The Cosmological argument

“Why is there something rather than nothing?” – Leibniz

Cosmological arguments claim that God’s existence can be established as the required explanation of what exists.

The main ingredients of cosmological arguments are typically:

  • A causal principle. Some variety of a claim like ‘every event has a cause’.
  • A denial of the infinite regress. The claim that time does not go back forever but had a starting point.

The combination of these two ingredients is meant to show that the universe must have had a cause. There are then a variety of methods used to try and show that a God is the best candidate for what that cause is.

The typical version of the argument is a posteriori, beginning with observations of the world and then concluding that some first cause or necessary being is the only explanation of their origin.

Cosmological arguments can be broadly categorised into those that focus on causation and those that focus on contingency.

Cosmological arguments from causation

Aquinas’ 1st way from motion

Following Aristotle, by motion Aquinas means any kind of change. Something can only change if it has the potential to change. So, we can understand change as the actualisation of a potential to change in a certain way.

P1. We observe motion.
P2. Motion is the actualization of a thing’s potential to be in motion.
P3. A thing can only come to be in motion by being moved.
P4. A mover must be something that is actual.
P5. A thing cannot move itself.
C1. So, all things in motion must have been moved by something else.
P6. If there were no first mover, there would be no motion now.
C2. Therefore, there must be a first mover which must itself be unmoved (pure actuality). That thing we call God.

Aquinas’ 2nd way from atemporal causation 

P1. We observe efficient causation.
P2. Nothing can cause itself.
P3. There is a logical order to sustaining causes: the first cause, then intermediate causes, then an ultimate effect.
P4. If A is the efficient cause of B, then if A doesn’t exist neither does B.
C1. There must be a first sustaining cause, otherwise P1 would be false as there would be no further sustaining causes or effects.
C2. As there is a first cause, there cannot be an infinite regress of causes.
C3. The first cause must itself be uncaused. That thing we call God.

The type of causation involved in the first two ways.

Aquinas employs Aristotelian efficient causation, which seeks the explanation for how a thing came into being. There are two types of efficient causation: sustaining and temporal. Aquinas’ claim that without a first mover/cause there could be no further motion or causation looks puzzling until we understand that his arguments involve sustaining rather than temporal causation.

Temporal (‘per accidens’ / ‘in fieri’ ) causation: A ‘horizontal’ sequence where effects are brought about at certain points in time. The continued existence of the effect does not depend on the continued existence of the cause. Each member of a temporal sequence has its own independent causal power to cause further effects. The causal chain can continue to exist and generate more effects regardless of whether the prior causes still exist.

Aquinas’ example: a father can create a son independently of his own father, who can then independently create their own son etc.

Sustaining (‘per se’ / ‘in esse’) causation: a ‘vertical’ hierarchy where effects are brought about continually by the sustained activity of higher causes. A first cause creates intermediate causes and then ultimate effects. Only the first cause has primary intrinsic causal power. The other members have merely secondary causal power because it is derived from the first cause. A sustaining series cannot continue to exist without the causal activity of the first cause.

Aquinas’ example: a hand (primary first cause) moves a stick (secondary intermediate cause) which moves a stone (effect). A stick cannot move a stone by itself, it can only derive that causal power from the hand. The causal activity of the hand and stick and their effect are all happening at once. The first cause is not ‘temporally prior’ to its intermediate causes and effects, they are all simultaneous. Nonetheless, the hand is clearly higher than the stick in a causal hierarchy because the stick gets its causal power from the hand. Without the hand, the other causes and effects would not happen.

This is why sustaining causation is ‘atemporal’. It cannot be properly understood as temporal events sequenced in time. It involves ontological relationships of dependence, where all except the highest member of a series depend on higher members.

The hand analogy cannot be perfect because ultimately there can only be one first sustaining cause of all things; God. The first and second ways attempt to show a first mover or cause must exist. Not ‘first’ in time, but ontologically first in the sense that all subsequent motion and causation ontologically depend on it.

“when Aquinas talks about an “order” of efficient causes he is not thinking of a series stretching back into the past, but of a hierarchy of causes, in which a subordinate member is here and now dependent on the causal activity of a higher member.” – Copleston

Aquinas’ & the infinite regress

Aquinas actually thought it was possible for a temporal series to have no first cause. His argument is that a sustaining series must have a primary cause. Members of a temporal series have independent causal power. Temporal causal events need not trace their explanation back to a first cause. So, a temporal series does not need a first cause and could be infinite.

