Arguments for the Existence of God

Scholasticism

• Scholasticism is a movement in philosophical theology which aimed to understand all of Christian doctrine as a rational system.
• Main principle: Theological knowledge can be attained by argumentative reasoning, not merely by faith and revelation.
  • In that sense, it includes and extends the project of perfect being theism to the whole of the Christian religion, not just its conception of God.
  • It shares with PBT too the desire to reconcile God with classical philosophy – especially Aristotle, whose works had only recently been retransmitted to the West, at the time of Scholasticism’s foundation.
• Methodologically, the techniques of conceptual analysis and attention to important distinctions has many affinities with contemporary analytic philosophy.

Anselm and Aquinas

• We won’t concern ourselves with Scholastic arguments that are directed at specifically Christian claims, but rather some influential arguments for the existence of a perfect being – God – by two of the most significant scholastic philosophers.
  • St Anselm (1033–1109), Benedictine monk at Bec in Normandy, France, 1060-1093, Archbishop of Canterbury 1093–1109. Considered the founder of scholasticism.
  • St Thomas Aquinas (1225–1274), Dominican friar who held many positions, notably as student and teacher at the University of Paris. Famed for bringing Aristotelian philosophy and Christian doctrine together, his writings remain the foundation for Catholic (‘Thomist’) theological training to this day.
• Both, but Aquinas in particular, wrote voluminously – we will not be surveying Scholastic thought, but trying to bring a couple of strands of Scholastic argument into modern form and context.

What is an Argument For the Existence of God?

• An argument is not a dispute or disagreement (though disputing people may make use of arguments).

• Rather, arguments are something we give in order to persuade:

  Giving an argument … is something more like making a case. An argument presents reasons [the premises] that purport to favour – or support – a specific claim [the conclusion]. … If the argument is well-constructed, the premises provide reasons in favour of the conclusion. (Eagle, Magnus, and Button 2022: §1)

• An argument for the existence of God is one with the conclusion God exists.

• Which God? Our Scholastic-inspired arguments aim at the ‘perfect being’ conception of God discussed in the previous lecture.

  • But we don’t have any armchair guarantee that every argument with the conclusion ‘God exists’ is aiming at the same conception of God. Premises grounded in specific scriptural evidence or revelation might be aiming at a quite different conception of God.

When is an Argument for God Successful?

• Consider the following virtues displayed by a good argument:
  1. It’s a valid argument: the conclusion really does follow from the premises;
  2. It’s a sound argument: the premises are in fact true;
  3. It’s a plausible argument: the premises are known to be true, or at least rationally believed, by the intended audience;
  4. It’s a persuasive argument when the premises are known/rationally believed by an audience of people who do not already accept the conclusion.
• Arguments can exhibit any virtues earlier on this list without exhibiting any of the later ones: validity doesn’t require soundness, soundness doesn’t require plausibility, etc.
• When is an argument successful? Does it have to be persuasive (e.g., can be used to convince unbelievers of God’s existence) – or does it merely have to be plausible (so it can be used to defend existing religious belief, even if it can’t be used gain new converts)?

Kant’s Taxonomy of Arguments for God

All the paths leading to this goal begin either from determinate experience and the specific constitution of the world of sense as thereby known, and ascend from it, in accordance with laws of causality, to the supreme cause outside the world; or they start from experience which is purely indeterminate, that is, from experience of existence in general; or finally they abstract from all experience, and argue completely a priori, from mere concepts, to the existence of a supreme cause. The first proof is the physico-theological, the second the cosmological, the third the ontological. There are, and there can be, no others. (Kant 1787: A 590/B 618)

• The ‘physico-theological’ are now known typically as teleological, but otherwise Kant’s distinction remains current. In this lecture we’ll consider Anselm’s ontological argument, and a cosmological argument due to Aquinas.
  • We leave teleological arguments for next lecture.

The Ontological Argument

Anselm’s Ontological Argument

Lord … we believe You to be something than which nothing greater can be thought. Or is there, then, no such nature [as You], for the Fool has said in his heart that God does not exist? But surely when this very same Fool hears my words “something than which nothing greater can be thought,” he understands what he hears. … So even the Fool is convinced that something than which nothing greater can be thought is at least in his understanding; for when he hears of this [being], he understands [what he hears], and whatever is understood is in the understanding. But surely that than which a greater cannot be thought cannot be only in the understanding. For if it were only in the understanding, it could be thought to exist also in reality – something which is greater. Therefore, if that than which a greater cannot be thought were only in the understanding, then that than which a greater cannot be thought would be that than which a greater can be thought! But surely this [conclusion] is impossible. Hence, without doubt, something than which a greater cannot be thought exists both in the understanding and in reality. (Anselm 1078: 93–94)

