he Cosmological Argument is one of the more popular arguments presented today for the existence of a god¹. In this essay I’d like to briefly explain why I think the argument is very much overrated, and ultimately even unsuccessful.

 

The Cosmological Argument is an argument that seeks to prove the existence of a god by demonstrating that the universe came into existence via a First Cause or Uncaused Cause. There are various versions of the argument. The Kalam Cosmological Argument is a more concise version that was developed in light of the Big Bang theory. This is the version I will be focusing on in this essay, although any version of the argument would include some form of the premise I think is in question.

 

I want to point out a fatal flaw in the argument that Christian apologists (and even some atheists) tend to overlook. Some Christians would surely like to have my head for this. Before even posting the final version of this essay I’ve already received more than one e-mail asking why on earth I would want to argue against an argument for the existence of a god. Surely I must be nothing but a needless instigator, trying to divide the Body and help the atheist, they say. But there are several reasons why I find this essay necessary.

 

First, it is actually with precisely the opposite intention in mind that I choose to examine and criticize classical arguments for the existence of a god. Christian apologists are already divided over how apologetics should be done, and they need to come to a common understanding that is based in Scripture. This classical argument, as popular as it may be, is not an argument for Christian theism. It is an argument for what we might call a general theism, or a theory that encompasses a very broad plurality of monotheistic ideas as possible conclusions. Christians use this argument, Muslims use this argument, I’ve heard some Jews use this argument, Mormons use this argument, and I’ve even met a self-professing Deist who has used this argument. The argument is not an argument toward the Christian God specifically (which holds to a unique doctrine of God that no other theistic idea shares), but rather a general idea of theism.

 

Now, how exactly does the Christian apologist fulfill his duty to “give an answer for the hope that is within [him]” (1 Pet. 3:15) by arguing toward a general theism, as opposed to the Christian God specifically? Does the Christian’s hope within him rest on the mere fact that he has intellectual knowledge of a god’s existence, or on the assurance of his salvation in Christ? (Of course, a much more unfortunate question that should be asked about some budding apologists is, do they have an assurance of their salvation, or do they simply enjoy debating or showing off their intellectual abilities?) Some will say that arguments like the Cosmological Argument are but a “first step” in the apologetic encounter, and that once we establish the existence of a god, we can then focus on narrowing down to the Christian God. But this is actually a self-refuting way of arguing for Christianity, as we will have to wait and see in further posts. Furthermore, very rarely have I actually seen the Cosmological Argument presented as a “first step.” Normally the argument is laid out and defended, and then that’s it.

 

Second, the nature of the criticism I have already received for questioning the soundness of a beloved argument should actually lead us to question the motives behind these critics. Consider the possibility that the critique I am about to provide is a sound one, and the Cosmological Argument therefore fails. That’s something apologists who favor this argument would need to take seriously! How ridiculous would it be to brush off something so serious simply because an apologist likes using the argument? If it fails, it fails, and we need to deal with it.

 

But maybe you can defend the argument against my criticism. I am certainly sympathetic to that possibility, and look forward to hearing legitimate responses. After all, I would not go so far as to say that there is never a point in the apologetic witness where the Cosmological Argument cannot be used soundly. My criticism is not against the validity of the argument, but its soundness. There is a point in the apologist’s witness in which the Cosmological Argument may be used as a sound argument for God. However, I argue that this point comes so late in the witness that it would be unnecessary to present the argument anyway.

 

The reason the argument appears to fail is because one of its premises is unjustifiable without God. This means either that this premise must beg the question of God’s existence, or that God’s existence must already be proven before the Cosmological Argument can be presented soundly. But what point would there be in presenting the argument at that stage if God’s existence would have already been proven?

 

In recent years, the Kalam Cosmological Argument has increased in popularity among Christians thanks particularly to William Lane Craig, who states the argument as follows²:

 

 

It will then be argued in addition that this cause is itself uncaused (for there must be a termination point in the causal chain in order to avoid an infinite regress of causes), and usually that this uncaused cause is personal. But for the purposes of this essay it is fine to simply consider the basic three premises stated above.

