion.

Chapter II begins with a familiar distinction. Something can be present in our thoughts without being present in reality. Understanding the description of an island does not make it appear in the ocean. On this point, Anselm and his opponents can agree completely.

Anselm then argues that God's case is unique. Suppose that the absolutely great being exists only in our thoughts. We could then think of this same being as also existing in reality. For Anselm, this second case would be greater than the first.

We would therefore have thought of something greater than the being assumed to be absolutely great. Our starting point would be contradicted. Anselm concludes that this being cannot remain an idea that exists only in thought. It must also exist in reality.

It is easy to see why this conclusion is surprising. Usually, adding qualities to a description never guarantees that what it describes exists. Yet Anselm claims that the description of God is an exception to this rule.

Chapter III explores this question further. It compares a being that could have failed to exist with a being whose absence would be impossible. According to Anselm, the second has greater greatness. The absolutely great being must therefore exist necessarily, rather than simply happen to exist.

This step gives the reasoning its deepest significance. A God who could have failed to exist would not be God in the intended sense. He would be an exceptional reality, but his existence would still be fragile. Anselm instead seeks the one whose existence cannot depend on a fortunate circumstance.

A difficulty arises at once. If God's absence cannot be thought of, how can anyone deny God? Chapter IV then distinguishes between saying an expression and fully understanding what it refers to. We can use a word without having grasped everything its meaning involves.

For Anselm, thinking of a God who might not exist therefore means missing what he intends to describe. An idea is coherent if the statements it contains do not contradict one another. This does not mean that every unbeliever is unable to think. It identifies the precise disagreement: is the idea of a necessary God coherent, and does it allow us to conclude that he exists?

It would therefore be misleading to sum up his reasoning as the power of imagination. Imagining a winged horse proves nothing about its existence. Anselm is talking about an absolutely perfect being whose necessity would be part of what it is. The whole debate concerns whether this particular step is sound.

Anselm's argument has been developed in several ways. Saint Bonaventure and John Duns Scotus are two Catholic thinkers from the Middle Ages. The first takes up the idea that the absolutely perfect being cannot remain a mere thought. The second places more emphasis on the possibility of a first being that brings other beings into existence. These developments show that the Catholic tradition has not followed just one way of reasoning.

René Descartes, a philosopher, proposes a new formulation in his Meditations on First Philosophy, in 1641. These meditations examine, among other things, what we can know about God. Descartes starts from the idea of a supremely perfect being. For him, such a being possesses every perfection.

His reasoning can be stated simply. God cannot lack any perfection. Now Descartes considers existence to be a perfection. He concludes that we cannot think correctly about God while taking existence away from him.

To clarify his idea, Descartes turns to mathematics. In the geometry he uses for his comparison, the angles of a triangle add up to two right angles. This relationship belongs to the triangle as soon as we understand what it is. Descartes argues that existence belongs to God in this way.

The comparison helps us understand his intention, but it also reveals a difficulty. Describing the features of a triangle does not show that a triangle really exists in front of us. Why would the description of God allow us to go further? Descartes replies that God's absolute perfection is precisely what sets him apart.

This formulation nevertheless rests on a much-debated point: is existing really a quality comparable to knowing or being good? If it is not, his reasoning needs to be re-examined. This question will be at the heart of Kant's criticism.

Gottfried Wilhelm Leibniz, another philosopher, identifies a further difficulty. Before drawing a conclusion, we must show that all perfections can belong to the same being. A description may seem clear even though its different parts cannot fit together. The idea of absolute perfection must therefore be examined, rather than simply put forward.

“It must be proved with the rigor of geometrical demonstrations that such a [supremely perfect] being is possible.”

Gottfried Wilhelm Leibniz, Quod Ens Perfectissimum Existit, November 1676.^[2]

Leibniz tries to answer by examining perfections. He calls qualities simple when they cannot be broken down into more basic qualities. He calls them positive because they contain no negation.

According to him, for them to conflict, one quality would have to affirm what another denies. This is why he argues that they can go together. This answer remains debated, but his question can no longer be avoided.

Not every version of the argument, however, leads to the Christian understanding of God. The philosopher Spinoza identifies God with Nature. This position is called pantheism. It means identifying God with the whole of nature, rather than distinguishing the Creator from his creation.

