This argument has become very popular, and mostly as a result of William Lane Craig’s defense of it in both academic and popular circles. The argument was originally formulated by a medieval Islamic philosopher and theologian called Al-Ghazali. The argument is very simple and its premises seem obviously true.
Everything that begins to exist has a cause.
The universe began to exist.
Therefore, the universe has a cause.[i]
The first premise is so obviously true it barely needs a defense. Everything that begins to exist has a cause, because anything that happens at all has a cause! Beginning to exist is something that happens, and so it should have a cause as well. It is striking that some atheist philosophers deny this premise, committing themselves to an absurdity. The premise has both very strong empirical and intuitive support. We know from experience that nothing whatsoever begins to exist or that nothing at all happens without a cause. We also have a strong intuition that this is the way it must be, comparable to our logical intuitions underlying, for example, the principle of non-contradiction. What atheists sometimes say is that this intuition does not serve us well when thinking about the universe as a whole, because everything within the universe, on which they claim the intuition is based, presupposes the conditions of the universe. But the universe does not necessarily have a wider context. But this objection, as we saw in the article on the Leibnizian cosmological argument, also rules out scientific cosmology. Scientific cosmology is also in the business of drawing conclusions of the universe as a whole based on basic intuitions like those underlying the first premise ( such as modus ponens and modus tollens).
Craig uses four arguments to defend the second premise: that the universe began to exist. Two are based on modern science and two are philosophical arguments. If the universe does not have a beginning, it must be infinite. So, firstly, Craig argues that an actual infinite cannot exist. He distinguishes between a potential infinite and an actual infinite. A potential infinite is a series of numbers that are indefinite, while an actual infinite is a collection that goes on forever. Things can be potentially infinite but not actually infinite. “You can keep on dividing parts in half forever, but you will never arrive at an actual “infinitieth” division or come up with an actually infinite number of parts.”[ii] The objections and counter-objections to this sub-argument can get extraordinarily complicated, because they involve difficult mathematical concepts. Suffice it to say, for us lowly laymen, the idea of an actual infinite certainly seems intuitively absurd. To say that one has an infinite “collection” of things sounds like a contradiction in terms. A collection of things implies something static, with defined boundaries, but infinitude implies something that is always getting larger and does not have any boundaries. Infinity is dynamic, while a collection or a set of things is static. So an actual infinite implies both a defined boundary, but also a dynamic process of getting larger or more, which is a contradiction. The closest we can come to actual infinity is to say that something is indefinite.
Secondly, Craig argues that, even supposing an actual infinite is not logically impossible (or a contradiction in terms) it would be impossible to form an actual infinite in time. An actual infinite can only be formed (in time) by successively adding one thing after another. As Craig notes this idea depends on what is called the A-theory of time. The A-theory of time suggests that the past or the future is not real (only the present is real). Temporal becoming (that things come into existence and go out of existence as time progresses) is real. This seems like the obvious view. But, in philosophy, there are always those who deny what seems to be obvious. There is also the B-theory of time, which says that past and future is just as real as the present and that temporal becoming is an illusion. The B-theory, of course, implies that time is not a sequence as it is in the A-theory, so an infinity would not have to be formed by adding one moment after another. The A-theory, however, is just common sense. Nobody really believes that the past and the future is just as real as the present or even that the future and past is real at all. The easiest way to see this is that the future depends upon what happens now. If the future were as real as the present, it would not depend on what we do and how things develop in the present. In other words, the reality of the events in the future depends on the reality of the events in the present. I suppose you can say that this could be achieved through simultaneous causation, but the events of the future are not simultaneous with the events of the present (by definition). So an actual infinite must be formed by successive addition. But we know that an actual infinity cannot be formed by successive addition, because you will never reach “infinity” the number. You will always be able to add one more number, meaning that the infinity remains potential and not actual. Thus, if an actual infinity in time can only be formed by successive addition, but successive addition cannot form an infinity, then an actual infinite cannot exist in time.