Edward Feser explains that for a sustaining series, the number of members is irrelevant. It could actually be infinite, or loop around on itself. Regardless, a primary cause would still be required to explain the causal power in the series. It cannot be explained by its other members, since they are secondary causes. Secondary causes (like sticks) cannot move themselves. They can be moved by other secondary causes, but those also require an explanation.

There must be a primary cause to explain the causal power that exists in a sustaining series, regardless of how long it is, even if it’s infinite. What’s really impossible for Aquinas is an infinite regress of explanation for the causal power of secondary causes. There cannot be an infinite derivation of the causal power we observe in secondary causes.

There must be an originator of the causal power which defines a per se / in esse series of sustaining causation. Otherwise, there would be no motion or causation in the series, i.e., it would not exist. That’s what Aquinas means when he says without a first cause there wouldn’t be any further causation or motion.

Edward Feser explains:

“as long as the members of such a circular or infinite chain of causes have no independent causal power of their own, there will have to be something outside the series which imparts to them their causal efficacy. (As the Thomist A. D. Sertillanges once put it, a paint brush can’t move itself even if it has a very long handle. And it still couldn’t move itself even if it had an infinitely long handle.) Moreover, if that which imparts causal power to the members of the circular or infinitely long series itself had no independent causal power, then it too would of necessity also require a principal cause of its own, relative to which it is an instrument. This explanatory regress cannot possibly terminate in anything other than something which has absolutely independent causal power, which can cause or “actualize” without itself having to be actualized in any way, and only what is purely actual can fit the bill.” – Feser

F. Copleston explains:

“unless there is a “first” member, a mover which is not itself moved or a cause which does not itself depend on the causal activity of a higher cause, it is not possible to explain the “motion” or the causal activity of the lowest member. His point of view is this. Suppress the first unmoved mover and there is no motion or change here and now. Suppress the first efficient cause and there is no causal activity here and now. If therefore we find that some things in the world are changed, there must be a first unmoved mover. And if there are efficient causes in the world, there must be a first efficient, and completely non-dependent cause. The word “first” does not mean first in the temporal order, but supreme or first in the ontological order.” Copleston

The Kalam cosmological argument from temporal causation

The Kalam cosmological argument focuses on the coming into being of the universe. This shows the argument explicitly involves temporal causation. Causation is conceived in a horizontal sequence of temporal events. Effects then can continue independently of the causal activity of the temporally first cause. W. L. Craig brought this argument to prominence in the late 20th century and named it ‘Kalam’ after the Islamic philosophy which first invented it in the 11th century.

P1. Everything that begins to exist has a cause of its existence.
P2. The universe began to exist (an infinite regress is not possible).
C1. So, the universe has a cause of its existence.

Further inferences are made to show that the cause of the universe is God. Craig firstly argues that scientific explanation applies within the universe and therefore cannot apply to its actual creation. The only other option is that it would have a personal explanation, i.e., intentionally created by an intelligent mind.

This being must have the power to create a universe from nothing (ex nihilo), which means it is omnipotent. A being which could create something from nothing it seems could do anything. It cannot be a temporal or extended being, since time and space didn’t exist until it created the universe. It’s impossible for it to be spatially or temporally bounded if space and time did not exist without the universe. A timeless is eternal, meaning it didn’t begin to exist. So, it doesn’t contradict P1 to say it doesn’t have a cause. Ultimately, these are qualities that God would have, so the cause of the universe is God.

The first premise is justified by the causal principle, which Craig claims is based on the metaphysical principle that something cannot come from nothing.

The Im/possibility of an infinite regress

Craig’s second premise is justified through a priori arguments against an infinite regress.

Craig rejects the possibility of an ‘actual infinite’, the idea that infinity could exist in reality. The issue is that sets with infinite members can paradoxically be equal in size to their subsets. This might make sense theoretically, but Craig claims problems arise when applying it to reality. He illustrates with an infinite library that has an infinite number of books, half of which are green. The green books are half of the total, so they are less than the total. But in mathematics, half infinity is still infinity. So, the green books are both less than and the same size as the total number of books. That is absurd, so infinity cannot exist in reality.

Craig also argues you cannot traverse an infinite through successive addition. We could never have gotten to this moment, if an infinite number of moments had to pass to get to it.