The philosopher Hegel also defends the importance of the argument against Kant. But his way of connecting God with the whole of being and history also falls outside the traditional Christian framework. For the Catholic faith, God is not the universe itself. Defending the argument therefore also means making clear which God we are talking about.

The first objection comes from Gaunilo of Marmoutiers, a monk who lived at the same time as Anselm. He proposes describing an island superior to all others. It would be wonderfully rich, beautiful and fertile. We understand this description, but no one is forced to conclude that the island really exists.

The objection targets exactly the move from thought to reality. If a definition is enough to make God exist, why would it not also make this island exist? Repeating that God is perfect does not answer this difficulty. We must explain why the two cases differ.

Anselm replies that an island remains something limited, located somewhere and capable of improvement. Its beauty or fertility do not make its existence necessary. God, in the proposed reasoning, has no limits at all. The word infinite here means this absence of limits, rather than an enormous size.

So the answer is not to imagine an even more wonderful island. It is to say that no island can meet the definition of the necessary being. Gaunilo's comparison warns against faulty reasoning. According to Anselm's defenders, however, it is not enough to refute the special case he examines.

Saint Thomas Aquinas raises a different objection. This great Catholic thinker does not treat God as something that might exist by chance. He does indeed hold that God cannot be separated from his existence. But he asks whether our minds know God well enough to start directly from what he is.

His answer is no. Understanding a definition given in words does not mean fully grasping the divine reality. For Thomas, we must begin with what is more accessible to us. We can examine the world and work back to God from his effects. His effects are what he produces.

The cosmological argument follows this approach, starting from the world. This objection retains its full force within a Catholic reading. Anselm's defenders can nevertheless give his argument a different role. Even if it is not our first proof, it clarifies what we mean by God. It helps us understand why God would not simply be one very powerful being among others.

David Hume, a philosopher critical of proofs for God, goes further. According to him, anything we conceive of as existing can also be conceived of as not existing. We therefore cannot prove an existence simply by reflecting on an idea. This objection rejects precisely the exception claimed for God.

The reply highlights a difference between forming a sentence and establishing a real possibility. Being able to speak of a world without God does not yet show that such a world is possible. If God is necessary, his absence is impossible. But this necessity must then still be justified, which brings us back to the heart of the debate.

Immanuel Kant offers the most famous criticism. He asks whether existence adds anything to the description of an object. According to him, the answer is no. Saying that an object exists affirms that it is real, without adding another quality to what we had described.

His example concerns a sum of one hundred thalers, a form of money. Whether it is merely thought of or actually owned, the sum described remains the same. Yet the difference is considerable for the person who owns the money. It concerns the real presence of that money, rather than a quality added to its description.

Kant thus touches on a weak point in Descartes' version. If existence is not an additional perfection, we cannot simply add it to a list. Defenders of modern versions agree to shift the discussion. They speak of the necessity of existing, rather than existence treated as an ordinary quality.

This difference needs to be understood carefully. Saying that something exists and saying that it could not fail to exist are two distinct claims. The second concerns its way of existing. According to these defenders, Kant's criticism is not enough to dismiss this new reasoning.

Gottlob Frege and Bertrand Russell, two thinkers who study the rules of reasoning, make the objection more precise. Saying that horses exist means that the concept of a horse corresponds to real beings. It does not add a quality to each horse. A description, however precise, therefore still does not guarantee that an object corresponds to it.

Modern replies do not deny this general difficulty. Formulations based on what is possible and necessary allow the necessity of existing to be examined rigorously. They do so without confusing description and reality. The debate then concerns their starting claims. Modern logic alone is therefore not enough to declare the argument refuted.

In the last century, Norman Malcolm and Charles Hartshorne put necessity back at the centre of the discussion. In 1960, Malcolm stresses the difference between chapters II and III of the Proslogion. The first concerns existence, the second the impossibility of not existing. According to him, Kant's criticism therefore does not affect them in the same way. For Malcolm and Hartshorne, such a being therefore either exists necessarily or cannot exist.

