The Ontological argument: Edexcel A* grade notes

Edexcel
Philosophy

AO1: Anselm

Anselm’s ontological argument involves a purely a priori analysis of the concept of God
It’s intended to be deductive, so that the conclusion follows from the premises with necessity.
Anselm illustrates: a painter has an idea in their mind before painting it in reality.
This distinguishes existing in the mind verses existing both in reality.
He then points to Psalm 14:1 “the fool says in his heart, ‘there is no God’.”
Atheists who reject belief in God must have an idea of God in their mind.
In Proslogion 2, Anselm argues:
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
God cannot exist in the mind alone. That would be incoherent, since then we could conceive of something greater: God also existing in reality.
Anselm develops the reasoning In Proslogion chapter 3:
P1. A necessary being whose nonexistence is impossible is greater than a contingent being whose nonexistence is possible.
C1. Therefore, God (as the greatest conceivable being), necessarily exists.
Malcolm interprets Anselm’s term ‘greater’ as referring to degrees of limitation, such as dependence on other things for existence.
God must be unlimited, without any of the contingencies of ordinary beings that make their non-existence possible.
So, a being greater than which none may be conceived is one whose nonexistence is impossible.
In his replies to critics, Anselm concludes that if such a being is logically possible, then it must exist.

AO1: The Ontological argument’s status as an (a priori deductive) proof

Ontological arguments are a priori, meaning they are based purely on reason, not the application of reason to experience.
They begin with a definition of God:
The greatest conceivable being (Anselm).
Supremely perfect being (Descartes).
An unlimited being (Malcolm).
The argument then involves a purely conceptual analysis of the logical implications contained within this definition.
The argument is deductive, meaning that if the premises are true, the conclusion must be true.
The truth of the premises logically entails the truth of the conclusion, rather than merely supporting it.
A priori deductive arguments are often associated with rationalism (e.g., Descartes), the view that knowledge of reality can be gained through reason alone.
Though, they can also be accepted by empiricists, who may regard God as a unique case where logical truth and matters of fact converge.
Deductive arguments aim at certainty.
If successful, they demonstrate what must be true, not merely what is probably true.
They are not defeasible in the same way as inductive arguments, since no further evidence could overturn a valid deduction.
Therefore, they can only be challenged by either rejecting the premises (denying soundness) or by denying that the conclusion follows from them with necessity (denying validity).
Such arguments rely on analytic entailment, where what is contained within a concept determines what must be true of it.
This is how the ontological argument claims that once the concept of God is properly understood, existence (or necessary existence) is entailed by it.
It intends to show that God’s non-existence is a contradiction.
That would establish the proposition “God exists” as an analytic truth (true by definition), the key criterion of which is that it cannot be denied without contradiction.
Standard deduction operates through analytic entailment of categories:
E.g.: “All men are mortal; Socrates is a man; therefore Socrates is mortal”
However in the ontological argument, the deduction proceeds not merely through categories but ‘modal’ concepts, such as possibility and necessity.
A necessary being is one that cannot fail to exist.
If such a being is even possible, then it must exist in all possible worlds, including the actual world.
The claim is that a priori reason discovers that analytic entailment functions just as decisively in this unique case.

AO1: Strength & weakness of the ontological argument: AO2 summary

Gaunilo’s ‘lost island’ objection
Weakness: The ontological argument appears to generate absurd conclusions when its logic is applied to contingent objects like a perfect island.
Strength: The argument distinguishes necessary from contingent beings and limits its applicability to necessary a being whose existence is contained within its definition.
Gaunilo: God beyond understanding
Weakness: The claim that God exists in the understanding is challenged by the view that God is beyond human comprehension.
Strength: The argument only requires understanding that whatever God is, God is the greatest conceivable being, rather than fully understanding God’s nature.
Kant: existence is not a predicate
Weakness: The argument is criticised for treating existence as a predicate, even though existence does not add anything to the concept of a being.
Strength: The argument specifically draws on the concept of a necessary being, a unique case where necessary existence could be defended as a predicate.
Kant: necessity doesn’t imply existence
Weakness: Defining God as a necessary being does not establish actual existence, since necessity may only describe how God would exist if God exists.
Strength: The argument is grounded in the powerful reasoning that God is a necessary being which therefore must exist, as its non-existence would contradict its nature.

AO2: Gaunilo’s ‘lost island’ objection

Gaunilo’s ‘reductio ad absurdem’ critique argues absurdity results if we apply Anselm’s logic to another case: the greatest possible island.
Following P2, if it’s greater to exist, then this island must exist.
This would work for the greatest possible version of anything, absurdly implying reality would be ‘overloaded’ with greatest versions of every possible thing.
Gaunilo is denying that the conclusion of God’s existence follows from the premises, so he is denying deductive validity.

Counter

However, Anselm replies that his argument is only intended to work in the case of God.
Descartes develops this line of thought by arguing that God’s essence includes necessary existence, whereas contingent beings like islands do not.
An island is contingent by definition (land enclosed by water), so its existence depends on water.
Therefore, like all things we observe, no matter how great or perfect an island is, it will still be contingent.