Craig’s arguments from scientific evidence that the universe had a beginning appeal to the expansion of the universe and the second law of thermodynamics. Most basically the “standard model” of the expansion of the universe (the Friedman –Lemaitre model) implies that the universe had a beginning. Several other cosmological models have been proposed to reject this conclusion, but they fail and have not been accepted by most cosmologists. Second, the second law of thermodynamics implies that the universe will eventually reach equilibrium and suffer “heat death.” This raises a problem for an infinite past. If the universe had been infinite in the past, then we should already have reached heat death, because there is a clear time limit to how long the universe can exist before reaching heat death. But, because we have not reached heat death, the universe must have existed only for a finite time in the past.
Why is the explanation God?
If it is established that the universe does require a cause, why should that cause be God? Whatever is responsible for the universe, must transcend what the universe is made of, because whatever it is, it existed apart from the universe. Thus, it can’t be constituted by things or be dependent on things that only came into existence with the universe. That is, it must transcend matter, space and time. Thus, it must be immaterial, timeless and spaceless. It must also be a person. William Craig notes “the personhood of the cause of the universe is implied by its timelessness and immateriality. The only entities we know of which can possess such properties are either minds or abstract objects, like numbers. But abstract objects do not stand in causal relations…Numbers, for example, cannot cause anything.”[iii] Also, the existence of the universe cannot be accounted for by science, for at least two reasons. First, science always explains things naturalistically. However, the universe cannot be explained naturalistically since it represents the totality of natural reality. It would need to be explained with reference to itself, which would just be illogical. One needs to appeal to something other than the thing needing to be explained in order to explain it. Secondly, science explains phenomena by identifying regularities or laws that operate on conditions. However, since there is nothing outside of the universe (or multiverse, whichever you prefer), it cannot be explained through regularities and laws. Natural laws “organize” preexisting material. But those laws cannot account for how the universe came into being, because there is nothing outside the universe for it to organize. The only other type of explanation we are familiar with is personal explanation: explanation with reference to the intentions and actions of persons. Thus, the universe must have a personal explanation.[iv] Thus, the universe must have been caused by a timeless, spaceless, immaterial person.
Who Designed the Designer?
If the universe requires a cause, doesn’t God require a cause as well? The first premise, however, is “Whatever begins to exist has a cause.” Classical theism has always maintained that God does not begin to exist. Or at the very least, we could claim ignorance about whether God begins to exist or not, and this would mean that the objection does not succeed. In other words, we would need an argument that shows why God needs a beginning. As it stands, we have good reason to think that God does not need a beginning, because of the attributes of God implied above. If God is spaceless, timeless, and immaterial, then God exists outside of time and he also is not subject to the impermanence of physical things. This seems to imply at least that he does not require a beginning. Nothing can begin or end in a timeless reality, because you would need to identify a time at which something begins. Since there is no time, this is impossible. There is no temporal becoming in a timeless reality. So if God does not begin to exist, he does not need a cause.
Secondly, as William Craig has noted, this objection distorts the real nature of explanation. If an explanation is proposed, one does not immediately demand an explanation of the explanation. If you always needed to provide an explanation of every explanation you come up with, you would never explain anything at all. You would, in every instance, need to explain your explanation, and then come up with an explanation of that explanation of that explanation… ad infinitum.
Why ‘God’ and not ‘Gods’?
Why shouldn’t we think that many gods came together to design the universe? The simplest conclusion is that there is only one. If we are satisfied that the argument proves some sort of designer, there is no need to multiply entities beyond what is metaphysically required (i.e. Ockham’s Razor). The principle of ontological parsimony would counsel us that where the evidence is the same for either of the above propositions, we should opt for the simpler one: the one that postulates the fewest entities. I could similarly postulate that many gods came together and then they had a huge supernatural fight. The universe was spawned out of this celestial clash of powers. This theory explains the conclusions of cosmological and teleological arguments, but it is invalid as a conclusion precisely because we have no evidence for such a complex story (and it therefore contradicts ontological parsimony). It is therefore unreasonable to suggest that natural theologians have to rule out every logically possible scenario of creation. Similarly, scientists do not need to rule out every logically possible explanation of the evidence but can comfortably settle with the simplest one.