 

This is an obviously valid argument, because the conclusion is entailed by the premises. It is not possible for both 1 and 2 to be true and 3 to also be false. However, is the argument sound? Regardless of an argument’s validity, it still does not work if the premises cannot themselves be shown to be true facts. And the demonstration of the truth of Premise 1 may be a problem for the apologist.

 

The statement, “Whatever begins to exist has a cause,” seems like an obvious truth. And I do believe that everyone feels in their heart of hearts that it is in fact true (even if we may even have reason to doubt it³). But the problem is that the only reason I think everyone believes this to be true is because everyone has a natural knowledge of God, having been made in His image. This would be a problem for the Kalam Cosmological Argument, because in order to avoid begging the question of God’s existence the apologist must be able to demonstrate how what was just stated is false. The apologist must be able to somehow demonstrate that Premise 1 is true even if it is not true that there exists a God whom all men know in their heart of hearts. Why? Because if it is assumed that the reason we can all agree that “whatever begins to exist has a cause” is because the law of God written on our hearts is what accounts for our intuitive awareness of logical laws (including a law of causality), then obviously the Christian apologist’s explanation of Premise 1 would ultimately assume God. Thus, in order to avoid begging the question in this manner, the apologist must be able to somehow prove Premise 1 without at all appealing to God. I do not think this can be done, and precisely for the same reason that the atheist’s espoused worldview is not able to adequately account for the laws of logic.

 

There are three possible ways in which the apologist can attempt to justify the truth of Premise 1. We can call these a priori, a posteriori, and transcendental methods of justification. An a priori justification is one that is demonstrable independently of experience (for instance, we can know a priori—that is, simply according to reasoning—that there are no three-sided squares. No kind of observational experience is necessary to know this.). An a posteriori justification is one that depends on experience (this would, for example, include all scientific theories or things discovered via the scientific method). The third option, a transcendental method of justification, is a form of “spiral” reasoning, beginning with the assumption that the Bible is true and then proceeding to indicate, indirectly, that this starting assumption must be shared by all in order to even be able to account for the very principles of logic and reason that make a statement like “Whatever begins to exist has a cause” intelligible to us.

 

A transcendental argument would in fact justify the truth of Premise 1 of the Kalam Cosmological Argument. However, we will not be considering that option here because it is essentially the alternative argument I spoke of above which would ultimately render the Cosmological Argument to be unnecessary. If we prove, transcendentally, that the very principle of causality that is assumed in Premise 1 is only accounted for by the existence of the Christian God, then we might have rational justification for believing the truth of Premise 1, but that justification would itself have already proven God’s existence. And with proof of God’s existence being necessary to even justify the first premise of another argument for God, this other argument for God becomes an unnecessary one.

 

So once again, in order for the Kalam Cosmological Argument to be sound (as a proof, itself, of God’s existence, that does not rely on another proof for its justification), it is necessary that the apologist be able to demonstrate that Premise 1 can hold true even if God does not exist. This means demonstrating that an objective principle of causality is not dependent upon God’s existence for its accountability. Or to put it in other words, the soundness of the Kalam Cosmological Argument relies on the assumption that law-like regularities in the world do not indicate the necessary existence of God. I would argue this assumption is false. But that is why I don’t accept the Cosmological Argument as a sound proof.

 

So the apologist is left with the options of an a priori justification or an a posteriori justification for the truth of Premise 1. But let’s be sure we are clear on what exactly either of these methods of justification would require. In order to avoid begging the question of God’s existence (or in order to avoid concluding, on the basis of having already proven God’s existence, that the Cosmological Argument basically serves no purpose), what the apologist must be able to demonstrate is how we can know the claim, “Whatever begins to exist has a cause,” to be a true and objective fact about the world even if a god does not exist, since we cannot begin with this assumption. The apologist ultimately places himself in the same arena as the atheist—needing to face the challenge of having to be able to account for the existence of objective law-like regularities in the world without an objective Law-Giver.

