“no one who understands what God is can conceive that God does not exist” – Anselm

Introduction

In 1077 AD, St Anselm created an argument for the existence of God which came to be known (thanks to Kant) as the Ontological argument. Ontology refers to ‘being’ or ‘existing’ or the nature of being / what exists.

Ontological arguments are a priori, based solely on an analysis of the concept of God. They essentially argue that if you think carefully about what God is, you’ll understand that God must exist.

The argument has proven controversial, with many of its critics actually being religious themselves. It’s less popular than the cosmological and teleological arguments amongst religious philosophers as a defence of theistic belief. Those arguments identify features of the world (motion, causation, complexity/purpose) and attempt, a posteriori, to infer God’s existence as the explanation for the existence of the universe, or for why the universe is the way it is.

The a priori ontological argument is trying to do something quite different. It’s actually difficult to put one’s finger on what the argument is doing. First impressions of the argument very widely, but a common one, shared by its first major critic Gaunilo, is to think the argument is borderline absurd. Nonetheless, something about the argument has provoked serious attention from many great philosophers, whether positive or negative. There could be more to the argument than first impressions might suggest. Russell captured this well:

“it is easier to feel convinced that [the Ontological Argument] must be fallacious than it is to find out precisely where the fallacy lies.” – Bertrand Russell

St Anselm’s Ontological argument

P1. God is the greatest conceivable being (by definition)
P2. It is greater to exist in reality than the mind alone
P3. God exists in the mind
C1. Therefore, God exists in reality

Anselm uses the illustration of a painter who has an idea of what they will paint in their mind before painting it in reality. This illustrates the distinction between our idea of something existing in the mind alone, verses existing both in the mind and in reality.

Anselm points to Psalm 14:1 “the fool says in his heart, ‘there is no God’.”

An atheist says they do not believe in God. That implies they at least have an idea of God in their mind.

The force of Anselm’s argument is that God cannot be an idea that exists in the mind alone. That would be incoherent, since then we could conceive of something greater, i.e., God also existing in reality. Yet, God is the greatest being, so conceiving of anything greater is incoherent. So, our idea of God must therefore be of a being that exists in reality. To say that God does not exist in reality is to say that the greatest being is not the greatest being. It is self-contradictory.

“that, than which nothing greater can be conceived, cannot exist in the understanding alone: then it can be conceived to exist in reality; which is greater. Therefore, if that, than which nothing greater can be conceived, exists in the understanding alone, the very being, than which nothing greater can be conceived, is one, than which a greater can be conceived. But obviously this is impossible. Hence, there is no doubt that there exists a being, than which nothing greater can be conceived, and it exists both in the understanding and reality.” – Anselm.

Proslogion chapter 3 & necessary existence.

Anselm’s ontological argument underwent a revival of interest in the 20^th century thanks to N. Malcolm and C. Hartshorne. They argued that the strongest form of the argument was in chapter 3 of the Proslogion. There, Anselm concludes not merely that God exists, but that God is a necessary being (one which contains its own reason for existence; whose existence doesn’t depend on anything else).

A necessary being whose nonexistence is impossible is greater than a contingent being whose non-existence is possible. So God, as the greatest conceivable being, must be necessary.

Malcolm interprets Anselm’s term ‘greater’ as referring to whether a being is limited, e.g., due to depending on something else for its existence. The greatest conceivable being would not have any dependence or limitation. God doesn’t have any of the contingencies which could result in non-existence. So, a being greater than which none may be conceived is one whose nonexistence is impossible. Hartshorne calls this insight “Anselm’s discovery”. Later, in his response to critics, Anselm writes:

“If it [a being greater than which cannot be conceived] can be conceived at all it must exist.” – Anselm.

God is a being whose non-existence is impossible. So, if such a being is logically possible, then it must exist.

Whether God is within our understanding

A strength of the ontological argument its definition of God

Anselm uses a theologically and philosophically convincing definition of God, carefully designed to avoid the problem of defining something that is beyond our understanding. Anselm presents an analogy. We can’t fully look at the sun but can still see daylight. Similarly, we can’t fully know God, but can at least understand that he is the greatest conceivable being.

