The Case Against The Cosmological Argument
Thomas Ash

Particularly relevant to this essay is my other response to the arguments put forward for God's existence, 'The Case Against The Design Argument'.

The cosmological argument is one of the most popular ways of proving God's existence. It is also the best. It has been philosophically influential, and was famously propounded by St Thomas Aquinas in three Ways in his Summa Theologica. We know that the concept itself dates back to Plato and Aristotle's "unmoved mover", and possibly before. Clearly it has great intuitive power, and the benefit of starting from an undeniable fact - the existence of the universe - which seems hard to explain. It is thus a partly a posteriori argument, rather than one based on the purely a priori reasoning of the ontological argument, which few can manage to beleive is up to the job of proving God's existence. Ordinary people also use it all the time, although they probably do not call it the 'cosmological argument', whenever they ask, "Why was there the Big Bang?", or "Why is there anything?", in a leading way.

There are really two key forms of the argument, illustrated by these two questions. The first implies that there must have been a first cause in time. This set off the chain of events: for instance, causes and effects (as in Aquinas' Second Way), or motion (as in his First.) In all three ways, Aquinas rejects infinite regress of motion [1], etc. more or less out of hand ("Now in efficient causes it is not possible to go onto infinity...") so he concludes that there must have been a first cause: "a first mover, moved by no other; and this everyone understands to be God."

The second question is more sophisticated, and does not need to talk about time, and so applies even if there is an infinite regress. It depends on the contingency of the universe, and is hinted at in the Third Way. Here, Aquinas says that all matter in the universe is contingent - depends on something else, and can "not-be" - and thus that there must have been a time when nothing existed, and that material things must have been brought into existence by God, a necessary being external to the universe. This relies on the universe as simply being the totality of all matter - a reasonable view, though one which a believer in souls might wish to question.

Clearly, these arguments at first look very powerful. But, when closely examined, are they really successful? And do they 'prove God'?

Perhaps the most obvious criticism you can make is to ask just why Aquinas is so confident there cannot be infinite regress. To be sure, we find it very hard to comprehend the notion of actual infinity, and it is possible to come up with all sorts of paradoxes which make infinity look like an 'impossible number.' The Islamic 'Kalam' version of the cosmological argument - developed by Al-Kindi and Al-Ghazali before Aquinas was even alive - does just this. Ed Miller's modern version points out that you cannot add a new day onto what is already an actual infinity of days, and that the present would never have been reached if an infinite number of days had had to be completed before it. Similarly, William Craig claims that as any event would happen an infinite number of times in an infinite stretch of time, the tortoise would have won the race as many times as it was run but also as many times as the hare won (this is my example, not Craig's!) The implication is that infinity is somehow paradoxical.

The idea seems to be to show that you cannot 'fit' the concept of indivisible infinity into our universe of finite things, so it is not a valid concept inside our universe. But all the argument shows is that we have intuitive difficulties imagining infinity, which is, after all, to be expected of such an advanced concept. Mathematicians can cope quite well with an infinite sequence of integers; appealing to the incomprehensibility of infinity is what Richard Dawkins in The Selfish Gene calls an "argument to personal incredulity."

The word 'personal' exposes the key flaw in this line of reasoning - our limited human imaginations are a poor guide as to what properties the universe can have. It's just arrogant to say "I have difficulty imagining or explaining this thing, therefore no one ever could." But that is what the Kalam argument effectively does.

Besides, the universe as a whole is not an ordinary finite 'object', it is what finite objects constitute parts of; this means we have no reason to assume that the same principles apply. This can be said of both contingency of everyday material objects and their obeying ofthe laws of cause and effect. As Kant pointed out, we have only ever witnessed these properties within the universe. It would be going far beyond what we know to conclude that the universe itself has a cause or is contingent.

Nonetheless, this assumption has great intuitive power as a result of all our experience of what happens within the universe. As J.L. Mackie points out, it is difficult to imagine an infinite train of carriages in motion without an engine providing a driving force, just as God is said to provide a creating and sustaining force. This reflects the intuitive power of Leibniz's 'Principle of Sufficient Reason': there must be some fact (or facts - but I shall pass over this) which completely explains existence. Despite the intuitive appeal of thinking that if there were no such sufficient reason(s) for everything to depend on, there would be nothing, there are several problems with this. A cause or reason must by definition exist (even if it is just an abstract principle, it is the case), so it cannot be the cause of all existence. Also, Leibniz's Principle found its backing in the assumption that God must have created everything for a reason, so cannot be used to argue for God's existence without circularity, unless we can find another reason to accept it besides a gut feeling formed by experience of living inside the universe.

Peter van Inwagen has pointed out that if there is some sufficient explanation of every fact, then everything in the world is necessarily true. This would mean that the putative sufficient reason could only logically fit with the actual world, and van Inwagen thinks the only such fact could be a complete description of the actual world - which is clearly no sufficient reason. However, a theist could think that God could choose to create this world alone. This does not, though, fit with the traditional theistic conception of free will, or with Christian doctrine that God did not have to create any world.

One problem which affects the first (temporal) formulation of the cosmological argument specifically is that a cause of the universe requires that there must have been a period of time before the universe began. But the universe is meant to encompass all time and space. This is also a problem for the Protestant view of God as the eternal rather than timeless creator of the universe - if God were in time, how could he have created all time and space? It is, however, avoided by the more theologically sophisticated Catholic view of God as timeless, and by the second formulation, which depends on the presumed contingency of the universe.

