[Note: this article assumes the reader is already acquainted with certain philosophical terms, phrases and concepts. Please enjoy.]

The Kalam Cosmological Argument, as popularized by philosopher and Christian Apologist William Lane Craig, goes something like this:

P1. Whatever begins to exist has a cause (of its existence).
P2. The universe began to exist.
C. Therefore, the universe has a cause (of its existence).

While much ado has been made of this argument’s second premise—that the universe began to exist—my purpose here is to offer a critique of (P1), or more specifically, the “three arguments” Craig offers in favor thereof:

(1) Something cannot come from nothing.

(2) If something can come into being from nothing, it becomes inexplicable why just anything or everything doesn’t come into being from nothing.

(3) Common experience and scientific evidence confirm the truth of [P1].

I will address each argument in order:

(1) Something cannot come from nothing.

It’s odd that Craig ever refers to this an “argument”. It’s not an argument; it’s a claim. And why should I, his intended audience, believe this claim? Craig’s answer is that this is a “metaphysical intuition” we all share. But what is an intuition? As an evolved primate whose cognitive abilities were naturally selected for survival and mating, why should I—an atheist for whom the Kalam Cosmological Argument is intended—take this seriously as an epistemic tool for determining the ultimate metaphysical nature of reality? What should I understand my intuitions to be, exactly, if they are not ultimately reducible to induction? How could this “metaphysical intuition” (which I don’t share, by the way) be anything other than an inference—conscious or subconscious—derived from my empirical knowledge of temporal, spatial, ex materia causality in the physical world? Isn’t (1) really just (3) in ‘metaphysical clothing’?

Craig is impatient with this kind of skepticism. He quickly warns that anyone who believes that the spacetime universe itself could be uncaused has “quit doing serious metaphysics and resorted to magic”, and leaves it at that. But one could dismiss Kalam offhand by pointing out that (1) simply fails to establish with any clarity that his first premise is true, and that intellectual intimidation of the sort he resorts to at this particular impasse simply won’t cut it.

Further, there’s some very tricky language being employed here, and it causes other problems. Ask yourself if (1) is logically equivalent to:

   (1′) Something cannot not come from anything.

Well if “nothing” means “not-anything”, then yes, the two statements (1) and (1′) are indeed logically equivalent. However, this presents a problem, since God by definition “didn’t come from anything”. So if God exists, then this proposition is false, and vise versa. Meaning the (logically equivalent) proposition, “Something cannot come from nothing” is also false, or else theism is false.

Perhaps he would grant that the terms “nothing” and “not-anything” are logically equivalent, but reject that the two statements (1 and 1′) are logically equivalent. He might say—and I’m just spitballing, here—that the true logically equivalent proposition to (1) would be:

   (1″) Something cannot come from not-anything.

Meaning we must turn our focus to the phrase “come from”. God, you may argue, didn’t “come from” anything at all, whereas an uncaused universe would have “come from” not-anything, and that’s the difference. But is that even coherent, let alone true? I think not! Why think the phrase “came from” applies to the universe any more than it applies to God? If the earliest moment of the universe was the first moment of time itself, it is incoherent to speak as if the universe did any “coming from” at all. It would simply be the case that the universe exists at all points in time, and that there exists no moment prior to its existence… just like God.

Therefore, (1) could be true yet have no implications for the universe. This is the problem with applying language that normally assumes time and sequence (such as “come into being”, “begin to exist”, “come from”, “appear”, “pop into existence”, etc.) to the beginning of time itself. It’s all incoherent. Of course, Craig wants to define “begins to exist” (or “comes into being”) in a way that protects God and Kalam from these problems, ad hoc:

“The Kalam cosmological argument uses the phrase ‘begins to exist.’ For those who wonder what that means I sometimes use the expression ‘comes into being’ as a synonym. We can explicate this last notion as follows: for any entity e and time t, e comes into being at t if and only if (i) e exists at t, (ii) t is the first time at which e exists, (iii) there is no state of affairs in the actual world in which e exists timelessly, and (iv) e’s existing at t is a tensed fact.” 

