During William Lane Craig’s recent Reasonable Faith tour to the UK where he debated philosophers such as Stephen Law and Peter Millican, Craig received a vast amount of publicity for having Dawkins refuse to debate him. However, what was more important to me was either a severe case of philosophical amnesia, or Craig has dropped the Kalam Cosmological Argument, which has been a standard part of his three / four / five pronged attack for decades. Why, I wonder. Well, let me explain.


 The KCA is one of the cosmological family of arguments which seeks to logically prove that this universe requires a cause (which is implied to be God). It goes like this, in its simplest form:

1) Everything which begins to exist has a cause for its existence.

2) The universe began to exist.

3) Therefore, the universe had a cause for its existence.

I have set out elsewhere the weaknesses in these
arguments (http://atipplingphilosopher.yo..., but in short the problems are as follows.



The circularity of the argument


Firstly it is circular in that the only thing, it can be argued, that ‘has begun to exist’ is the universe itself (i.e. all the matter and energy that constitute the universe and everything in it). Thus the first premise and the conclusion are synonymous – the argument is entirely circular.


So how do I establish that the only things which has begun to exist is the universe? We may think that things like tables, chairs, humans, rocks, lemmings and so on exist. Well, they do in one sense (an arrangement of matter / energy), but in the sense of the abstract labels of ‘rock’ or ‘chair’, they are exactly that, abstract labels. Their existence, in Platonic terms, as some kind of objective entity, requires the philosophical position of (Platonic) realism. This is not a position that Craig adheres to. All we have on a nominalist or conceptualist worldview (as opposed to realist) is a transformative coming into existence. What this means is that what makes the chair, the molecules and atoms, already existed in some other form or other before the ‘chair’ came to be. So the matter or energy did not ‘begin to exist’. This merely leaves the label of ‘chair’.


Let’s now look at the ‘label’ of ‘chair’. This is an abstract concept, I posit, that exists only in the mind of the conceiver. Most philosophers agree that the part of the definition of abstracts is that they are causally inert. We, as humans, label the chair abstractly and it only means a chair to those who see it as a chair – ie it is subjective. My idea of a chair is different to yours, is different to a cat’s and to an alien’s, as well as different to the idea of this object to a human who has never seen or heard of a chair (early humans who had never seen a chair, for example, would not know it to be a chair. It would not exist as a chair, though the matter would exist in that arrangement). I may call a tree stump a chair, but you may not. If I was the last person on earth and died and left this chair, it would not be a chair, but an assembly of matter that meant nothing to anything. The chair, as a label, is a subjective concept existing in each human's mind who sees it as a chair. A chair only has properties that make it a chair within the intellectual confines of humanity. These consensus-agreed properties are human-derived properties, even if there may be common properties between concrete items – i.e. chairness. These properties are arguable and not objectively true themselves. Thus the label of ‘chair’ is a result of 'subjectively human' evolution.


If you argue that objective ideas do exist, then it is also the case that the range of all possible entities must also exist objectively, even if they don't exist materially. For example, a 'forqwibllex' is a fork with a bent handle and a button on the end (that has never been created and I have ‘made-up’). This did not exist before now, either objectively or subjectively. Now it does - have I created it objectively? This is what happens whenever humans make up a label for anything to which they assign function etc. Also, things that other animals use that don't even have names, but to which they have assigned 'mental labels', for want of better words, must also exist objectively under this logic. For example, the backrubby bit of bark on which a family of sloths scratch their backs on a particular tree exists materially. They have no language, so
it has no label (it can be argued that abstracts are a function of language). Yet even though it only has properties to a sloth, and not to any other animal, objectivists should claim it must exist objectively. Furthermore, there are items that have multiple abstract properties which create more headaches for the objectivist. A chair, to me, might well be a territory marker to the school cat. Surely they same object cannot embody both objective existences: the table and the marker!


When did this chair ‘begin to exist’? Was it when it had three legs being built, when 1/2, 2/3, 4/5, 9/10 of the last leg was constructed. You see, the energy and matter of the chair already existed. So the chair is merely a conceptual construct. More precisely a human one. More precisely still, one that different humans will variously disagree with.