Another important point to make is that Aquinas says that nothing can be infinite, and then goes on to say that God is infinite. He says that nothing is the cause of itself, then says that God is the cause of Himself. Admittedly, he does say that God is outside the universe. Perhaps this means he is not bound by the limitations which apply within it. Even so, this involves conceding that something can be infinite or indendent, undermining Aquinas' basis for arguing that the universe must have a cause. And we can question whether it makes sense to speak of something outside the universe. If not, the response I have offered to Aquinas is an appeal to mystery, or - less charitably - a dodge.

A more powerful case against the cosmological argument depends on the difficulty of seeing God as "a necessary being", the cause of Himself. As discussed in my examination of the ontological argument, this is a highly problematic notion. We don't know that necessary existence is a meaningful concept. And even if it is, why shouldn't the Universe or the Big Bang (the most basic, earliest thing we actually know about) itself be considered the first, necessary cause? Since we cannot begin to comprehend the notion of necessity, we have no reason to assume it can belong to God's but not to universes, or even 'bangs.'

Bertrand Russell famously suggested taking the existence of the universe as a "brute fact." This approach is often defended by appealing to 'Occam's Razor', the (widely accepted) principle first put forward by the monk William of Occam. Occam's Razor reminds us not to multiply entities unless they really help us explain our experience. At first sight, it might seem like positing God as a first cause of the universe does just this. But unless we can explain why He created the universe, and why He is necessary (something the ontological argument singularly fails to do in its concentration on the psychological), God helps explain nothing. It is best to follow Occam's Razor and not posit a mysterious, complex being outside the universe like God without good warrant. We should accept that the Big Bang is the first cause that we know about.

This is not to claim that the Big Bang is the last word; the first cause. That would be to misunderstand the logical force of the Razor - that if you believe in one entity more than other people, you are less likely to be right, so need a good reason for doing so. Putting God forward as the first cause, and then claiming he was always there, is just to treat Him as a "brute fact." This just shifts the problem to why He - rather than the universe - has no cause. It's never very satisfactory to call something a brute fact. But since we have no alternative for now, we had best make it the universe, which we at least know exists.

This effectively neuters the cosmological argument's intuitive force. But even if (and I hope I have shown that this really is a big if) it succeeds completely and shows the existence of some first cause of the universe, outside the universe, it does not in any way show the existence of God. It is surprising the number of people who use it as their proof for God without acknowledging that the first cause could be anything. If parsimony is a consideration, it is simpler to posit a simple, axiomatic principle as that cuase, instead of the complex entity that is God. In no way whatsoever must it be the God of the Bible and traditional religion. The most honest answer to give to the question of why the universe is here is: "I don't know." The truth is, religions have never known, either - merely claimed they do.


1. Aquinas says that whatever is in motion has to be set in motion by something else that is already in motion. Even if something has the potential to move, it cannot move itself. 'Motion' here is a broad concept, encompassing not only A to B movement but also change, like that of ice into water. This is why Anthony Kenny's point that humans move themselves is not valid - chemical energy from food is changed into movement by our bodies.

© 2001

About the author...
Tom Ash is the webmaster of Big Issue Ground and Atheist Ground. He studied philosophy at Oxford and Cambridge, where he was president of the University Atheist and Agnostic Society. He currently does non-profit work and freelance web development. See more information about Tom Ash.

Feedback on the site, or on this essay, is always welcome. If it's complimentary, you may also want to put it in the Guestbook, for all to see!


Essays on related topics...

Arguments For and Against God's Existence
Title Author
How serious an evidential problem is evil for theism? Thomas Ash * top quality content
How convincing is the 'fine-tuning' argument for God's existence? Thomas Ash * top quality content
The Case Against The Cosmological ArgumentThomas Ash
The Case Against The Design Argument Thomas Ash
The Problem Of EvilThomas Ash
Does It Make Sense To Believe In Miracles? Thomas Ash
Cosmological Kalamity Dan Barker [off site]
Theodicy (Against the Problem of Evil) Kenneth Cauthen
Arguments to Design Anthony Flew [off site]
Did Life Have an Intelligent Designer? Steve Kangas [off site]
The Gap In Theistic Arguments Michael Martin [off site]
Human Suffering and the Acceptance Of God Michael Martin [off site]
An Atheological Argument from Evil Natural Laws Quentin Smith [off site]
The Anthropic Coincidences, Evil and the Disconfirmation of Theism Quentin Smith [off site]
Atheism, Theism and Big Bang Cosmology Quentin Smith [off site]
A Big Bang Cosmological Argument For God's Nonexistence Quentin Smith [off site]
Can Everything Come To Be Without A Cause? Quentin Smith [off site]
Causation and the Logical Impossibility of a Divine Cause Quentin Smith [off site]
Internal and External Causal Explanations of the Universe Quentin Smith [off site]
A Natural Explanation of the Existence and Laws of Our Universe Quentin Smith [off site]
Quantum Cosmology's Implication of Atheism Quentin Smith [off site]
The Uncaused Beginning of the Universe Quentin Smith [off site]

Number of unique visits to Big Issue Ground so far...