 

William Lane Craig has offered the following explanation for Premise 1 on his website⁴:

 

 

Craig offers here both an appeal to pure reason, as well as an appeal to experience. First he states that we know intuitively that something cannot come into being out of nothing. This might be a true statement, but I think it misses the point. We do not need to know that we know intuitively that something cannot come into being out of nothing so much as we need to know how we account for, or explain why we intuitively know this. I can rationally justify, by dismissing the Cosmological Argument and using a Transcendental Argument, the belief that the reason we know this intuitively is because we have all been made in the image of God. But this reason cannot be provided as justification for this intuitive knowledge when using the Cosmological Argument without begging the question of God’s existence, since to say that we have all been made in the image of God obviously assumes already that God exists. It is necessary, then, that in order to argue that we can know the truth of Premise 1 intuitively that it can be demonstrated how Premise 1 is a priori knowable, even if there exists no god. In other words, the apologist must be able to somehow demonstrate that it is a necessary truth in all possible worlds that everything that begins to exist has a cause, and that there can be no such thing as a spontaneous beginning or occurrence.

 

But it does not seem to me that Craig has offered a serious a priori justification for such a belief. Why is it that we should believe that in all possible worlds there can never be a case in which something spontaneously comes into existence? Is it at least possible, in any possible world, that something could begin to exist without a cause?

 

Consider the possibility of a chaotic world, in which the uniformity of nature does not hold. A chaotic universe is in fact a priori quite possible. The uniformity of nature is not a self-evident, a priori truth. It is a belief established through observation, and we can at least imagine a possible world in which there is no uniformity of nature (imagine, for instance, simply a world where scientific observation would not work; where there would be no apparent regularity sustained among a set of observations of any particular thing, at least not for any other reason than mere chance).

 

It is not necessarily true, therefore, that nature be uniform in order for a universe to exist. But if this is so, then a problem arises when we consider that the apparent universal validity of causal relations in our world would seem to rely on the assumption that nature is uniform. If it is at least a priori possible that a chaotic world might exist, then why is it not also possible in such a world that there exist no causal law? The fact that it is at least conceivable how a universe could exist in which there is no causal law is enough to conclude that the proposition “Whatever begins to exist has a cause” is not knowable a priori. It is a proposition that must be determined through experience. Hence, we cannot say that we do know intuitively what Dr. Craig claims, at least not without assuming the existence of God first.

 

Consider a common example of a causal relation: Lightning and thunder. We tend to reason that thunder (call this a B-event) is caused by lightning (an A-event), because we have consistently observed this to be the case throughout the past. When there is an A-event present in the future, therefore, we believe that there will be a B-event following it.

 

But why do we believe this? We don’t know it to be a priori true, because these are independently observed events, and in order for the causal relation between the two to be knowable a priori we would have to be able to validly deduce from “When there is an A-event” that “There will be a B-event.” However, “There will be a B-event” is not entailed in the premise that “There is an A-event.” We observe that B-events tend to follow A-events, and from that observation infer inductively that probably all B-events will follow A-events. We do not deduce from the occurrence of the A-event that there will necessarily be a B-event. But an inductive inference is an inference from experience, which is another way of saying that something is knowable a posteriori, not a priori. It is therefore possible, as far as reasoning alone can tell us, that a B-event (i.e. thunder) can occur without the A-event (lightning) preceding it. 

 

This is a problem. The fact that we cannot deduce the future occurrence of B-events from the present or past occurrence of A-events would indicate that the causal relation is not a necessity, but only an empirical generalization based on observation. This would seem to lead to the devastating conclusion that (again, without first assuming God’s existence) there is no purely rational reason to believe that the fact that a B-event occurs necessarily indicates that its occurrence is connected at all with a prior cause. Hence, we cannot know a priori that all events must have causes. And if we cannot know a priori that all events must have causes, then on what basis does Dr. Craig suggest that it is “intuitively obvious” that everything that begins to exist must have a cause? It might be intuitively obvious to the Christian. But you can’t argue this if you avoid assuming God’s existence from the start. The consistent non-Christian position would reason that our belief in a causal law is based entirely on observation, and not a priori knowledge.