The Argument

• The broad idea is clear: Anselm thinks that there is something self-contradictory about the claim something than which a greater cannot be thought does not exist in reality, and so its negation must be true.
  • Since he identifies God with this greatest thinkable (conceivable) being, this is an argument for the existence of God.
• It is an ontological argument because it is purely a priori – nothing from experience is needed, since this argument is supposed to work just because of the content of the concept god (Oppy 2021: §8).
• ‘The fool’ is a reference to Psalm 14: ‘The fool hath said in his heart, There is no God’ – so this is supposed to be a persuasive argument.
• But does this argument work, and how? What, precisely, is this supposed self-contradiction in the idea of a non-existent God?

The Argument, Reconstructed (Oppy and Scott 2010: 74)

(AN1)

The greatest conceivable being (‘that than which a greater cannot be thought’) exists in the understanding. (Premise)

(AN2)

The greatest conceivable being can be conceived to exist in reality. (From AN1)

(AN3)

‘to exist also in reality … is greater [than existing only in the understanding]’ (Premise, Anselm (1078), pp. 93–4)

(AN4)

It is impossible to conceive of a being which is greater than ‘that than which a greater cannot be thought’. (Premise, obvious?)

(ANH)

The greatest conceivable being exists only in the understanding. (Supposition for reductio)

(AN5)

We can conceive of a being which is greater than the greatest conceivable being. (ANH, AN2, AN3)

(ANC)

(Hence) the greatest conceivable being exists not only in the understanding, but also in reality. (AN5, AN4, by reductio of ANH)

Gaunilo’s Perfect Island Parody

• If AN1 is correct, then so is the greatest conceivable \(F\) exists in the understanding, for pretty much any \(F\):

  it is said that somewhere in the ocean is an island, which, because of the difficulty, or rather the impossibility, of discovering what does not exist, is called the lost island. And they say that this island has an inestimable wealth of all manner of riches and delicacies in greater abundance than is told of the Islands of the Blest; and that … it is more excellent than all other countries….

  Now if someone … went on to say, as if by a logical inference: ‘You can no longer doubt that this island which is more excellent than all lands exists somewhere…’ … I know not which I ought to regard as the greater fool: myself, supposing that I should allow this proof; or him, if he should suppose that he had established with any certainty the existence of this island. (Gaunilo 1903: §6)

Diagnosis?

• Gaunilo’s parody, if we accept that it is parallel to Anselm’s, allows us to infer the existence of a greatest conceivable island iff Anselm’s argument allows us to infer the existence of a greatest conceivable being.
• So presumably we will reject both as unsound and perhaps invalid. But that doesn’t show us why the argument is flawed. We still need to diagnose where the fault lies.
• I think the reconstructed argument is arguably valid (I will return to the step at AN5 below, however). So we shall have to address the premises, to see whether they are true, and whether they are persuasive.
  • You might worry about how AN2 follows from AN1, but I suggest: If the idea of a greatest conceivable being is coherent, then it is likewise coherent to suppose that being to be actualised:
    • Just as believing that \(p\) is believing \(p\) to be true, so too conceiving of \(X\) is conceiving \(X\) to exist.

Worries about AN1

• Is AN1 true?
  • Consider a number than which no greater can be conceived – we can understand what this would involve, but it is not a coherent idea, given the axioms of arithmetic. So not obvious on grammatical grounds that AN1 is true.
  • Presumably Anselm will say that, for finite beings it is always possible to conceive a greater (still finite) one, but God is absolutely infinite – and for an infinite being, it is coherent to suppose it has inexceedable greatness.
• Is this conception of God common ground?
  • Does the Fool really accept the same conception as the believer? For isn’t it part of the fool’s conception that this being does not exist in reality, which is no part of Anselm’s conception?
  • If not, is this a problem?