Alvin Plantinga then develops a particularly debated formulation in The Nature of Necessity, in 1974. His reasoning uses modal logic. This way of examining arguments distinguishes what is possible, impossible or necessary. It helps make the steps we have just encountered more precise.

We also need to understand the expression possible world. It means a complete description of reality that could be true without contradiction. It does not necessarily mean a universe existing somewhere else. It is a way to compare different ways reality could be.

Think of Gaunilo's island. One possible description of reality may contain this island, while another may not. The island would therefore not be necessary. A necessary being, by contrast, would be absent from no genuinely possible description of reality.

Plantinga then distinguishes two expressions. Maximal excellence means perfect knowledge, power and goodness in one possible world. Maximal greatness means this same excellence present in all possible worlds. The second expression therefore includes necessary existence.

We can then follow the reasoning in a few steps:

1. It is possible that a being possessing maximal greatness exists.
2. This greatness includes existence in all possible worlds.
3. Such a being therefore could not exist only in some worlds.
4. If it is really possible, it exists necessarily.
5. It then also exists in the actual world, which is one of the possible situations.

This transition uses a system of reasoning called S5. Here, this name refers to rules according to which whatever is possibly necessary is necessary. In this system, a truth that is necessary in one possible world also holds in all the others.

This does not mean turning any possibility into reality. A possible island does not become real through this reasoning. What makes the difference is the necessary existence included in maximal greatness. Without it, the step presented by Plantinga does not work.

The difficulty therefore lies at the beginning. A premise is a starting claim on which a line of reasoning rests. Here, the decisive premise states that a being possessing maximal greatness is really possible. It seems modest, but it already contains a very strong claim about what can exist.

Plantinga himself acknowledges that this claim can be rejected. He does not present his argument as a proof that would force every opponent to give in. Instead, he defends the reasonableness of his starting point and his conclusion. Here, he is speaking about several formulations of this argument.

“Perhaps one cannot say that they prove or establish their conclusion. But since it is rational to accept their central premise, they show that it is rational to accept that conclusion.”

Alvin Plantinga, The Nature of Necessity, p. 221.^[3]

Kurt Gödel, a specialist in the rules of reasoning, proposes another version of the argument. His work is published after his death, in 1995. He seeks to write every step with a precision comparable to that of a mathematical proof. This precision makes it possible to check the steps of the reasoning in detail.

His proof begins with axioms. These are starting claims accepted in order to build the proof. They must be distinguished from the results obtained later. A line of reasoning can follow its axioms correctly without their truth having been established.

Gödel uses, in particular, the expression positive property. In his reasoning, it means a quality treated as a perfection rather than a limitation. This meaning is fixed by the rules he adopts. So the word positive does not simply mean pleasant, useful or reassuring.

Among his starting points, Gödel accepts that necessary existence is a positive property. He also accepts that possessing all positive properties by nature is itself a positive property. From his full set of rules, he concludes that a divine being exists necessarily. The focus is on the precise connections between these claims. These two claims are among the starting points that must be defended philosophically.

In 2013, Christoph Benzmüller and Bruno Woltzenlogel-Paleo use a computer to check a version of this proof. This check confirms that the conclusion follows from the chosen starting claims. It does not show that each of these claims correctly describes reality. The computer checks the steps, rather than the philosophical truth of the whole starting point.

This distinction matters a great deal. If we accept certain rules, a conclusion may become unavoidable within that framework. But we can still ask why we should accept those rules. Gödel's work thus makes the debate more precise: it makes clearer what has been proved and what still needs to be justified.

The strongest contemporary objection returns to Plantinga's first sentence. Why accept that the existence of an absolutely great being is possible? An opponent can propose the opposite starting point. They can argue that it is possible that no being of this kind exists.

This objection is called the symmetry problem. It sets against each other two starting points that seem to mirror one another. The first affirms the possibility of God. The second affirms the possibility of his absence.

In the S5 system, these claims cannot both be true. We are still talking about the same being, defined by maximal greatness. Such a being either exists necessarily or cannot exist. If its absence is possible, it therefore cannot exist. The reverse argument thus leads to a conclusion opposite to Plantinga's.

Graham Oppy, a philosopher critical of the argument, stresses this difficulty. The reasoning will not convince someone who rejects its first claim unless we have further reasons to accept it. The correctness of the later steps is not enough. We must defend what we assumed at the beginning.