Craig provides a posteriori reasoning against the infinite regress. The evidence from modern cosmology suggests the universe had a beginning due to the evidence for the big bang theory.

Counter: the possibility of an infinite regress

Critics of the cosmological argument attack it by arguing there could be an infinite regress and therefore no first cause. The universe could simply have always existed in some form. God’s existence cannot be appealed to as the explanation of the origin of what exists if there was no origin.

Logical possibility. Hume thinks an infinite regress cannot be ruled out a priori, since a finite regress can be denied without contradiction. The concept of ‘time’ does not seem contradicted by ‘infinite’. An infinite regress may be difficult to imagine, but there is no obvious logical contradiction in it.

Craig argues that in an infinite library with half its books being red, a subset (red books) would be both smaller than and equal to its set (all the books). For Craig that proves that an ‘actual infinite’ cannot exist in reality. However, mathematician G. Cantor that for infinite sets it’s not absurd, it’s actually a defining characteristic of them. They are radically different to finite sets in their mathematical properties. When we think of libraries or hotels, we have in mind our ideas about finite sets of things, but Cantor argued such intuitions are not applicable to infinite sets. Infinite sets simply have different mathematical properties, one of which is the possibility of a one-to-one relation between the number of members of infinite sets and their subsets. So, Craig’s library or Hilbert’s Hotel are not absurd.

Metaphysical possibility. Then there is the question of metaphysical impossibility. Hume thinks we simply lack the evidence to judge here.

Support for Hume over Craig can be found from scientific theories of the origin of the universe which involve the possibility of an infinite regress.

One is that the universe eternally cycles between expansion and collapse, where a new timeline begins each cycle. In that case, an infinite amount of time never passes and in fact a timeline containing an actual infinite never existed. Yet, an infinite series of cycles existed, just not on any one timeline.

Another possibility is Alan Guth’s inflation theory, which proposes that there is eternally existing quantum energy that can sometimes create a universe like ours.

Like all scientific theories, they could turn out to be wrong. The point to take from them is that physicists take seriously the possibility of an infinite series. Russell points out that scientists being able to conceive of something indicates that it is possible.

So, there is philosophical and scientific disagreement about the possibility of an infinite regress. This means we cannot rule it out, which undermines the cosmological argument since it is premised on ruling it out.

Evaluation defending the cosmological argument:

Aquinas & Leibniz’ versions of the argument are not defeated by the possibility of an infinite temporal series. They say that such a series would still either:

  • Be contingent and thus still require a God to create it (Leibniz & Aquinas’ 3rd way)
  • Contain merely secondary causes and thus require a primary cause (Aquinas’ 1st & 2nd ways)

Aquinas & Leibniz accept the possibility of an infinite regress of beings and/or time. They argue that what’s really impossible is an infinite regress of explanation. Scientific hypothesis about the possibility of the universe stretching back forever does not undermine their argument.  

Evaluation criticizing the cosmological argument:

Hume’s stance is successful because it takes the right stance on the relationship between philosophy and science.

Philosophical analysis is useful to the degree a concept is fully understood. The issue is, we know very little about what time actually is scientifically. So, Aquinas, Leibniz and Craig’s philosophical reasoning about time and infinity is limited. Their claims about the causal principle and PSR are limited for the same reason.

The history of science has proven many metaphysical intuitions false, such as the discovery that space is non-Euclidian or the apparent paradoxes that arise from quantum mechanics.

Science has shown reality to be far stranger than past philosophers thought possible.

Cosmological arguments therefore fail as they rely on the philosophical assumption of the impossibility of an infinite regress which is really a scientific question.

“There are more things in heaven and earth … than are dreamt of in your philosophy” – Shakespeare’s Hamlet

Debates over the causal principle

A strength of cosmological arguments is their support from the causal principle, which is the idea that every event must have a cause. It seems impossible for an event to happen without a cause, otherwise something could come from nothing, which is absurd. Ex nihilo nihil fit (nothing comes from nothing) is an idea going back to ancient Greek Philosophers like Parmenides. So, everything that begins to exist and every contingent being must have a cause. William Lane Craig claims that the causal principle is:

“based on the metaphysical intuition that something cannot come out of nothing.” – W. L. Craig

Counter: Hume’s objection to the causal principle.

According to Hume’s fork, propositions such as the causal principle can either be analytic or synthetic.