Is Existence a Greatness-Making Quality? (AN3)

• Maybe it would be better that the greatest conceivable being exist than not. But isn’t that a good feature of the possible situation in which he exists, rather than a good feature of the greatest conceivable being?
  • The argument could not show that God is better in those possibilities in which he exists than in those in which he does not – how is that even coherent?
  • Perhaps Anselm is mislead by his talk of ‘existing in the understanding’.
• Suppose on the other hand that the greatness of existence does accrue to the existing thing. Consider me, and consider a handsomer, smarter counterpart whose only flaw in comparison to me is that he fails to exist.
  • We may say my counterpart is greater than me, no matter how slight the degree of superiority he has over me (Oppy and Scott 2010: 80).
  • But then I can be exceeded in greatness by a non-existing thing with arbitrarily small improvements from me.
  • So my existence seems to contribute nothing to my degree of greatness.
• Finally: wouldn’t God be even greater if he created the world without even existing‽

Which Being Are We Talking About? (Oppy and Scott 2010: 79)

• At AN5, we have a contradiction with AN4: conceiving of a being greater than the greatest conceivable being.
• But aren’t we just conceiving of the very same greatest conceivable being – just conceiving of it being slightly greater than it actually is?
  • The contradiction needed for a reductio requires that when we conceive of a being B existing under the hypothesis that God doesn’t exist, B is distinct from God.
  • But if all we are conceiving is the possible situation in which God exists from the point of view of our supposition that the actual situation is one in which God does not exist, then we are making a consistent supposition about how the greatest possible being could have been, rather than an inconsistent supposition about another being that is greater than the greatest possible being.
  • If so, AN5 doesn’t follow: all we should actually deduce is this claim, which appears to be consistent with AN4:

  (AN5*)

  We can conceive of the greatest conceivable being as having more greatness than it actually has.

Existence in the Understanding

• The argument relies extensively on existence in the understanding.
• Anselm’s text spells this out in a way that makes it look like existing in the understanding is a kind of existence which just follows from being a coherent idea.
  • The example of the painter who imagines his work before creating it – Anselm suggests that the painting already exists in some weaker sense before it is created.
• But is this really a kind of existence?
  • There is a coherent idea of an extinct creature. Implausible to think that, just by being conceived, this creature that no longer exists nevertheless exists.
• But the argument does not go through equally well if we replace premise AN3 with the claim that it is greater to exist than to be merely conceived.
  • That may be true, but then I am already greater than a merely hypothetical greatest conceivable being – though it would be greater than me, were it to exist.

Being and Existing

• Probably the clearest way I know of to develop the idea that existing in the understanding is a kind of existing is the Meinongian approach.
  • Meinong was interested in claims like some things do not exist, and he drew a distinction between being and existing to account for them.
• Some have tried to reconstruct Anselm’s argument using this distinction (Oppenheimer and Zalta 1991). The logical argument they offer is valid, and (if one accepts the framework) even may be sound – the essential premise corresponding to AN3 is that anything which merely is without existing is less great than something which is and exists.
  • Again, though, we might find this premise plausible only if we think that things which don’t exist nevertheless have being, so that existence is a feature, like being bald, that some things have and others lack, and which therefore can contribute to comparisons of greatness.

The Cosmological Argument from First Causes

Aquinas’ ‘Second Way’ to God

The second way is based on the notion of an efficient cause:

We find that among sensible things there is an ordering of efficient causes, and yet we do not find – nor is it possible to find – anything that is an efficient cause of its own self. For if something were an efficient cause of itself, then it would be prior to itself – which is impossible.

But it is impossible to go on to infinity among efficient causes. For in every case of ordered efficient causes, the first is a cause of the intermediate and the intermediate is a cause of the last – and this regardless of whether the intermediate is constituted by many causes or by just one. But when a cause is removed, its effect is removed. Therefore, if there were no first among the efficient causes, then neither would there be a last or an intermediate. But if the efficient causes went on to infinity, there would not be a first efficient cause, and so there would not be a last effect or any intermediate efficient causes, either – which is obviously false. Therefore, one must posit some first efficient cause – which everyone calls a God. (Aquinas 1274: 15)

Reconstructing the Argument

(AQ1)

Events form an actual causal chain (‘there is an ordering of efficient causes’).

(AQ2)

Causal chains cannot involve infinitely many causes.

(AQI)

The actual causal chain is finite (it is ‘obviously false’ that it ‘go on to infinity’). (AQ1, AQ2)

(AQ3)

If there are only finitely many causes in a causal chain, and each has a cause, then at least one of them can be found on its own causal chain.

(AQ4)

Nothing can be found on its own causal chain (nothing is ‘prior to itself’).

(AQ5)

A causal chain with finitely many causes has a first cause: something which causes without being caused itself. (AQ3, AQ4)

(AQ6)

The actual causal chain has a first cause. (AQI, AQ5)

(AQ7)

This actual first cause is God.