Its defenders reply that the two starting points are not equally justified. Within this framework, allowing the absence of the necessary being to be possible amounts to saying that it cannot exist. They therefore ask what precise conflict would prevent its qualities from going together. This reply does not remove their duty to justify their own claim of possibility.

Another objection asks whether the qualities attributed to God can really go together. The philosopher C. D. Broad raises this problem in particular. Is perfect knowledge compatible with our freedom? Is perfect power compatible with the inability to do evil?

Traditional replies distinguish between knowing a choice and removing that choice. Knowing what someone chooses does not, in itself, mean forcing them to choose. For the argument's defenders, perfect knowledge therefore does not necessarily take away the freedom of the person acting.

As for evil, the power being examined is that of a perfectly good being. Attributing an evil act to him would contradict this moral perfection. His defenders therefore do not consider this inability a lack of power. They distinguish between doing what is possible and bringing about a contradiction. Evil is not contradictory in itself: the problem lies in attributing it to a perfectly good being.

The question of time calls for a further distinction. Can God know events without changing as they happen? The traditional reply explains that he knows them without himself being carried along by the succession of moments. This explanation needs to be defended, but it offers a way to understand these qualities together.

The philosopher J. N. Findlay takes the objection further. A being worthy of worship should, according to him, exist necessarily. But he considers this necessary existence incomprehensible according to the philosophical standards he adopts. He concludes that such a God cannot exist.

Malcolm and Hartshorne refuse to treat necessary existence as an empty expression. They distinguish between what could fail to exist and what could not. This distinction leads to the alternatives we have already encountered: for God, either necessary existence or impossible existence.

Alexander Pruss, a Catholic philosopher, brings together several reasons to take the possibility of God seriously. These reasons seek to show why this possibility deserves to be accepted.

The natural desire for the infinite is a first reason. It expresses a longing for an unlimited good that would lack nothing. Its defenders see this as a sign that this good can be real. The reasoning does not say that everything we desire exists. It suggests that what this natural desire seeks is possible, without establishing on its own that it exists.

Converging religious experiences provide another reason. These are experiences that point towards the same divine reality. Their convergence is offered as a reason to take seriously the possibility of a real God. They are neither produced nor verified by the rules of the S5 system alone.

Examining divine qualities provides a further reason to think they are compatible. The models proposed to understand God bring these qualities together in a single explanation. Here, a model is an explanation whose different claims can fit together. The distinctions concerning knowledge, freedom, goodness and time play this role. They prevent us from immediately concluding that these qualities contradict one another, and their coherence must be examined.

Taken separately, these reasons are not conclusive. They supplement the analysis of the concept with reasons to take it seriously.

We must nevertheless keep the crucial distinction in mind. Not having found a contradiction does not automatically prove that something is really possible. Its defenders therefore seek positive reasons, rather than merely the absence of a refutation. This is where the ontological argument joins the other paths to God.

A cumulative argument brings together several independent reasons that support the same conclusion. The cosmological argument examines, in particular, why the world needs an explanation. The argument from contingency asks why things exist that could have failed to exist. The moral argument examines what grounds the real difference between good and evil.

The fine-tuning of the universe, for its part, examines the order of the world's physical conditions. These approaches begin with different questions. If they make a necessary, intelligent and good being plausible, they strengthen the reason to examine the possibility of God. They do not prove this possibility simply by being numerous.

This is how the ontological argument finds its strongest place. The other paths give reasons to believe that we are talking about a reality, rather than just a definition. It asks what this reality must be if it is truly God. The arguments then support one another without becoming interchangeable.

A further reason suggests that there would be only one God. If a single necessary being is enough to explain the world, why add others? This economy of explanation favours the idea of one God, without proving it on its own.

The Catholic Church does not require the ontological argument as a proof of God. Nor has she condemned it. Some Catholic thinkers have defended it, while others have judged it insufficient to establish God's existence. This disagreement concerns a method of reasoning, rather than whether people are allowed to believe in God.