“If you say that what is not entirely understood is not understood and is not in the understanding: say, then, that since someone is not able to gaze upon the purest light of the sun does not see light that is nothing but sunlight.” – Anselm

Weakness: God is not ‘in’ the mind/understanding

Gaunilo raises an objection to P3; the premise that the greatest conceivable being exists in the mind/understanding. Gaunilo draws on the traditional Christian premise that God is beyond our understanding to argue that God therefore cannot be in the understanding.

Anselm cannot then proceed to reason about whether it would be greater also in reality. The ontological argument seems to fail because it relies on our ability to understand and reason about things that are beyond our ability to understand or reason about.

Aquinas also made this argument against Anselm – that God’s nature, such as the ‘eternal law’ is beyond our understanding and that people have different understandings of God.

“Perhaps not everyone who hears this word “God” understands it to signify something than which nothing greater can be thought” – Aquinas.

Gaunilo even doubts that we can understand this idea of the greatest conceivable being: 

“of God, or a being greater than all others, I could not conceive at all” – Gaunilo.

“So much for the assertion that this supreme nature already is in my understanding.” – Gaunilo.

Gaunilo’s ‘lost island’ response to Anselm

Gaunilo denies that the ontological argument is actually a valid deductive argument, attacking the inference from the premises to the conclusion of God existing in reality

“I have in my understanding all manner of unreal objects” – Gaunilo.

“I should not admit that this being is in my understanding and concept even in the way in which many objects whose real existence is uncertain and doubtful, are in my understanding and concept. For it should be proved first that this being itself really exists somewhere; and then, from the fact that it is greater than all, we shall not hesitate to infer that it also subsists in itself.” – Gaunilo.

Anselm’s argument could succeed in showing that if God exists, then God is the greatest being and even that it subsists in itself, i.e., has necessary existence. However, this is not enough to show that God does exist necessarily.

“he who says that this being exists, because otherwise the being which is greater than all will not be greater than all, does not attend strictly enough to what he is saying. – Gaunilo.

Gaunilo illustrates this with the case of a perfect lost island, an illustration of a thing whose real existence is ‘uncertain and doubtful’ yet exists in his understanding as a concept.

Applying the logic of Anselm’s argument to this island has an absurd result (reductio ad absurdum). It is greater for this island to exist in reality, so it must exist. This would work not just for an island. The greatest or supremely perfect member of every category must exist. This is sometimes called the ‘overload’ objection because it suggests that reality would be overloaded with greatest/perfect things.

Counter

Firstly, Gaunilo’s critique is unsatisfying as far as it goes. Gaunilo asserts, but does not demonstrate, the absurdity of Anselm’s logic proving the existence of a perfect island. He merely asserts that such reasoning must either be a joke or a symptom of foolishness.

To demonstrate absurdity requires showing a contradiction, which Gaunilo has not shown. Proving existence a priori might seem counter-intuitive. Gaunilo remarks that the logic seems like joke or sign of foolishness when applied to the island. Nonetheless, Gaunilo has not demonstrated it actually absurd. Perhaps a perfect island does indeed exist.

Anselm himself made a different response to Gaunilo. He insisted that a proper understanding of his argument showed that it can only prove the existence of God. Testing the logic through applying it to a different case like an island is not valid.

Something is greater if it doesn’t depend on anything for its existence. By definition an island is land enclosed by water. Definitionally then, no matter how great or perfect an island is, in order to be an island it will be dependent on something else to exist, such as an ocean, planet, sun, etc. So, the greatest possible Island will be contingent, which means by definition it could either exist or not.

This is why a priori analysis of its definition cannot prove its existence. The existence of contingent beings cannot be proven a priori because their existence is not a matter of definition. Their existence is a matter of whether what they depend on exists.

There is nothing in the concept of the greatest being that involves dependence, making it a necessary being. So, the reason for the logic not working in the case of the island (or any contingent being) does not apply for God.

Descartes’ Ontological argument

Descartes aimed to strengthen the ontological argument through founding it on his rationalist epistemology. This claimed that we can gain certain knowledge of some truths a priori.

Anselm is often called the father of Scholasticism, a theological movement influenced by Aristotle’s approach to argumentation. At its core is subject-predicate analysis. Propositions are combinations of subjects and predicates which assert something as true or false.