(AQC)

God exists. (AQ6, AQ7)

Evaluating the Argument: AQ7

• This is an argument that something started it all off – there is an uncaused cause, and that is something to which ‘which everyone calls a God’. But we mostly agree with the first claim, though not the second!
  • Not because of anything about the metaphysics of causation (though that plays a part), but because there is good scientific evidence that the universe began in a Big Bang, a physical event with no prior cause (there was nothing prior to it), which begins the causal chain of every event now existing, but which was neither an intelligent person itself, nor created by such a person.
• If this is right, Aquinas’ argument is not an argument for God, since we have good reason to think that the actual first cause was not a supernatural person, and hence was just not the kind of thing deserving of the name ‘God’.
  • This premise is not part of a persuasive argument – only those already convinced of God will be inclined to grant that God is to be identified with the first cause.

Are causes in a chain? Evaluating AQ1

• Aquinas seems to have the picture that every event can be fitted into a single causal chain.

  • But that doesn’t seem right: many events \(A\) and \(B\) are such that neither is causally prior to the other.

• So maybe there are many actual causal chains – and possibly many first causes? To get his desired conclusion we need a substantial additional premise:

  (AQAdd)

  Any two actual causal chains share some initial segment in common.

• But rather than thinking of the causes Aquinas is talking about as ordinary localised events, let us take them to be maximal simultaneous states.

• Since time is linearly ordered, it is true that these world states come ordered in series, and plausible that each is causally dependent on those earlier than it.

  • Setting aside relativistic issues mostly (Eagle 2021)!

Causal loops – AQ3

• We can prove AQ3. Suppose there are finitely many events \(A_{1},…,A_{n}\), each of which has a cause.
• Start with \(A_{1}\), and then successively put the remaining events in their place in the chain until all are placed.
• To ensure that all have a cause, we will have either done the following:
  1. Arranged them all in a circle.
  2. Arranged at least some of them in a loop.
  3. Made at least one of them its own cause.
• We don’t have enough events to ensure that each event has a distinct cause that it is not causally prior to.
  • If we had infinitely many, we could do it: we could have an infinite descending chain of causes, and we can always draw a ‘new’ causal from our infinite supply to be the cause of any event in the chain.

No causal loops – AQ4

• Aquinas simply states that it is ‘not possible’ that something precede itself.
• Either this is because he is simply ruling out causal loops by fiat, or because he thinks that causal precedence entails temporal precedence, and he thinks that nothing could be earlier than itself.
  • At least, no event can be earlier than itself.
• Are there really no causal loops? What about time travel?
  • In some scenarios compatible with general relativity, there are what are known as closed timelike curves (Arntzenius and Maudlin 2013). These are trajectories through space and time which, though they continually move ‘forward’ in time from moment to moment, globally end up closing back on themselves.
  • In such scenarios, we can have physical objects interacting with earlier ‘stages’ of themselves; those interaction events are part of their own causal history, violating Aquinas’ principle.

Infinite Causal Chains AQ2

• The bulk of Aquinas’ text is taken up by what he seems to regard as an argument that a series of causes cannot ‘go on to infinity’.
• The argument relies on the claim that ‘when a cause is removed, its effect is removed. Therefore, if there were no first among the efficient causes, then neither would there be a last or an intermediate’.
• He seems to think that an causal chain without a stopping place is a finite causal chain which is lacking its first cause – and he rightly points out that there would be no chain at all in that case.
• An infinite causal chain also lacks a stopping place (because for each potential stopping place there is another event precedes it), but not because we’ve taken it away – it’s because we’ve put more causes in. And this doesn’t seem to be susceptible to Aquinas’ argument.
• As Oppy and Scott put it, ‘it is quite clear that the argument that Aquinas actually gives against the possibility of an infinite regress of causes is simply question-begging’ (Oppy and Scott 2010: 85).

Identifying God(s)

• We’ve said some negative things about both Aquinas’ and Anselm’s arguments.
• But even if we accepted them both, and agreed with their conclusion, we would need something further: some argument that the greatest conceivable being and the first cause of our universe are one and the same thing.
• It seems perfectly conceivable that they are not – specifically, that there could be a first cause/creator who was pretty great, but not the greatest conceivable being.
  • Surely the greatest conceivable God would have made a better world than this! (We’ll return to this point when we discuss the argument from evil in lecture 4.)
• And moreover, it is not obvious that either is the same as the God of the Bible, who seems neither so abstract as to be a first cause, nor so perfect as to be the greatest conceivable being.