Evaluation

This reply defeats Gaunilo because a priori reasoning can only demonstrate existence of a necessary being.
A priori reasoning solely involves the analysis of the definition of concepts.
The definition of contingent beings is that their existence depends on something else.
The existence-conditions of contingent beings are therefore separate to their definition.
So, a priori reasoning cannot determine the existence of contingent things.
E.g.:
I can know a priori that a perfect island depends on water to exist, but not whether that water does exist.
So I can’t know a priori whether a perfect island, or any other contingent thing, exists.
For necessary beings, by contrast, there is no supra-definitional existence requirement.
The ontological argument therefore works exclusively in the case of the greatest conceivable being.
So, there’s a relevant difference between God and contingent things like Islands which explains why the logic works exclusively for God.
Its logic cannot be tested where it doesn’t belong, through reference to contingent beings.

AO2: Gaunilo’s critique that God is beyond our understanding

Gaunilo objects to P3, the claim that God is in our mind/understanding.
He invokes the classical notion that God is beyond our understanding.
In that case, Anselm can’t go on to conclude that God being the greatest being requires that he is not just in our understanding, but also in reality.

Response:

Anselm replies with an analogy: just because we cannot look directly at the sun, doesn’t mean we can’t see by sunlight.
Similarly, just because we can’t understand what God is, doesn’t mean we can’t understand that whatever God is, God is the greatest conceivable being.
We can understand that in a hierarchy, like greatness, there must be a highest point or ‘intrinsic maximum’, to use Plantinga’s modern term.
That’s all we need to understand for Anselm’s argument to work.
When we combine that with the premises that it is greater to exist, we can then deduce the conclusion that the greatest conceivable being exists.
We don’t need to understand what a maximally great being actually is, only that whatever it is, it is the pinnacle of the greatness hierarchy.

Evaluation:

Firstly, Gaunilo’s critique doesn’t work against Anselm’s 2nd form of his argument, which doesn’t rely on a premise about God existing in the mind.
Secondly, Gaunilo misunderstands P3 and commits a straw man fallacy, attacking a claim Anselm didn’t make.
In P3, Anselm uses ‘in intellectu’, meaning ‘can be conceived or thought about’, not ‘fully understood’.
Anselm didn’t mean God is in the mind in the sense of us having full knowledge of God’s substantive nature.
He just meant that we understand something proper to the form of God; that God is maximally great.
So, Anselm’s defence works because knowing God to be the greatest being, and knowing what it is to be the greatest being, are two distinct things.

AO2: Kant’s 2nd critique: existence is not a predicate

The ontological argument claims that denying God’s existence is an incoherent denial of what God is (a maximally great/perfect being).
Kant responds that this misunderstands what existence is by treating it like a predicate (a description of what a thing is).
He illustrates: there’s no conceptual difference between 100 coins in reality versus only in the mind.
If existence were a predicate, it would describe a quality of the real coins.
They would then have more qualities and thus be conceptually different to the mental coins.
But, they are not: 100 coins is just 100 coins, defined by the predicates of 100, round, shiny, etc.
So, existence is not a predicate.
This attacks the premise that existing is greater or more perfect.
If existence is not part of what God is, we can deny God’s existence without incoherently contradicting what God is.

Response:

However, Descartes’ epistemology doesn’t rely on Anselmian scholastic attribution of predicates to subjects, but on intuition.
We rationally appreciate that God is inseparable from existence, just as a triangle is inseparable from three sides.
So Kant’s criticism fails to target Descartes’ formulation.
Malcolm also defends Anselm:
Kant is correct, but only about contingent existence.
The reason for the existence of a contingent thing is dependence on something else, so is external and not a defining part of it.
However a necessary being contains the reason for its existence within itself.
So, necessary existence is a defining quality of a thing, in a way contingent existence is not.
So necessary existence is a predicate.

Evaluation:

So, both Anselm and Descartes’ approaches succeed against Kant’s criticism.
Kant makes the same mistake as Gaunilo, thinking an argument for a necessary being could be undermined by showing it fails when applied to contingent things like coins.

AO2: Kant’s 1st critique: necessity doesn’t imply existence

Gaunilo criticised the conflation of necessity in mental judgement with necessity in reality.
The Island (and Kant’s 100 coins) misapplied this to contingent things.
Kant’s first critique refines Gaunilo’s underlying point by applying it to necessity itself:
A triangle necessarily has three sides.
This proves that if a triangle exists, then it necessarily has three sides.
Similarly, God is defined as necessarily existing.
But again, this only shows that if God exists, then God exists necessarily.
So here Kant grants that necessary existence can be predicated of God, but denies this establishes God’s actual existence.
If God exists, it’s contradictory to deny God’s necessity.
But if God does not exist, then neither does God’s necessity.

Counter:

Malcolm responds that Kant’s criticism is incoherent because a necessary being must exist.
If God is a necessary being, then God must exist.

Evaluation

However, Hick uses modern distinctions between types of necessity to counter Malcolm.
Anselm & Malcolm claim a maximally great/unlimited being cannot have contingent dependencies.
They then treat non-contingency as entailing logical necessity.
Hick objects that non-contingency only means that God is defined as a non-dependent, eternal, self-explaining being (aseity), which Hick calls ‘ontologically necessary’.
This does not entail that such a being must exist, but only how it would exist if it did.
Its existence is not logically required but would be a metaphysical ‘sheer fact’ (Hick).
So, the ontological argument cannot establish the incoherence of God’s non-existence.
It at most proves that if God exists, then God exists in a metaphysically special way (with aseity).
Gaunilo and Kant’s point that conceptual necessity doesn’t entail actual existence was right, but needed these newer distinctions between types of necessity to make clear why.