Hume argues that the causal principle cannot be analytic, because it can be denied without contradiction. Logically absurd ideas are things like four-sided triangles. A triangle is clearly contradicted by ‘four sides’. Contrastingly, the idea of an event (something that happens) doesn’t seem contradicted by the idea of ‘no cause’. So, the causal principle doesn’t seem supportable by a priori reasoning like Craig had thought. So, the universe or any of the events or secondary causes we experience could have no mover/cause/explanation.

The other option is that the causal principle is synthetic and justified a posteriori. Philosophers like Swinburne and even Leibniz argue this approach could work. Every event we have experienced has a cause. So, the evidence supports the causal principle. Not as a necessary truth, but as an empirical inference.

However, Hume still objects that we don’t know whether the current state of the universe is anything like its origin. So, we cannot infer from observations made within the universe to the nature of its origin. To have valid evidence would require observing the origin itself, or the origin of other universes, which we currently can’t and may never do.

So, we have no good reason, either a priori or a posteriori, to believe the universe had a cause.

Evaluation defending the cosmological argument:

However, this criticism is unsuccessful because it fails to attack Aquinas’ arguments.

Hume’s critique works better against the Kalam cosmological argument, which does claim the whole universe, like all things that begin to exist, has a cause. Aquinas doesn’t claim that. He only claims that secondary causes we observe must have a primary cause.

The idea of ‘secondary cause’ is contradicted by the idea of ‘no cause’. This is because by definition, a secondary cause is derived from a primary cause.

You couldn’t even imagine a secondary cause come from nothing because then it wouldn’t have the causal power to be what it is; a secondary cause.

So, Aquinas’ argument is successful because it focuses on the secondary causation of everyday things we observe around us. It doesn’t rely on being able to prove anything general about the universe.

Evaluation criticizing the cosmological argument:

The success of Hume’s sceptical stance is borne out by modern science.

Russell points out that in quantum mechanics there are events with no cause This could be true of the origin of the universe too.

Copleston tries to respond that only some interpretations of quantum mechanics propose uncaused events, which could be wrong. But this misses Russell’s point. Scientists being able to conceive of uncaused events shows that they are possible.

Cosmic inflation theory claims the universe has zero total energy because the positive energy of matter equals the negative energy of gravity. This means the universe required no energy to be created. Alan Guth, one of the theory’s developers, said the universe is the ‘ultimate free lunch’.

Physicist Lawrence Krauss says inflation theory shows how the universe can come from nothing. He claims this answers Leibniz’ question of why is there something rather than nothing, without needing God.

Inflation theory could turn out false, but it’s scientific possibility shows Hume took the right position. The origin of the universe could be so different to anything we experience within it. Neither a priori nor a posteriori reasoning can justify believing that the universe needed a cause.

So, perhaps the universe or the causal power we observe in either temporal or sustaining causes, might all have no ultimate explanation, since such a thing is possible.

Cosmological arguments from contingency

Beings have causal relations, but also an ontological status, i.e., contingency or necessity. Contingency arguments claim that contingent beings & series must have an external explanation since it is their nature to depend on something else for their existence.

A strength of cosmological arguments from contingency is that their conclusions achieve more than arguments from causation. They can establish God’s necessity, meaning inability to cease existing, which is a key element of Christian theology. Aquinas understands necessity to mean the inability to cease existing, which fits with the concept of omnipotence.

Another strength is their seeking of an ultimate explanation rather than only a first cause. This focus more fundamentally on the nature of things.

Aquinas’ 3rd way (contingency)

P1. We observe that there are contingent beings (things that can possibly not exist).
P2. If it is possible for something to not exist, then there is some time in which it doesn’t exist.
C1. If everything were contingent, then at one time nothing existed.
P2. If nothing once existed, nothing could begin to exist, so nothing would exist now.
C2. So, there must be something that is not contingent, “having of itself its own necessity … That thing we call God.”

Philosophers have puzzled over the truth of P2 and the inference to C1. It’s not obvious why everything being contingent means that at one time nothing existed. On the contrary, it seems possible for there to be an infinite series of contingent beings, each giving rise to the next, none of which always existed. Then, everything would be contingent, yet it wouldn’t follow that nothing once existed.

However, this presupposes the possibility of an infinite regress of dependence and explanation, which Aquinas has already given reasons for doubting in his first and second ways and elsewhere in his writings. Consequently, many philosophers interpret Aquinas to have meant the following:

If everything were contingent and there couldn’t be an infinite regress of contingent beings, then a contingent series must have had an ultimate beginning before which would be nothing. From there the argument can proceed that since nothing comes from nothing, yet there is something now, we can conclude that it cannot be the case that everything is contingent.