Let’s work backwards. Right away, notice that (iv) assumes a theory of time which has been rejected by 85% of professional philosophers, yet which is crucial for Kalam to work. (iii) has clearly been added to the definition in order to rescue God from ‘beginning to exist’ per (i) and (ii), but it would be entirely unnecessary if only (ii) were a tad more precise:

   (ii*) there was a time prior to t at which e did not exist

Not only is this adjustment more parsimonious insofar as it eliminates any need for (iii), it is intuitive and universally affirmed by our experience (can you think of an exception?). Of course, we know why Craig can’t define “begins to exist” this way: if he did, it would mean that the universe, despite having a first moment of existence, would not require a cause.

It shouldn’t be controversial to point out that Craig is tailoring his definition of “begins to exist” to suit the conclusion of his argument, rather than formulating the best possible definition for “begins to exist,” and then deciding whether there is any argument to be made.

The real problem, though, is that something “beginning to exist”, once defined by Craig, no longer entails “coming from” anything. So the premise, “Whatever begins to exist has a cause” cannot be derived from, “something cannot come from nothing”. In other words, Craig’s definition makes it logically possible that “something cannot come from nothing” is true while “whatever begins to exist has a cause” is false. Therefor, (1) fails to support Kalam’s first premise.

Now what about…

(2) If something can come into being from nothing, then it becomes inexplicable why just anything or everything doesn’t come into being from nothing.

There are numerous problems with this, the most obvious of which is that it’s an appeal to ignorance: “If x, then we have no explanation for y” does not imply that x is false. It just implies we have no explanation for y. So? If God exists, it would mean there are many things we have no explanation for, yet this is not an argument for God’s nonexistence.

But that’s the least of it: the inference here is predicated upon a false dichotomy. The implication in Craig’s remarks is that if the causal principle as expressed in Kalam’s first premise is false, then nothing requires a cause of its existence. But that clearly doesn’t follow. Craig’s causal principle may just be incorrectly phrased, or some other causal principle may be true instead, or it could be that the scope of the causal principle is not “metaphysical”, but physical. It may be false that anything beginning to exist has a cause, yet true that anything beginning to exist ex materia within space and time has a cause. Whatever the case may be, it’s hardly “inexplicable” why we don’t ever see tigers and horses popping into existence uncaused, ex nihilo.

The more succinct response to Craig’s argument here is that it commits a fallacy of composition—the mistake of inferring that something is true of the whole, merely because it is true of some, or even every, part of the whole. Craig has a standard rejoinder to this:

“I have heard atheists respond to this argument in the following way. “Well, premise 1 is true of everything in the universe, but it is not true of the universe.” I think you can see that this is just the old Taxi Cab Fallacy again that we talked about with regard to Leibniz. You cannot dismiss the causal principle like a cab once you have arrived at your desired destination. Premise 1 isn’t just a physical law of nature, like the law of gravity, which only applies in the universe. Rather, it is a metaphysical principle which applies to being as being – it applies to being as such. Therefore, it governs all of reality, all of being.”

There are three problems with this: firstly, as discussed above, it is unclear how this “metaphysical” principle could have been derived from anything other than Craig’s empirical knowledge of causality within the physical universe. How is the “metaphysical” status of this proposition anything more than a bare assertion?

Secondly, Craig is begging the question here. When someone objects that the causal principle may only apply within the universe but not to the universe itself, they are clearly challenging Craig’s contention that the causal principle is metaphysical (as opposed to physical) in scope. That’s the whole point of the objection. It therefore does Craig no good to respond to this by re-asserting the very thing he is being asked to support. If the atheist were to agree that Craig’s causal principle is indeed metaphysical in scope, but argue that the universe is nonetheless exempt from it (for some reason), that would be the Taxi Cab Fallacy. And yet, nobody I’ve ever heard of has objected in this way.

Thirdly, Craig shouldn’t so incautiously draw attention to dismissing philosophical “taxi cabs,” as it were, when his inductive defense of Kalam’s first premise (3) dismisses both our empirical experience and our intuitions—like a cab—as soon as they have delivered him to his desired metaphysical conclusion. (More on this later.)