 

So we must turn to the last option then. We cannot use a transcendental method of justification and still hope to make anything useful out of the Cosmological Argument when we are finished. We cannot use an a priori method of justification because it simply is not true, unless you beg the question by first assuming God, that we know independently of experience that “Whatever begins to exits has a cause.” So we are left with the option of an a posteriori justification of Premise 1; a demonstration of the truth of Premise 1 from experience.

 

Dr. Craig also appeals to this option in the above quote. But first he says that “if things really could come into being uncaused out of nothing, then it becomes inexplicable why just anything and everything do not come into existence uncaused from nothing.”

 

I just want to take a brief moment to consider this statement, because I’m not sure how it is suppose to be so evidently true. The mere idea that it might be possible for an event to occur without a cause does not necessitate that we should expect to see such events, or that we even can recognize such uncaused events if and when they do occur. There is no reason that it cannot be by chance that the only events we have observed have happened to have causes (as unlikely as that might be), nor is there reason to doubt that it is possible that we do observe events that are uncaused, but simply fail to realize that we do because of the expectation we have for there to be causes, based on our past observation of causal relations. In other words, it seems possible that we could observe events all the time that are uncaused, but simply infer believed causes onto those observed events since we assume that they must have one.

 

Additionally, I want to just make an interesting note on the wording of his first premise:  Dr. Craig supposes that every event has a cause. This seems to assume a very simplistic view of the universe. Why can’t an event have multiple causes? And why should we leap to conclude that a god is the cause of the universe, or that a particular god is the only cause? Why can’t multiple gods have worked together to cause the existence of the universe? Again, I’m only playing devil’s advocate here, to demonstrate the short-falls of an unbiblical apologetic method.

 

Furthermore, I am sincerely curious as to whether Dr. Craig would even fully believe in Premise 1 himself. William Lane Craig is a Molinist. He holds to a view of libertarian free-will that would seem to itself eliminate the principle of sufficient reason. Dr. Craig believes that many, if not all of our choices are predetermined by nothing. But if this is the case, then aren’t all of our free choices themselves events that come into being without any cause? Or perhaps our ‘free’ choices have influencing factors, but not one particular cause that determines them. But this would simply seem to confirm that there are events that come into being with multiple causes. So why could the beginning of the universe itself not be one of those events?

 

But let’s consider Dr. Craig’s attempt to provide an a posteriori justification of Premise 1. He states that the first premise is “constantly confirmed in our experience, which provides atheists who are scientific naturalists with the strongest of motivations to accept it.”

 

The obvious problem with this statement is that it immediately runs into the infamous problem of induction. Like the problem of justifying the concept of causality or the uniformity of nature, the problem of induction is a problem that points out the epistemological dilemma that we have no rational justification for our belief in the reliability of inductive reasoning. “Induction” is that type of reasoning which moves from particular observations to generalizations about the world. Hence, the problem of induction is the dilemma that asks, what reason do we have to believe that our past observations are a good indicator of future or unobserved observations? Why do we believe that the future will be like the past? Why do we assume that just because every past observation of a particular event has yielded a certain constant result, that therefore every other observation of a similar event will, or is even likely to, also yield the same result?

 

Consider an example. It has always been observed in the past that water freezes at approximately 0 degrees Celsius at sea level. Because there has been no significant variation among the results of these observations, we therefore infer inductively that all unobserved and future cases of water at sea level in 0 degree Celsius temperature will likewise result in freezing. But the question proposed by the dilemma is, what reason do we actually have to believe this? Why do we believe that these past observations give us good reason to assume that similar observations in the future will yield similar results?

 

The first and most obvious reason we will want to give is that we have so far been very successful in generalizing into the future from particular observations. (After all, science and common sense rely very heavily on this principle.) But this actually begs the question. We cannot point to our past success in using induction in order to justify our decision to continue using induction in the future, when whether or not we are rationally justified in generalizing about the future on the basis of past observation or experience is the very principle that’s validity is in question. On the other hand, we cannot offer a successful a priori justification for induction either, for very similar reasons as those discussed above. It would involve needing to somehow demonstrate that it is necessarily true that in all possible worlds, including worlds in which the uniformity of nature does not hold, past observations (so long as they yield relatively constant results) will serve as a good or likely indication of future events. But because it is perfectly conceivable how this might not be so, a principle of induction is not a self-evident, a priori truth.