Leibniz’ principle of sufficient reason

Leibniz’ argument is a priori in the pre-Kantian sense of inferring from causes to effects. It doesn’t begin with inference from experience of effects and infer to a cause.

The argument is based on the principle of sufficient reason:

“we can find no true or existent fact, no true assertion, without there being a sufficient reason why it is thus and not otherwise, although most of the time these reasons cannot be known to us” – Leibniz

This is the first premise. It does both the job of a causal principle and an argument about the infinite regress. This strengthens the argument by making it dependent on only one claim.

“we cannot find in any of the individual things, or even in the entire collection and series of things, a sufficient reason for why they exist” – Leibniz

P1. For every true fact or assertion, there is a “sufficient reason why it is thus and not otherwise.” (PSR)
P2. It is a contingent fact that a series of contingent beings exists
C1. So, the contingent series that exists must have a sufficient explanation.
P3. The sufficient reason for contingent beings cannot be found in other contingent beings
C2. So, it must be found in a necessary being, which is God.
C3. So, God exists

Leibniz claims that the principle of sufficient reason (P1) can be known as a necessary truth. Even if we can’t know or even find out what the reason for a true fact is, there must be one.

Leibniz takes a similar approach to the infinite regress to Aquinas. He accepts it is possible, but still says a God would be required to explain any series of contingent beings, even an infinite one.

This is because an infinite series without a necessary being would have no sufficient reason for its existence. Each being’s reason for existence would consist something for which its reason for existence also consists in something else. There would simply be an infinite deferring of the reason for existence and thus there would not actually be a reason for the existence of the series.

So, an infinite series could exist, but not by itself. It must have been created by a necessary being which exists outside of the series.

Leibniz illustrates with a geometry book:

“Let us suppose that a book on the elements of geometry has always existed, one copy always made from another. It is obvious that although we can explain a present copy of the book from the previous book … this will never lead us to a complete explanation, no matter how many books back we go, since we can always wonder why there have always been such books” – Leibniz

Leibniz argues the same is true of the world. Even if the world had always existed, we can still ask why it has always existed. There must be a sufficient explanation for every true fact, such as a universe existing, even if it’s infinite. That reason cannot be found in any of the contingent beings themselves. This is because since they are contingent, they require an explanation. The chain of explanation can only terminate and be sufficient with a being that contains its own reason for existence. A necessary being.

The fallacy of composition

“I can illustrate what seems to me your fallacy. Every man who exists has a mother, and it seems to me your argument is that therefore the human race must have a mother, but obviously the human race hasn’t a mother — that’s a different logical sphere.” – Russell

The fallacy of composition is the assumption that what’s true of a thing’s parts is also true of the whole. Experience shows that parts of the universe have an explanation (because they are contingent or have a cause/mover/explanation). To infer that means the universe as a whole must have an explanation commits the fallacy of composition.

However, Edward Feser points out that Aquinas is only making a claim about the beings we experience, not the whole universe.

Also, although cosmological arguments often start with observations of the parts, Copleston points out it’s not obvious they infer from them to a whole. Even less so with Leibniz’ argument, which starts with the a priori principle of sufficient reason.

Leibniz & Aquinas argue a series must have an explanation, not because its parts do, but because all contingent things or true facts must have an explanation. They rely on the impossibility of an infinite regress of explanation. Any series we observe, even if it’s infinite, requires an explanation in a primary cause (1st & 2nd way) or necessary being (3rd way & Leibniz).

To deal with this, Hume & Russell deepen the objection.

A series is just a mental construction. We observe individual causes or beings and group them together in our mind. A series doesn’t need its own explanation because it isn’t anything over and above a mentally grouped set of sufficiently explained things.

Hume illustrates with 20 particles, a finite collection of contingent beings. Each part has an explanation, but the collection is not itself a thing which needs an explanation. Hume argues the same is true of an infinite series of contingent beings. Each being has an explanation in the being it depends on. The idea that the ‘whole series’ needs an explanation is fallacious because a series is nothing over and above its parts, which each have an explanation.