Descartes rejected scholasticism. He instead argued that the foundation of knowledge was intuition. Intuition operates through direct intellectual awareness, not the indirect analysis of linguistic representation employed by logical terms. Intuition provides absolute certainty. We can bring ideas before our mind and apprehend truths about them due to the psychological character in which they strike us.

E.g., we intuitively know that a triangle has three sides, because it is impossible to bring a triangle before our mind separated from having three sides. Similarly, we cannot conceive of a supremely perfect being separated from existence. We thus rationally appreciate that God contains the perfection of existence. Intuition shows us that God exists.

“the idea of God, or a supremely perfect being, is one that I find within me just as surely as the idea of any shape or number. And my understanding that it belongs to his nature that he always exists is no less clear and distinct than is the case when I prove of any shape or number that some property belongs to its nature” – Descartes

Descartes did put it into the form of a deductive argument:

P1 – I have an idea of a supremely perfect being which contains all perfections
P2 – Existence is a perfection
C3 – God exists

The argument is deliberately short, suggestive of Descartes’ view that God’s existence can be known intuitively.

Hume’s empiricist response to the ontological argument

Hume is an empiricist who rejects a priori demonstrations of existence and the concept of a ‘necessary being’.

Truths of logic/definition are true or false no matter the factual state of the universe. There is no logically possible factual state of affairs that is incompatible with ‘1+1=2’, for example. Hume thinks this shows that logical truth and factual truth are distinct, including our means of knowing them. This is called ‘Hume’s fork’.

Analytic: true by definition. Cannot be denied without contradiction. E.g. “a bachelor is an unmarried man”.
A priori reasoning involves the analysis of the relation between ideas. So, only analytic truths can be known a priori.

Synthetic: true because of the way the world is. Can be denied without contradiction.  “E.g. “The sun will rise tomorrow”.
A posteriori reasoning involves experience of the factual state of the world. So, only synthetic truths can be known a posteriori.

Applying this to the ontological argument:

“there is an evident absurdity in pretending to demonstrate a matter of fact, or to prove it by any arguments a priori” – Hume.

“It will always be possible for us at any time to conceive the non-existence of something we formerly conceived to exist; the mind can never have to suppose some object to remain always in existence, in the way in which we always have to conceive twice two to be four” – Hume

A necessary being must exist. So, we shouldn’t even be able to conceive of it not existing. However, Hume claims that whatever we conceive of as existing, we can conceive of as not existing. It follows that there is no being that we cannot conceive to not exist. So, our mind is incapable of giving meaning to the idea of a being existing necessarily. Hume concludes:

“The words, therefore, necessary existence, have no meaning.” – Hume.

Any belief we have about what exists could be imagined as either true or false. Therefore, we cannot coherently understand any being to be logically necessary.

Whether something exists is a contingent matter of fact. It cannot be logically necessary. The term “necessary existence” seems to ignore the disconnect between logical and factual truth established by Hume’s fork.

The consequence is that any claim about what exists (existential propositions, like ‘God exists’) can be denied without contradiction.

Existential propositions are therefore always synthetic. So, they can only be known a posteriori. The premises of ontological arguments are claimed to be known priori. In that case, they cannot allow us to know the conclusion that God exists.

Kant’s objection that existence is not a predicate

Kant develops Hume’s criticism, arguing that the reason any being is conceivably non-existent is that existence is not a property a thing possesses. Existence thus cannot be an essential property of a thing, inconceivably separate from it.

Descartes implies that perfection is a defining attribute of God. Anselm argues God must exist in order to be God. They both try to show that denying God’s existence denies what God is. This seems to treat ‘existence’ as if it described a defining property a thing possesses. That would make the word ‘exists’ a predicate.

Kant objects. If existence were a predicate, it would be added to our concept of a thing that exists. A thing that exists would be conceptually different to that same thing when not existing.

If I say my cat exists, I do not describe a feature of the cat. I may be describing reality in a general sense, so Kant allows that existence can be a ‘logical predicate’. However, existence is not a ‘real predicate’, meaning it does not describe an attribute of a thing itself.

To use Kant’s example, if existence were a predicate, then 100 thalers (coins) in reality would be conceptually different to 100 thalers in the mind.