Let's take the completed chair. When will it not become a chair? When I take 7 molecules away? 20? a million? The fallacy of the beard / sand dune / slippery slope will tell you that this is entirely subjective.

Now let's take an animal - a cat. What is this 'chair' to it? I imagine a visual sensation of 'sleep thing'. To an alien? It looks rather like a shmagflan because it has a planthoingj on its fdanygshan. Labels are conceptual and depend on the conceiving mind, subjectively.

So, after all that, what has begun to exist? A causally inert abstract concept.


You see, once we strip away the labels and concepts, all we have left is matter and energy which is only ever involved in what can be called 'transformative creation', meaning it doesn't begin to exist, but is being constantly being reformed throughout time. It only began to exist at the Big Bang or similar (in Craig’s model). So now let's look at the KCA again:


1) Everything that begins to exist has a cause for its existence

2) the universe began to exist

3) therefore, the universe had a cause for its existence.

Let's look more closely at premise 1. We have agreed, then, that abstract concepts might begin to exist, but these are causally inert and do not exist objectively - only in the minds of the conceiver. So that leaves matter and energy, which has always existed because it is, in effect, the universe itself. It is not that the universe is 'made up' of lots of matter and energy making it something, it simply IS a quantity of matter and energy. This, then, makes the KCA look like this:


1) The universe that begins to exist has a cause for its existence

2) the universe begins to exist

3) therefore, the universe has a cause for its existence

The syllogism becomes entirely circular, and thus logically invalid. 

What is happening is you are trying to make a generalised rule about other things to apply that rule to the universe. But there are no other things, so the only thing that has begun to exist is the universe itself. We cannot seek to prove a conclusion about the causal behaviour of a one-off event merely by asserting it in the premise.



Category Error and Fallacy of Composition


Craig could be accused of committing further fallacy in two ways. Firstly, if you can establish that abstract ideas can be causally active and agree that energy and matter has existed since the Big Bang, you cannot then apply the behaviour of abstracts to the behaviour of matter. This would be a category error, and it does not necessarily follow that this would be the case.

As we can see from the Craig vs Perter Millican debate of 2011, Craig getd round this by merely asserting that it is intuitively and metaphysically impossible that you could have creation ex nihilo. There are two points to this. Firstly, Craig also claims that God can do this, so there must be some mechanism whereby this is possible. This leads some philosophers to claim that even God cannot create ex nihilo, since, in a metaphysical and logical sense, as Craig seems to imply, it is impossible. Secondly, he relies on assertion. He is not using anything other than intuition based on everyday efficient causality to assert this. However, as has been argued, this efficient causality is the Big Bang itself. Thus we return to the circularity again – Craig is just falling into the same trap. The demand for evidence and observation that ex nihilo creation is impossible is simply not there. The appeal to intuition (after all, we intuitively though the sun revolved around the earth) may be a correct intuition, but it is not logically sound. It is not deductive logic, and that intuition is asserted from illusory independent causations.


Furthermore, one cannot necessarily make a conclusion about the universe based on the behaviour of things in the universe, whether it be matter or abstracts. I would actually not hold much store in this, since I define universe and matter synonymously. In other words, the universe IS matter, rather than being ‘made up’ of matter. It is a quantity of matter, rather than being a different entity made from matter, such that a human is made up of human cells. The cell and the human are different entities. I would argue that the universe and matter are the SAME entity. If you don’t hold to my view there (that ‘universe’ is simply an abstract label to represent all energy and matter that exists), then Craig is committing said fallacy.

As Peter Millican (who believes this does commit this fallacy) said in his 2011 debate with Craig.


“Every change that I have experienced in the world has had , as far as I can tell, a physical cause. Even personal action as I have said before, happen through physical intermediaries. All our evidence suggest that mental activity is dependent on brain activity. And in any case, I have never seen mere thought create a new object. So a more precise version of Bill’s first premise would be, “Whatever begins to exist has a physical cause. However, if the universe is understood to include all physical things, then it seems obvious that it, as a whole, cannot have a physical cause. And I have excellent reason to doubt whether that principle can be valid as applied to the entire universe…

The only sorts of causes that we can understand, including intelligent causes, act in time. If time does not exist without the physical universe, then it is hard to understand how the notion of causation can even apply to the creation of that universe.”