“in such a chain or series of items, each part is caused by the part that preceded it, and causes the one that follows. So where is the difficulty? But the whole needs a cause! you say. I answer that the uniting of these parts into a whole, like the uniting of several distinct counties into one kingdom, or several distinct members into one organic body, is performed merely by an arbitrary act of the mind and has no influence on the nature of things” – Hume

Russell agrees, noting that our concept of a cause comes from experience of individual things. We don’t have a basis for applying it to a series of things, as if a series was a thing itself that somehow needs its own cause.

“I think the word “universe” is a handy word in some connections, but I don’t think it stands for anything that has a meaning … The whole concept of cause is one we derive from our observation of particular things; I see no reason whatsoever to suppose that the total has any cause whatsoever … it doesn’t need to be its own cause, what I’m saying is that the concept of cause is not applicable to the total.” – Russell.

The cosmological argument takes a concept we gain from the parts and fallaciously assumes it applies to the whole.

Russell concludes the universe is “just there, and that’s all”. It is neither necessary nor contingent, since it neither contains its own explanation nor depends on anything else. This makes it a brute fact. We have no basis for thinking the concept of explanation is even applicable to it, whether causal explanation or otherwise. An infinite regress of explanation may be impossible, but we don’t have a reason for applying the notion of explanation to the universe.

So, for Hume, a series is a mental construction and thus requires no explanation. The explanation of its parts is sufficient. This applies regardless of whether the series only involves what we observe, is identified with the whole universe, or is infinite.

Counter: Aquinas’ arguments avoid this issue

However, arguably this can only work as a critique of the cosmological argument regarding a temporal series, not a sustaining series. A temporal series is a group of events whose causal power can operate independently of one another. Grouping independent parts into a whole series does seem like a mental construction. Explaining each independent part explains everything without need for a God.

A sustaining series seems different because the members are not causally independent. Every secondary cause depends for its causal power on the continuous operation of any higher secondary causes and/or a primary cause. So, the grouping of members in a sustaining series is not mentally constructed but refers to objective causal relations each member has with every other member.

Aquinas’ argument is that the parts of a sustaining series can only be sufficiently explained by a primary cause. Even an infinite number of secondary causes must have been created by a primary cause.

Evaluation defending the cosmological argument:

Aquinas seems to present the strongest form of the cosmological argument due to his use of sustaining causation. It shows a series  must be explained by a more than the type of parts we directly observe.

Furthermore, shows that the concept of a ‘brute fact’ is self-defeating. Science and philosophy are about finding out why things are the way they are; their causes and explanations. If we say that there is no reason, then we undermine the purpose of science and philosophy itself.

Feser goes further, arguing the concept of a brute fact leads to radical scepticism because all our beliefs could be caused by brute facts, including their not seeming so to us.

The requirement for explanation behind the cosmological argument therefore seems defensible as the very assumption behind asking and answering any scientific or philosophical questions at all.

Evaluation criticizing the cosmological argument:

This defense of Aquinas is unsuccessful because sustaining causation faces its own issues. It’s not clear that it actually exists. All the effects we observe could be the result of temporal causes.

Aquinas believed the existence of the universe and all causes we observe in it are sustained by a primary cause, God. However, firstly there’s no obvious reason to think the universe’s existence needs sustaining in that way.

Secondly, there are cases like Aquinas’ hand example. But they need not be explained by a hierarchy of sustaining causation. It could be explained temporally, through each moment involving a simultaneous transition of kinetic energy between different objects at multiple points in space.

Aquinas’ use of sustaining causation avoids many of the standard critiques of the cosmological argument, but it assumes sustaining causation actually occurs. Objects we observe could simply be part of an infinite temporal series which isn’t an entity over and above its parts, and therefore needs no explanation. So even Aquinas’ versions are based on assumption.

The Im/possibility of a ‘necessary being’

Hume argues we have independent reasons to reject the concept of a ‘necessary being’. This targets Aquinas’ 3rd way, Leibniz and Descartes’ cosmological arguments. It would also limit the sort of God inferable by any argument, if a necessary being is incoherent.

“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

No being’s non-existence is an inconceivable contradiction. So, no being’s existence can be logically necessary. 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 as non-existent. So, our mind is incapable of giving meaning to the idea of a being existing necessarily.

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

On Hume’s empiricism, we gain our ideas from experience. It is then only through imagination that we can apply modal concepts like possibility and necessity to those ideas. Through acts of imagination, we can add ideas of beings into our conception of reality and subtract them. The idea of a necessary being is one we are unable to subtract or remove from our imagined concept of reality. Hume thinks we can’t make sense of such an idea. Anything we can project onto our conception of reality, we can also take away or simply not project. Therefore, we cannot coherently understand any being to be logically necessary.