The path emphasised by Catholic teaching starts from the created world. The First Vatican Council affirms this in Dei Filius, on 24 April 1870. A council is an assembly of bishops gathered to address, among other things, the Church's faith. Here, its teaching concerns what human reason can know.

“This same Holy Mother Church holds and teaches that God, the beginning and end of all things, can be known with certainty by the natural light of human reason from the created things.”

First Vatican Council, Dei Filius, chapter 2.^[4]

The natural light of reason here means our human ability to think and understand. A revelation means what God makes known to us about himself through his deeds and words. This statement therefore does not require us to believe a revelation before beginning any inquiry into God. It recognises the real reach of our understanding. The world can lead us to its origin and end.

The Catechism of the Catholic Church takes up this approach. This book presents the teaching of the Catholic faith. It mentions, in particular, change, the order and beauty of the world, and our awareness of good. The ontological argument can complement these reflections without taking their place.

We must then clarify what it would contribute if it succeeded. It would not merely conclude that something invisible exists. It would point to a necessary and absolutely perfect being. In the analysis offered by its defenders, this conclusion calls for several clarifications about how to understand God.

Transcendence means that God is beyond the world and is not an element that depends on it. An island belongs to the universe, and its presence depends on particular conditions. God would not be one more element in this whole. He would be the being on whom other realities depend for their existence.

Divine simplicity means that God is not made by assembling several parts. A composite whole requires its elements to be brought together. God would then depend on whatever brings his parts together. He would no longer be the being on whom everything depends. So the word simple means neither lacking in richness nor easy to understand.

Eternity here means that God does not depend on the passing of time. It does not simply mean an extraordinarily long duration. His knowledge and goodness are not qualities added later, either. In Plantinga's formulation, they belong to the very perfection the reasoning examines.

Its defenders also argue that God would be unique. According to this analysis, two absolutely perfect beings could differ only if one had a limit that the other did not. The one with this limit would no longer be absolutely perfect. The defenders conclude that there cannot be two beings of this kind.

These developments nevertheless require us to examine each quality and the connections between them. Not all historical formulations of the argument present them in the same way. Saying that God is necessary does not remove the need to understand what we go on to claim.

This inquiry belongs to natural theology. This expression means what our reason can seek to know about God without starting from a revelation received in faith. The argument therefore does not directly give us the whole content of Christianity. On its own, it does not demonstrate what God has done in history.

This argument is not enough to establish Christian teaching about God's inner life. To examine the Christian faith, we must also look at the deeds and words through which God has made himself known. The resurrection of Jesus belongs to this historical inquiry.

The lesson of the argument nevertheless remains important, even for someone who prefers the paths of Thomas Aquinas. God is not to be understood as an exceptionally great creature. Our minds can seek to know him without claiming to contain him in an image. Greater understanding must therefore go hand in hand with an awareness of our limits.

• The ontological argument starts from the idea of God, rather than an observation of the world.
• Anselm asks whether the absolutely perfect being could exist only in our thoughts.
• A necessary being cannot fail to exist. It differs from something that simply exists today.
• Gaunilo, Thomas Aquinas, Hume and Kant raise precise difficulties, which different formulations address.
• Plantinga shifts the debate towards the real possibility of a necessary and absolutely perfect being.
• Checking the steps of an argument is not enough to establish the truth of its starting claims.
• In Catholic thought, this argument complements the other paths without replacing the examination of the world and history.

The strength of this argument lies in a question that is hard to avoid. If God is truly God, can he depend on a fortunate circumstance to exist? Anselm replies that this would still mean thinking of a creature, however extraordinarily powerful. God must be the one whose existence cannot fail.

The difficulty remains showing that such a being is really possible. Acknowledging it does not remove all value from the reasoning. It shows where to look, what questions to ask and why the other arguments matter. The Catholic faith can welcome this inquiry without turning it into a formula that settles everything.

At the burning bush, God reveals to Moses a name that this reflection helps us ponder:

“I am who am. (...) This is what you shall say to the children of Israel: I-Am has sent me to you.”

Exodus 3, 14.^[5]

This biblical text is not an extra step added to force the proof to work. It opens up a depth that the reasoning seeks to illuminate. God is not simply greater than the realities around us. He is the one who cannot fail to exist, and towards whom our minds can move with confidence.