Anselm's Proslogion: The Ontological Argument

A painter can plan a picture before painting it. Understanding the plan does not make the finished picture exist. Anselm uses that distinction in chapter II of the Proslogion, just before arguing that one particular object of thought cannot exist only in thought. What could make that case different?

His proposed answer concerns a being beyond which no greater can be conceived. He argues that denying its real existence conflicts with understanding what the description means. The resulting argument, commonly called ontological, asks whether reflection on a concept can establish the existence of what it describes. Evaluating it requires keeping the initial description, the reasoning, and the conclusion apart.

In the preface, Anselm explains that his earlier Monologion connected many arguments. He wanted to find a single argument that would establish God's existence and support an account of the divine nature. His search for a proof was deliberate. Presenting the Proslogion as a prayer that accidentally became a proof would misrepresent his stated intention.

He describes the work as written in the person of someone seeking to understand what he believes. Its earlier title was Faith Seeking Understanding. The speaker already believes in God and asks for understanding of that belief. This starting point explains the devotional setting without deciding whether the reasoning can convince a reader who does not share the belief.

Chapter I asks the reader to withdraw from distracting occupations and seek God. It closes by placing belief before understanding. Chapter II then addresses God while introducing a challenge involving the scriptural fool who denies God's existence. Prayer and argument belong to the same project; they are not successive genres joined by an accidental interruption.

Two questions should remain distinct. What does the reasoning do for a believer seeking understanding? What reasons does it offer someone who disputes its conclusion? An argument can have a religious purpose while presenting premises whose force remains open to examination. Neither the speaker's faith nor a reader's disbelief settles the validity of the inference.

Anselm describes God through an unsurpassability condition: no greater being can be conceived. Greater concerns excellence, rather than physical size. The description is intended to rule out a superior that thought could place above its subject. It is not simply a claim that God happens to outrank the beings currently known to us.

There is a difference between the greatest member of a familiar group and a being that cannot be surpassed in thought. The fastest runner in a particular race can still be slower than a conceivable runner. A winner selected from an existing group therefore need not meet Anselm's condition. This hypothetical comparison illustrates the scope of his description; it does not prove that the description is coherent or instantiated.

Nor should the condition be treated as a complete mental picture of God. Saying what would rule out a supposed candidate differs from comprehending every feature of its subject. Chapter XV makes the difficulty explicit by saying that God exceeds what can be conceived. The book therefore combines a condition governing thought about God with an acknowledgment of the limits of that thought.

That combination creates a question for the reader: how much understanding of a description is enough to support the argument? We can understand a rule without surveying everything it concerns. Yet merely pronouncing a sequence of words does not guarantee that those words describe a possible being. This gap is one of the points the debate needs to examine.

The painter in chapter II explains a distinction that Anselm accepts. Before making the picture, the painter has the planned work in the understanding. After making it, the painter understands that it also exists in reality. Thinking about something and recognizing its real existence are different conditions.

The denial of God is supposed to concede at least the first condition. According to Anselm, someone hearing and understanding the description has its subject in the understanding even while denying that it exists in reality. The argument does not begin by saying that every intelligible description already has an external instance.

This terminology can mislead a modern reader. A picture planned in the mind is not a tiny physical canvas inside the painter's head. Having something in the understanding names its availability to thought. Whether that availability licenses every later step is a further issue, rather than a fact settled by the spatial word in.

The distinction also gives a critic a starting point. Anselm must show why the unsurpassability condition changes the case. Repeating that the concept is understood cannot by itself supply that explanation, since the painter's unmade picture is understood too. The work of the proof occurs in the move from understood content to a contradiction in denying real existence.

A reconstruction of the argument can separate its main commitments:

1. A hearer understands the description of a being than which no greater can be conceived.
2. What is understood is in the understanding, in the sense just explained.
3. Suppose that the being described exists only in the understanding.
4. It could then be conceived as existing in reality, which Anselm treats as greater.
5. The supposed unsurpassable being would therefore be surpassable in thought.
6. Anselm rejects that contradiction and concludes that the being exists in reality as well.

This is a reconstruction, rather than a quotation. It makes visible the dependence of the conclusion on the comparison in the fourth step. The first two steps alone do not establish that an object exists outside thought. A critic can grant those steps and still question the later comparison or its use.

The method is a reduction to contradiction. It temporarily accepts the denial to show that it conflicts with another accepted commitment. Such reasoning is familiar outside theology: if an assumption entails incompatible conclusions, there is reason to reject it. The dispute here concerns whether Anselm has established the incompatibility without already assuming what he wants to prove.

One way to state the challenge is to distinguish the content of a description from whether anything satisfies it. If a description requires a certain feature, we can say that any actual instance must have that feature. It does not follow merely from that conditional statement that there is an instance. Anselm's argument needs its unsurpassability comparison to do more than restate a requirement.

Another question concerns the comparison itself. In what sense is an existing being greater than a merely conceived being? If the comparison concerns two actual beings, the denial has already been abandoned. If it concerns two descriptions, the critic can ask how comparing descriptions establishes a real object. These are local questions about the inference, not a general rule that reason can never establish existence.

Chapter III strengthens the result Anselm claims. It argues that a being whose nonexistence cannot be conceived is greater than one whose nonexistence can be conceived. A supposedly unsurpassable being that could be thought not to exist would consequently be surpassed. Anselm concludes that God cannot be conceived not to exist.

Necessary existence, in this discussion, contrasts with an existence that can fail. It does not mean being very old, lasting a long time, or resisting destruction exceptionally well. A thing could last throughout an indefinitely long period while its existence still depended on conditions that might not have obtained.

Chapter II and chapter III therefore have distinguishable targets. The former argues against existence only in the understanding. The latter argues against conceivable nonexistence. Reading the second as merely repeating the first loses the further claim Anselm wants to make.