 

Now of course, the problem of induction is very easily solved by using a Transcendental Argument for the existence of God. The Christian understanding of God can account for our belief in the reliability of induction. But once again, we cannot point to God as the solution to the problem when attempting to justify the first premise of the Cosmological Argument, lest we beg the question or defeat the purpose of the argument. Additionally, many proposed solutions to the problem of induction have been made from non-Christian perspectives, but none of them have ultimately been successful⁵. Thus, it is only on the basis of a Christian worldview that an appeal to past experience can consistently be used as a justification for any proposition. Yet, that Christian basis is the very thing the classical apologist must abandon in his reasoning if he is to have any hope of presenting a sound Cosmological Argument without begging the question of God’s existence.

 

So Dr. Craig cannot point to our past experience in observation in order to justify the premise in question, and it surely does not provide atheists “with the strongest of motivations to accept it”! The first premise of the Kalam Cosmological Argument, therefore, would seem to be unjustifiable without already assuming God. If you can show me otherwise, I will be glad to change my impression of this argument. But as it stands I do not find the Cosmological Argument to be a sound proof for the existence of a god. Because the principles of induction, causality, and the uniformity of nature, principles which are all presupposed to some degree in the first premise of this argument, would seem to already rely on the existence of the Christian God for their accountability, the argument is ultimately question begging. It must be demonstrated therefore how Premise 1 can be justified on the basis of a non-Christian framework.  But this cannot be done.  And the fact that it cannot be done is itself an even more powerful argument for God.  Hence, the Transcendental argument should be preferred instead of these classical arguments for God.  The only way you can rationally justify a belief in Premise 1 is by first proving that the Christian God exists, as the very thing which accounts for such law-like principles. At that point only in the apologetic witness can the Cosmological Argument be presented soundly, if my analysis here has been accurate. But again, what use would there be in presenting the argument at all at that stage if God’s existence has already been demonstrated?

 

In conclusion I want to make sure we are clear on what is being argued in this essay and what is not. I have not suggested here that the apologist who presents a Cosmological Argument for a god has no basis for the belief that “whatever begins to exist has a cause.” He does indeed have a rational basis at least for some kind of belief in a causal law (in fact, he does, and the atheist does not). But what has been suggested here (and what the Transcendental Argument argues) is that this rational basis is the already-established existence of God, and that no other attempt to justify this belief without appealing to God will ultimately prove to be successful."The apologist ultimately places himself in the same arena as the atheist—needing to face the challenge of having to be able to account for the existence of objective law-like regularities in the world without an objective Law-Giver." Hence, you cannot rationally justify the belief that whatever begins to exist has a cause without already appealing to God’s existence, since the uniformity of nature and the concept of causality cannot be accounted for without Him. And hopefully by now it has been made clear that the problem with this is that we cannot present an argument for God that uses a premise that relies on God’s existence for its justification. And this is one of the critical downfalls of the evidentialist/classical method of doing apologetics. By stepping off of the Christian worldview and onto a non-Christian worldview in the apologetic task, the classical apologist commits himself to reasoning as if he were also an unbeliever, and therefore sets himself up for failure.  The price for refusing to submit to the Lorship of Christ even in argumentation and reasoning is ultimately a self-defeating apologetic.

 

Indeed, the presentation of the Cosmological Argument and arguments like it is a self-refuting endeavor.  To justify the first premise of this argument we must be able to demonstrate the truth of a causal law and inductive principle independently of God.  If atheists can adequately demonstrate how a non-theistic worldview can account for law-like principles of causality and induction, then the first premise of the Cosmological Argument can be justified, since the non-Christian epistemology assumed in its starting point would be able to account for its presuppositions.  However, if atheists cannot adequately demonstrate this, then there is no hope for the Cosmological Argument.  Thus, the fate of the Cosmological Argument actually rests on atheistic arguments!  The soundness of an argument for God that does not begin by relying on Christian principles depends on Christianity not being true, because the only way this argument will ever be sound is if it can be shown that the non-Christian worldview can make sense of its usage of law-like principles of reasoning.  But if that can be shown, it would demonstrate that you do not need God in order to account for objective standards or law-like regularities in the world (a demonstration that would both cripple many other Christian arguments as well as suggest that the non-Christian worldviwe is an accurate representation of reality).  It is not I who is helping the atheist.