However, the concept ‘100 thalers’ is the same whether a mere concept in your mind or instantiated in reality. A thing is what it is, regardless of whether it exists or not. 100 thalers is just 100 thalers. It has the predicates of shininess, roundness, 100, etc. Being only in the mind doesn’t make the concept somehow less of a complete description of what 100 thalers is. So, existence is not a description of a thing. It is not a predicate.

We cannot determine whether a thing exists merely through understanding what it is. A thing is not less great or perfect simply for not existing. So, Anselm and Descartes seem incorrect when they claim it’s incoherent to think of the greatest or supremely perfect being not existing. 

Counter: Kant’s criticism faces two counters.

Firstly, Kant’s objection fails to target Descartes’ version of the argument. Anselm understands ‘God exists’ as a subject-predicate relationship.

Descartes’ rejection of Aristotelian subject-predicate analysis means he can’t be accused of inferring God’s existence by assuming that existence is a predicate of God.

Descartes’ argument doesn’t operate by assigning predicates to subjects, but by determining whether the idea of a supremely perfect being can be clearly and distinctly perceived while excluding necessary existence from it through a purely intellectual operation.

Furthermore, Malcolm defended Anselm’s approach, arguing that Kant only shows that contingent existence is not a predicate.

Something is contingent if it is dependent on something else for its existence. The reason for the existence of a contingent thing is external to it and so does not describe or define it. However, a necessary being doesn’t depend on anything else for its existence. It contains the reason for its existence within itself. ‘necessary existence’ therefore does describe something about a being. It is a defining part of a thing in a way that contingent existence is not. So, necessary existence is a predicate.

Malcolm’s ontological argument

Norman Malcolm created his own version of the ontological argument, referring to God as an unlimited being.

Malcolm uses modal logic, which involves analysis of the logical consequences of necessity and possibility.

P1. God either exists or does not exist.
P2. If God exists, God cannot go out of existence as that would require dependence on something else. So, if God exists, God’s existence is necessarily
P3. If God does not exist, God cannot come into existence as that would make God dependent on whatever brought God into existence. So, if God does not exist, God’s existence is impossible.
C1. So, God’s existence is either necessary or impossible
P4. The concept of God is not self-contradictory (like a four-sided triangle), therefore God’s existence is not impossible.
C2. Therefore, God exists necessarily.

Malcolm points to this quote from Anselm’s reply to Gaunilo:

“If it [the thing a greater than which cannot be conceived] can be conceived at all it must exist. For no one who denies or doubts the existence of a being a greater than which is inconceivable, denies or doubts that if it did exist its non-existence, either in reality or in the understanding, would be impossible. For otherwise it would not be a being a greater than which cannot be conceived. But as to whatever can be conceived but does not exist: if it were to exist its non-existence either in reality or in the understanding would be possible. Therefore, if a being a greater than which cannot be conceived, can even be conceived, it must exist.” – Anselm

Malcolm remarks:

“What Anselm has proved is that the notion of contingent existence or of contingent nonexistence cannot have any application to God. His existence must either be logically necessary or logically impossible. The only intelligible way of rejecting Anselm’s claim that God’s existence is necessary is to maintain that the concept of God, as a being a greater than which cannot be conceived, is self-contradictory or nonsensical” – Malcolm.

Malcolm’s interpretation of Anselm is that neither contingent existence nor contingent non-existence can apply to God. The only way to deny God’s existence is for God to be necessarily non-existent, i.e., incoherent.

Kant’s objection that necessity doesn’t imply existence

Gaunilo’s underlying point was to show the difference between existing in the mind and existing in reality. His lost island was an illustration of that. Anselm defeated the relevance of the lost island, but arguably not Gaunilo’s underlying point. Kant developed this type of objection.

Kant takes Descartes example of a triangle. It is necessary that ‘having three sides’ is part of the concept of a triangle. This doesn’t mean ‘three sides’ are necessary. It means that if a triangle exists, then it necessarily has three sides. We could not accept a triangle, but deny three sides, without contradiction. Yet we could deny the triangle exists, and then also its three sides.

Similarly, the ontological argument shows that ‘necessary existence’ is part of the concept of God. Kant’s objection is that this only shows that if God exists, then God exists necessarily. It doesn’t show that God-the-necessary-being does exist. If God does not exist, then neither does God’s necessity.