The further claim does not remove every difficulty. A reader can ask whether inconceivability here tracks reality or a limitation in the person doing the conceiving. Failure to imagine an object's absence is different from establishing that its absence is impossible. Anselm's argument invokes what belongs to understanding the unsurpassable being, rather than a report that he personally finds imagining its absence difficult.

This is also a reason to avoid turning the passage into a modern possible-worlds proof without explanation. The text proceeds through greatness and conceivability. A later formal reconstruction would need to state its own interpretation of possibility and the rules connecting possibility with necessity. Those additions should be assessed as additions, not silently attributed to Anselm.

Chapter IV addresses an apparent conflict. If God's nonexistence cannot be conceived, how can the fool deny it? Anselm distinguishes thinking through words from understanding the thing those words signify. A person can utter or entertain words that do not express a fully grasped possibility.

The distinction has an intelligible use. Someone can repeat a description before working through its implications. Recognizing that possibility does not show that Anselm's own description has the implications he claims. The distinction explains how he reconciles denial with his conclusion, but it cannot independently establish that conclusion.

A critic therefore need not accept being classified as someone who merely fails to understand. The critic can identify a specific step, such as the real-existence comparison, and explain why it remains unproved. This directs attention toward reasons that can be examined rather than toward the presumed intelligence or sincerity of either participant.

Gaunilo's response, usually titled On Behalf of the Fool, challenges the reasoning through a lost island. He describes an island surpassing other lands in riches and delights, then asks whether its being understood obliges us to accept its real existence. The intended pressure comes from trying to reproduce the disputed move in a less plausible case (section 6).

The example is a parody of an inference, rather than geographical evidence against an island. Its force does not depend on a survey of every ocean. Gaunilo asks whether someone would have supplied a proof by reasoning from the island's described superiority to its existence. If that procedure appears inadequate, the advocate of Anselm's proof must explain the relevant difference.

Calling the example a refutation without qualification would move too quickly. A successful parody must preserve the feature on which the original inference depends. If it changes that feature, the defender can reject the analogy without accepting the island. The disagreement therefore concerns the scope of the reasoning as well as the plausibility of its conclusion.

Gaunilo also questions whether understanding words about an unsurpassable being gives the required understanding of their subject. That challenge reaches an earlier stage of the argument than the existence comparison. It is useful to separate the challenges, since defending the intelligibility of the description would not automatically validate the inference from it.

Anselm insists in his reply that the crucial description concerns what cannot be surpassed in thought. He resists replacing it with an ordinary claim that something is greater than all others. The distinction connects the reply with chapter III: a being meeting his condition cannot be treated like a contingent object whose nonexistence remains conceivable.

In the reply's island passage, he challenges Gaunilo to find something other than the unsurpassable being to which the argument properly applies. His answer should not be reported as the simple observation that one could always add another palm tree. That is a possible modern illustration of why island perfection is difficult to specify, not the wording or entire reasoning of his reply.

For example, imagine defining the best possible holiday island. More sunshine might improve one visitor's experience and worsen another's. Additional development might provide comforts while reducing solitude. These hypothetical tradeoffs show why an apparently clear comparative expression can conceal an unsettled standard. They do not demonstrate that every conception of an unsurpassable island is contradictory.

The corresponding question for Anselm is whether divine unsurpassability is sufficiently coherent and distinctive to block the parody. Saying that God is special does not yet show that the particular inferential step is justified. A defense needs to identify which feature makes the step succeed for God and why the feature is genuinely available to the argument.

The reply thus leaves a productive disagreement. Gaunilo asks how thought reaches an existing subject; Anselm insists that the relevant subject cannot be assimilated to contingent things. A reader can understand both positions without pretending that the island settles every issue or that invoking necessity automatically settles the reply.

The chapters following the existence argument investigate divine attributes, including justice and mercy. Chapter XV returns to the limit of conception, while the later chapters consider the good and joy sought by the speaker. These discussions show why identifying the work with a short proof can obscure the inquiry of which the proof is a part.

Its religious setting also limits a tempting conclusion about its effect on readers. Anselm intends understanding to belong to a life of seeking God. It does not follow that performing the argument reliably changes someone's psychology or produces religious conviction. Those would be additional claims requiring different evidence.

The philosophical work a reader can do is more specific. Identify the description, distinguish understanding from real instantiation, and locate the comparison that is supposed to connect them. Then ask whether necessary existence supplies an independent reason or restates a disputed condition. These questions preserve the argument's difficulty without requiring premature agreement or dismissal.

A useful reading exercise is to write two separate objections. In one, grant that the description is intelligible but deny that real existence increases greatness in the required way. In the other, question whether the description supplies a coherent possible subject at all. The first challenges the inferential comparison; the second challenges what the argument has available before making it. Keeping them separate prevents a reply to one objection from being mistaken for a reply to both.

The same discipline applies to the island. Ask whether it reproduces the relevant description, then ask whether the conclusion would follow even if that description were granted. These questions are deliberately open. They locate the work a defense must do without assuming either the success of the proof or the success of the parody.

Read the preface and chapters I-IV together. The preface states the intended proof, chapter I locates it within faith seeking understanding, and chapters II-IV develop the reasoning about existence, necessity, and denial. Chapter XV is a useful companion because it complicates any claim that Anselm thinks the human mind completely comprehends God.

• Anselm, Proslogium, Sidney Norton Deane's translation, for the complete work.
• Gaunilo, In Behalf of the Fool, and Anselm's reply, in the same translation, especially Gaunilo's section 6 and the reply's discussion of the island.

Keep the two texts beside one another when assessing the disputed inference. A paraphrase can clarify the steps while also changing them; returning to the description each writer uses helps reveal where that change has occurred.