 

There is one, and maybe only one sort of argument that can be used to prove God’s existence without running into the same or similar problems that have been briefly considered in this essay. And that is what is called the Transcendental Argument for the existence of the Christian God, based out of the presuppositional method of doing apologetics. By beginning with the assumption of the truth of the Bible and God’s existence (and therefore reasoning on the basis of a Christian worldview, demonstrating the truth of its epistemological claims even in practice), the presuppositionalist temporarily steps onto the unbeliever’s worldview for the sake of argument in order to demonstrate, indirectly, that these principles which we all assume in reasoning—induction, uniformity in nature, causality, as well as other logical laws, moral standards, etc.—cannot be adequately accounted for by the worldviews espoused by non-Christians, and thus by arguing from the impossibility of the contrary we demonstrate that to deny the existence of the Christian God can only inevitably lead to self-contradiction and absurdity. We further argue that the only reason that non-Christians assume such principles that allow for things to be made at all intelligible to them is because of the inescapable knowledge of God that is written on their hearts, and hence all non-Christian arguments are inevitably self-defeating because they actually unconsciously borrow presuppositions from the Christian worldview in order to even form or make sense of their arguments against the Christian worldview.

 

This presuppositional approach is the biblical way of doing apologetics, as Paul himself demonstrates in his usage of such an argument in Acts 17. It is the most God-honoring way of doing apologetics, as it is the only method that takes seriously the beliefs that “the fear of the Lord is the beginning of knowledge” (Prov. 1:7) and “in [Christ] are hidden all the treasures of wisdom and knowledge” (Col. 2:3). And it is the most philosophically sound argument that can be given in defense of the existence of the Christian God, as it is the only way of arguing for God that reasons in a way that is consistent with the epistemological claims of the biblical worldview being argued toward. Any other arguments, which refuse to submit to the epistemic Lordship of Christ, but attempt instead to meet the unbeliever on his own turf and reason toward God on the basis of unbelieving principles, will ultimately prove to be unsuccessful, as we have seen an example of here.

 

 

¹ The Cosmological Argument does not argue for the Christian God specifically, but for a general theism; hence the argument has been used in the attempt to prove various monotheistic deities. "A god" is therefore more grammatically correct than "God" when speaking of the general argument.

 

² Craig, William Lane. Reasonable Faith. Wheaton: Moody Press. 1994, pg. 92

 

³ Much of quantum physics actually suggests there may be uncaused events in the world. The position of an electron, for example, or the movement of a gas molecule, is not determined by any apparent preceding factors (see: Kaku, Michio. Hyperspace: A Scientific Odyssey Through Parallel Universes, Time Warps, and the Tenth Dimension. New York: Oxford University Press, 1994.). Of course, on a biblical worldview not even the apparently uncaused events are truly uncaused, for everything is ordained by God (cf. Matt. 10:29-30).

 

⁴ http://www.reasonablefaith.org/site/News2?page=NewsArticle&id=5705

 

⁵ See: (BonJour, Laurence. In Defense of Pure Reason: A Rationalist Account of A Priori Justification. Cambridge: Cambridge UP, 1998.), Chapter 7, for a brief critical critique of solutions offered by Karl Popper, Hans Reichenbach and Wesley Salmon, P. F. Strawson, Frederick L. Will and Max Black, and see (Beebe, James R. "Can Rationalist Abductivism Solve The Problem of Induction?" Pacific Philosophical Quarterly 89 (2008): 151-68) and (Wrenn, Chase B. "Inter-World Probability and The Problem of Induction." Pacific Philosophical Quarterly 87 (2006): 387-402) for a critical critique of BonJour's own proposed solution to the problem.