“All existential propositions are synthetic” – Hume

Hume’s argument is epistemological, i.e., focused on what we can know. Arguably Hume isn’t technically saying a necessary being is impossible, just that it is impossible for us to conceive of or therefore know that such a being exists. The imagination is our only tool for knowledge of what is possible. This means the limits of our imagination are the limits of our ability to think about what is possible or necessary. There is no being we are unable to imagine not existing. So, we cannot form the concept of a necessary being. It is inevitably meaningless to us.

Counter: Hume’s critique misses

Saul Kripke introduced the concept of ‘metaphysical necessity’. This occurs when it’s not possible that the world could have been otherwise regarding a thing existing due to the kind of thing it is. Cosmological arguments only need to claim God’s existence is metaphysically necessary, not logically necessary.

The only explanation of causation or contingency is a being with a unique metaphysical status. One that can be an explanatory terminus by containing its own explanation for existence, a necessary being. God’s existence is argued necessary for reasons that are metaphysical, not purely logical. God is the metaphysically necessary explanation of the causation and contingency we observe in the world.

We could imagine no God, though then to be consistent we’d also have to imagine no world existing either. So, we could accept Hume’s claim without any damage to the cosmological argument. It doesn’t need to claim that an inconceivably non-existent being exists, just that if a world like ours exists, then we have metaphysical reasons for thinking a necessary being must have created it and therefore exists.

Evaluation defending the cosmological argument:

Hume’s initial objection is unsuccessful because on closer analysis, it attacks a claim the cosmological argument doesn’t make, that God’s existence is logically necessary.

There are even some religious philosophers, like Swinburne and Hick, who believe God is not logically necessary. However, they still defend the cosmological argument as giving us a reason to believe that God exists.

For Hick, God is a being we can’t imagine ceasing to exist if it existed. He calls that an ontologically necessary being. Hume appears to have no way to disregard that sort of being.

So, the cosmological argument doesn’t strictly rely on God’s logical necessity. Hume’s critique of that concept is irrelevant.

Evaluation criticizing the cosmological argument:

However, Hume develops his argument which protects it against even these modern defences.

Suppose we were to grant meaning to the concept of a necessary being. It could simply be the universe itself or some sort of matter that is necessary, rather than a God.

This resonates with the ideas from modern cosmology about the possibility of eternal quantum energy which sometimes creates universes.

The point is not that we should believe that is the explanation. The point is that the cosmological argument cannot provide a reason to think the necessary being must be mind, let alone God, rather than some form of necessary matter. So, it cannot give us a reason to believe God exists.

Sources & further reading

Web pages

The Stanford encyclopedia is the best philosophy resource online. But it will be quite challenging for A level students:

https://plato.stanford.edu/ENTRIES/cosmological-argument/

Professer Edward Feser’s blog posts, mostly focusing on Aquinas’ versions of the arguments with some discussion about the argument and its objections more broadly:

https://edwardfeser.blogspot.com/2011/07/so-you-think-you-understand.html
https://edwardfeser.blogspot.com/2010/12/hume-cosmological-arguments-and-fallacy.html
https://edwardfeser.blogspot.com/2010/08/edwards-on-infinite-causal-series.html
https://edwardfeser.blogspot.com/2012/05/natural-theology-natural-science-and.html

The famous debate between Russell and Copleston which heavily features the Cosmological argument:
http://www.scandalon.co.uk/philosophy/cosmological_radio.htm

Copleston on Aquinas’ 5 ways – starts on page 80:
https://archive.org/details/TheExistenceOfGod_201702/page/n87/mode/2up

Hume’s ‘dialogies concerning natural religion’
https://web.archive.org/web/20180423030326id_/http://www.earlymoderntexts.com/assets/pdfs/hume1779.pdf
(Note that Hume’s critique of the ontological and cosmological arguments are a bit blurred together in ways modern commentators typically find problematic. They can be found on page 39 & 40).

Media

William Lane Craig & Physicist Sean Carroll debate “God and Cosmology”.

William Lane Craig discusses the Kalam cosmological argument.

Lawrence Krauss’ lecture “a universe from nothing”.(Very heavy on the physics/cosmology!).

Edward Feser on the argument from motion.

Books