Another Ontological Argument

The Modal Ontological Argument

1. It is of the essence of God that He is perfect. (Premise, definition of God from PBT)
2. It is more perfect to exist necessarily, than to exist contingently. (Premise)
3. It is of the essence of God that He exists necessarily. (from 1, 2)
4. It is possible that God exists. (Premise)
5. It is possible that God exists necessarily. (from 3, 4)
6. Whatever is possibly necessary is actual. (Premise)
7. God exists. (from 5, 6)

• This is broadly inspired by the modal ontological argument devised by Plantinga (1974).

Evaluating the modal argument: God as a necessary being

• The premises (1) and (2) are broadly familiar from Anselm’s argument; issues we’ve already raised about PBT or controversies over the correct conception of ‘perfection’ will recur in essentially the same form.
  • The phrasing in terms of essence is new; it serves to suggest that necessary existence is part of God’s nature, part of what it takes to be God. ‘It is of the essence of X to be F’ doesn’t entail that X exists; only that in any situation in which X exists, X is F in that situation.
• In any case, the role of these premises is only to justify the intermediate conclusion (3) that God is to be understood as a necessary being.
  • In Plantinga’s original form of the argument he begins with the premise that ‘God’ to be defined as being of ‘maximal greatness’, where to have maximal greatness is to be a necessary being and to have greatness-making qualities of omnipotence, omniscience, etc., essentially – i.e., in every situation in which a maximally great being exists, it has those features.
  • Whether we argue for (3) or accept it as true by definition, as Plantinga wants, the distinctive novelty of the argument comes in its second half.

Evaluating the modal argument: possible necessity

• Premise (6) says that what is possibly necessary is actual, so the possibility of God’s necessary existence (5) would entails the actuality of God’s existence (7).
• Perhaps surprisingly, (6) is an orthodox principle of modal logic, the logic of ‘necessarily’ and ‘possibly’.
• The basic idea in modal logic is that ‘possibly \(p\)’ means something like ‘there is a possible scenario in which \(p\)’; and ‘necessarily \(p\)’ means ‘every possible scenario is one in which \(p\)’.
• Importantly, in modal logic we can consider not just what is necessary and possible, but what would be necessary/possible from the perspective of another possible world.
  • So ‘necessarily \(p\)’ is true at some scenario \(W\) just in case at every scenario which is possible from the perspective of \(W\), \(p\) is true.
• This is what allows us to understand nested modalities like ‘\(p\) is possibly necessary’.

The Brouwerian intuition

• Premise (6) is the so-called Brouwerian axiom B of modal logic.
  • The intuition is this: suppose \(p\) is possibly necessary.
  • Then there is a scenario \(G\), which is possible from our perspective, in which ‘necessarily, \(p\)’ is true.
  • So from the perspective of \(G\), every possible scenario must be one in which \(p\).
  • Since \(G\) is possible relative to actuality, actuality should be possible relative to \(G\) – it would be weird if \(G\) was a way things could be, but according to \(G\), the way things are isn’t a way things could be!
  • But then the actual scenario is among the possibilities-relative-to-\(G\), and hence \(p\) is true at it.
• Kane (1984: 341–44) discusses B in the context of the modal ontological argument; I won’t question it, or (6).

Evaluating the modal argument: the possibility of God

• If we accept God’s nature involves his necessary existence, and we accept the B axiom, then ‘Possibly, God exists’ is true iff ‘Actually, God exists’ is true.
  • The proponent argues: it is possible that God exists (4), so God exists.
  • The opponent will argue: God doesn’t exist, so it is not possible that God exists.
• Is there anything that breaks this symmetry?
• The proponent will point to the long history of religious belief, that intelligent and thoughtful people down the ages have accepted the existence of God. Were these people not just mistaken, but so confused that they believed something impossible?

Epistemic possibility and genuine possibility

• The opponent replies: God is epistemically possible: for all we know there could be a God.
• But ‘epistemic possibility’ is not really a species of genuine possibility.
  • Consider some mathematical conjecture that turns out to be true. Perhaps some smart mathematicians thought it might turn out false. But they were wrong: in fact, it was later proved, which shows that it couldn’t have been false. We might say: it was epistemically possible that it be false, for all they knew; but in fact it had to be true.
    
• Crucially, the B axiom is not plausible for epistemic possibility.
  • If for all I know, \(p\) must be the case, that doesn’t entail that \(p\) is the case.
  • Suppose for all I know (which is very little), the butler must have done it; my ignorance doesn’t entail that the butler did it!
• So the opponent says the argument equivocates.
  • On the epistemic reading of ‘possibly’, (4) is plausible but (6) is false.
  • On the metaphysical reading of ‘possibly’, (6) is correct, but why should we think (4) is?

References