God’s infinity is invalidated by the arguments against an actual infinity
So, Craig’s formulation of the cosmological argument contends that an actual infinite not merely does not exist (in the physical universe) but that it cannot exist. But then, God can also not be infinite, as he is traditionally conceived. One can avoid one aspect of the objection by pointing out that there are two ways of conceiving of God’s infinity in philosophical theology: eternality and timelessness. If we suppose that God is timeless, then the objection would not be successful, because the second philosophical argument against an actual infinite only supposes that an actual infinite cannot be formed in time. An actual infinite can only be formed in time through successive addition. However, if God is not in time, then his infinitude would not need to be formed by successive addition. Secondly, the first philosophical argument against an actual infinite would also not apply because it would need to suppose that God’s infinity represents a set of specific things and we are constantly adding to its number. However, if time does not exist, or if God exists timelessly, there does not exist any temporal units that can be numbered in a set. You would also need to assume that God consists of a definite, discreet set of objects to whose number is always being added. But, of course, God need not consist of definite, discreet objects, since this would be a very physical way to conceive of God (when the argument shows he is immaterial).
Causation can’t apply to the universe as a whole because time does not exist apart from it
This objection supposes that the cause needs to precede the effect. While this may be true in some instances it certainly doesn’t need to be true in all. For example, think about a chandelier being held up by a chain leading to a ceiling. This is simultaneous causation: the cause does not precede the effect, but is simultaneous with the effect. You may suppose that simultaneity itself presupposes time. The chandelier and the ceiling may represent simultaneous causation because we are in time, but in a timeless reality it would simply be a dependence relation. Such a dependence relation is perfectly conceivable apart from time. A mere dependence relation does not presuppose time even though dependence may seem simultaneous in a world where there is time. So we can see that there is nothing in the concept of causation that requires time. Something bringing something else into being might normally happen in time, but that doesn’t mean causation is conceptually inseparable from time. Finally, to suppose that any and all time came into existence at the beginning of the universe may beg the question in favour of the universe being all there is. The physical space-time we’re used to may come into existence at the big bang, but perhaps God, even though existing timelessly, caused the existence of some sort of time prior to the universe.[v]
An actual infinite does not have to have a beginning.
J. L. Mackie objects to the cosmological argument based on the fact that a true infinite past would not have a beginning point. “It assumes that, even if past time were infinite, there would still have been a starting-point of time, but one infinitely remote, so that an actual infinity would have had to be traversed to reach the present from there. But to take the hypothesis of infinity seriously would be to suppose that there was no starting-point, not even an infinitely remote one, and that from any specific point in past time there is only a finite stretch that needs to be traversed to reach the present.”[vi] Thus, any particular point in the past will be a finite time from the present. It doesn’t seem that this really improves the problem. Even if there is no beginning, we still would need to traverse an infinite time to get to the present. We don’t need to identify a particular point in the past in order to show that it is impossible to traverse it if it is infinite. As Craig notes, “For the issue is how the whole series can be formed, not a finite portion of it.”[vii] In other words, the argument does not depend upon assuming a starting point. A lack of a starting point does not imply that an infinity can suddenly be formed by successive addition. It’s a red herring.
The world does not conform to our intellectual preferences
It is somewhat common for atheists to attempt to refute cosmological arguments by saying that our rational intuitions cannot be applied to the very macro or the very micro. Or, our rational intuitions cannot be applied to the conditions of the cosmos, because these intuitions developed in a world of medium-sized objects. For example, J.L. Mackie contends, “We have no right to assume that the universe will comply with our intellectual preferences.”[viii] It is astounding that he should say this however, since the scientific method and all our reasoning about the world, is ultimately based on intuitions like those that underlie the principle of sufficient reason (which he calls our “intellectual preferences”). Daniel Dennett and the popular atheist vlogger Thunderfoot, make similar points when considering the arguments of William Craig.