So Craig’s “taxi” response fails to engage the objection that he is committing a fallacy of composition. But we can make other objections to (2) as well. Supposing he is not just appealing to ignorance, there are actually two distinct arguments Craig could be making here—(a) and (b), we’ll call them—and because of his phrasing, it’s unclear whether he’s making a “would” argument or a “could” argument. If we formalize them both, the problem becomes clear:

1a. If the universe began to exist uncaused, then objects within the universe would begin to exist uncaused.

2a. Objects within the universe do not begin to exist uncaused.

3a. Therefore, the universe did not begin to exist uncaused.


1b. If the universe began to exist uncaused, then it is possible that objects within the universe would begin to exist uncaused.

2b. It is not possible that objects within the universe begin to exist uncaused.

3b. Therefore, the universe did not begin to exist uncaused.

(1a) is false, since even in principle, one object (if we can call the universe that) beginning to exist uncaused does not necessitate that any other object follow suit. It may entail the logical possibility of other uncaused objects, but hardly their necessity or actuality. Clearly, it is not inconsistent with my position to leave open the possibility of an uncaused object (say, some discovery to do with quantum physics); it is the party defending Kalam who shoulders the burden to demonstrate that whatever begins to exist must have a cause.

But what about argument (b)? Even if we grant (1b), the burden is still upon Craig to demonstrate (2b)—a difficult task without begging the question in favor of the conclusion for which he is arguing, viz., whatever begins to exist has a cause.

And that leaves us with….

(3) Common experience and scientific evidence confirm the truth of [premise 1].

Right out of the gate, this is patently untrue. Even if we follow along with Reasonable Faith or the Blackwell Companion and grant all of Craig’s arguments, the conclusion which follows is not that common experience and scientific evidence confirm the first premise of Kalam; it’s only that they cannot disconfirm it.

In any case, Craig says my experience of causality confirms that whatever begins to exist has a cause. But my experience of causality also confirms that causality is a spatial and temporal process, and that whatever is caused to exist has both a material and efficient cause. What should follow from this that the spacetime universe’s existence can’t have have been “caused” at all—efficiently or materially—at least not in any sense recognizable or meaningful to us.

But Craig won’t entertain this for even a moment (nor will he allow his audience to), and this unveils some troubling inconsistency: though he believes that material causation of the universe “may be simply overridden by arguments for the finitude of the past”, Craig refuses to extend this exact inference to an efficient cause of the universe. But why not? Especially since all the empirical data we have tells us that efficient causes, like material causes, exist and operate temporally and spatially—in fact we don’t even know what it would look like for them not to.

Still, Craig insists that the universe must have had some kind of “cause”, somehow, some way. Again, why cling to this? As a Christian, Craig unabashedly discloses that he is “committed to the doctrine of creatio ex nihilo”. But Craig’s personal motivations are irrelevant here; what reason does he give me, his audience? We already know the answer: the “metaphysical intuition” that (1) “something cannot come from nothing”. But as discussed above, (1) does not withstand critical scrutiny, and what’s worse, it is entirely unclear how the epistemology behind this “metaphysical intuition” does not ultimately reduce to—that’s right—induction.

It’s worth noting that Craig attempts to head off this reduction by asserting (not arguing; just asserting) that while (3) is inductive, (1) is “metaphysical”. But in so doing, Craig makes the surprising blunder of comparing apples to oranges: the term “metaphysical” describes the scope of the claim, not how we came to know it (a la “inductive”). Craig may contrast “inductive” with “intuitive,” or “physical” with “metaphysical,” but whatever his intent, there is no special epistemic privilege which attends a given claim, just because that claim speaks to metaphysics rather than physics. Protestations such as this will remain impotent until Craig can provide good reasons for thinking not only that (a) our intuitions are not reducible to induction, but that (b), our intuitions defeat induction.

So it seems that (1) and (3), far from being independent arguments, stand and fall together.  Craig’s experiential argument for Kalam’s first premise must presuppose the truth of his metaphysical intuitions, which, paradoxically, cannot have been conceivably derived from anything other than his experience of the world around him.

(There are other problems with Craig’s inductive argument for Kalam’s first premise, several of which I discuss here.Suffice it to say, the evidence given for Kalam’s first premise is not terribly compelling, at least not to anyone who doesn’t already share Craig’s theological commitments.

Tags: , , , , , , , ,