It would be contradictory to say that God exists, but not necessarily. Yet we can still deny that God exists, and with that, deny that God’s necessity exists. God may be necessary, but if God does not exist then God’s necessity does not exist.

Like Gaunilo, Kant is drawing a distinction between judgement and reality. A priori reasoning showing that existence is necessary to the definition of God in our minds is not the same as showing that God necessarily exists in reality.

“The unconditioned necessity of judgements is not the same as an absolute necessity of things” – Kant.

“the illusion of this logical necessity has proved so powerful that when one has made a concept a priori of a thing that was set up so that its existence was comprehended within the range of its meaning, one believed one could infer with certainty that because existence necessarily pertains to the object of this concept, i.e., under the condition that I posit this thing as given (existing), its existence can also be posited necessarily” – Kant.

Counter: Malcolm argued this critique from Kant is incoherent.

“I think that Caterus, Kant, and numerous other philosophers have been mistaken in supposing that the proposition ‘God is a necessary being’ (or “God necessarily exists”) is equivalent to the conditional proposition ‘If God exists then He necessarily exists’ … Can anything be clearer than that the conjunction ‘God necessarily exists but it is possible that He does not exist’ is self-contradictory?” – Malcolm

Kant seems to accept that the ontological argument shows that God is a necessary being. Malcolm argues this means God is a being which is characterised by the impossibility of non-existence. In that case, it can’t be possible for God to not exist.

Malcolm concludes It is incoherent of Kant to grant necessity to God while maintaining the possibility of God’s non-existence. So, the Ontological argument does show that God-the-necessary-being actually exists.

Hartshorne agrees with Malcolm, adding that the only valid way to suppose that God does not exist is to suppose that God’s existence is self-contradictory (logically impossible).

If one accepts the logical coherence of a being which contains the impossibility of non-existence, one must accept that it necessarily exists.

This is the insight behind Malcolm’s premise that God is either necessary or impossible.

Kant seems to want to propose a third option, that God is necessary and yet could not exist. Yet Malcolm argues that is a contradiction in terms.

The question of God’s logical impossibility

Hartshorne claims there are two ways one could attack the ontological argument. One is the ‘empirical’ method of Hume, Kant & Hick. They argued that existence can never be logically necessary, even for a being which contains the impossibility of non-existence. Hume says the concept of a necessary being is not even conceivable. Kant says the necessity is just in our mental judgement, Hick says it only involves ontological necessity, not logical.

Since Leibniz, proponents of the ontological argument have accepted that it depends on God being logically possible. If empiricist approaches fail, Hartshorne argues this is the only remaining other way to attack the ontological argument. That is, to argue that the God of classical theism, or the idea of a being that contains the impossibility of non-existence, is actually incoherent and thus impossible.

There are numerous philosophical debates about the coherence of God, including:

• The paradox of the stone
• The Euthyphro dilemma
• The incompatibility of free will and omniscience
• The logical problem of evil

These debates over the logical coherence of God are ongoing and have a long history. Neither side seems to have a knock-down criticism of the other.

Modern defenders of the Ontological argument, Malcolm, Hartshorne and Plantinga, agree that our inability to know for certain that God is coherent does limit what it can achieve by itself.

The ontological argument at most shows that if God is logically possible, then God necessarily exists.

Malcolm’s version incorporates this dependence, making God’s logical possibility a premise of the argument.

Plantinga accepts that the ontological argument can at most make religious belief rational, but cannot prove that God actually does exist.

“reformulated versions of St. Anselm’s argument … cannot, perhaps, be said to prove or establish their conclusion. But since it is rational to accept their central premise, they do show that it is rational to accept that conclusion” – Plantinga.

Plantinga admits it is rational to believe that God’s existence is logically impossible. However, he maintains that it is also rational for a theist to believe that God’s existence is logically possible, from which the ontological argument then shows that it is rational to believe that God exists. If a being whose non-existence is impossible could exist, then it must exist. It must exist, because its non-existence is impossible. So it could only not exist if such a being is somehow logically absurd.

This was Anselm’s initial insight, that if God is even conceivable, then God must exist.

“if a being a greater than which cannot be conceived, can even be conceived, it must exist.” – Anselm

 

Articles & web pages

Media

 

Books