Firstly, this objection assumes a naturalistic understanding of human beings and is therefore circular. It assumes that only natural causes are responsible for our intellectual intuitions, which assumes naturalism in order to argue for it. Secondly, this objection has a fairly absurd implication. Scientific cosmologists probably use modus ponens, modus tollens and other basic syllogisms to make conclusions about the universe as a whole. But, if our intuitions about modus ponens and modus tollens are empirically constituted and only developed through natural causes, then these scientists are also doing what these atheists have said we cannot do. The intuitions of those scientists are not special – they also developed in a world of medium-sized objects. So why can scientists use those same intuitions in scientific cosmology? Scientists may have some advanced reasoning at their disposal, but all this is based on intuitions very much like the one underlying the principle of sufficient reason. In other words, no matter how advanced a mathematical theorem these scientists may deal with, they are guided at every step of implication by these basic intuitions. Scientists regularly make use of the more basic forms of reasoning in forming their conclusions. Those complex mathematical models are all based on rational intuitions, which are supposedly useless when dealing with the very small or the very large. The problem with contending that our intellectual intuitions are suddenly invalidated when we deal with the “very large” or “very small” is that scientific endeavor in these areas are also then problematic.
The argument is circular
This is an objection you will only find in popular circles, because it is not a good objection. A decent formulation of it comes from Dan Barker, the atheist activist and author. Barker contends that the first premise of the argument (everything that begins to exist has a cause) “implies that reality can be divided into two sets: items that begin to exist and those that do not.”[ix] According to Barker, this is supposed to imply that God exists, because God is the only being that does not begin to exist, and the first premise is supposed to imply that beings that do not begin to exist do exist. So, the premise assumes that God exists. First, God is not the only being that does not begin to exist – abstract objects and universals also do not begin to exist ( although this does assume a realist metaphysics). Much more importantly, what Barker contends goes beyond what the premise both says and implies. The premise does not say anything about the set of things that do not begin to exist and it doesn’t imply that such a set is empty or not empty. In other words, the first premise does not imply that there are beings that do not begin to exist. It is unclear why Barker thinks that the first premise depends upon the idea that the set of things that do not begin to exist is not empty. He gives no argument for supposing this. He seems to be suggesting that in order for n to be meaningful, there have to exist things that are not-n. This is irrational. I can claim that there is a dog with a heart, and this does not mean that there are dogs without hearts. By claiming “If dogs have hearts, they have heart valves”, I am not “dividing reality into” dogs with hearts and dogs without hearts, and I’m not implying that the category of dogs without hearts is even logically coherent. If I claim that “round circles have no edges” I am not implying that a non-round circle is a coherent concept or that they exist. But it is perfectly true that round circles have no edges (even if it is tautological). Also, the argument wouldn’t be circular if the premise did presuppose that a being that does not begin to exist is logically possible. It would only be circular if it presupposes the existence of that being.
[i] William Lane Craig, Reasonable Faith, (Wheaton: Crossway, 2008) 111
[ii] Ibid., 117
[iii] Ibid., 153
[iv] Ibid., 152
[v] William Lane Craig, “Causation and Spacetime” Q&A with William Lane Craig, #148, February 15, 2010, Retrieved from http://www.reasonablefaith.org/causation-and-spacetime
[vi] J.L. Mackie, The Miracle of Theism, (New York: Oxford University Press, 1982) 93
[vii] William Lane Craig, Reasonable Faith, 123
[viii] J.L. Mackie, The Miracle of Theism, p. 87
[ix] Dan Barker, Godless: How an Evangelical Preacher Became one of America’s Leading Atheists, (Berkeley: Ulysses Press, 2008) p. 130