While I actually don’t buy the notion that it commits the fallacy of composition in this way, and I disagree with his first paragraph, the second part I have included here is very important.


Causality and time


So this brings us on to further issues. If causality depends upon the notion of time, then if the universe with its spacetime did not exist, how can causality take place?

Does, then, causality depend on time? David Hume seems to have thought so. In his Treatise on Human Nature, Part III, section XV, he set out that  "The cause and effect must be contiguous in space and time." Causal processes are often defined whereby a process is the “development over time of an object. Processes are usually extended in time[1]”.

Without time, we have simultaneous causation, a notorious problematic notion. If two things occur simultaneously with each other, then one cannot define which one causes the other. William Wharton concludes this[2]:


In closing, special relativity makes very clear that causation has a close connection with the time coordinate, unlike the spatial coordinates. There is no lower limit to the speed at which a causal chain can traverse space-time between two events. This means the two events can be at the same spatial location at different times. However the two causally connected events can not be at different spatial locations at the same time, because the maximum speed at which the causal chain traverses space time is the speed of light, c. Causal chains must traverse time but not necessarily space.


It does seem very problematic to apply the behaviour of causation that we know only, and coherently, to operate in time, and then apply this to an a-temporal framework.

As the Satanford Encyclopedia of Philosophy declares[3] about the notion of simultaneous causation:


The main reply to the simultaneous causation argument is that the cases appearing to exemplify it are misdescribed (Mellor 1995). The iron ball takes time to depress the cushion, and in general all bodies take time to communicate their motions. There are no perfectly rigid bodies, at least in any nomologically possible world. Without the intuitive support of this sort of case, the simultaneous causation argument may be charged with begging the question.



Craig’s response


William Lane Craig has claimed that there is no problem with this type of refutation – claims of circularity and existence. He sets out Adolf Grünbaum’s[4] argument here[5]:


(i) The concept is used equivocally, since in the premiss it refers to causes which transform previously existing materials from one state to another, whereas in the conclusion it refers to a cause which creates ex nihilo…


And goes on to say about it:


To which it may be answered: (i) The univocal concept of "cause" employed in premiss and conclusion alike is the concept of efficient causality, that is to say, something which produces or brings into being its effects. Whether such production involves transformation of previously existing materials or creation ex nihilo is completely incidental. That this is so is evident from the fact that the proponent of the argument must confront and deal with the objection that the first cause may not have created ex nihilo, but instead transformed an eternal, quiescent universe into a universe in change (Goetz [1989]). So the argument is clearly not equivocal.


I simply do not think this is good enough on Craig’s part. He is ignoring the whole point about ‘begins to exist’. This is clearly an equivocation. The fact that he implies that this same causality applies to previously existing material means that he is ignoring the terminology of the first premise. ‘Everything that begins to exist has a cause for its existence’. If everything is previously existing, then Craig is tacitly accepting that the universe is the only thing which began to exist; ergo, a circular argument arises. His red herring in bringing in Goetz is irrelevant. He then merely asserts that the argument is not equivocal. I’m not sure he has provided and defence of his claim. I simply don’t think that Craig extrapolated the logic of Grünbaum’s argument correctly.

He attempts to get round this by simply appealing to efficient causality. But Grünbaum, and my, contention is that the efficient cause of all these ‘things’ (which did not begin to exist anyway) is the universe itself. One can agree or disagree over whether the universe began to exist. Assuming it did, though, we have a finite amount of matter / energy which began to exist. This opening Big Bang causes everything to happen. And I say that in the present tense. The causality of things happening now is that initial singularity. So nothing has begun to exist, and no cause has begun to exist, other than that first cause – the Big Bang singularity. Craig has done nothing to dispel this issue.

Let me analogise this as follows. Imagine there are 5 billiard balls A-E and nothing else. These came to exist at point t0 with an ‘introductory force’. At each point t1, t2 etc, every ball hits another ball. At point t5, B hits E at 35 degrees sending it towards C. Craig’s point seems to be this: the cause for B hitting E at 35 degrees is the momentum and energy generated in B as it hits E. That is his ‘efficient cause’. My point is this: the cause of B hitting E is at t0. No cause has begun to exist or has been created out of nothing. The causes transform – what is called transformative creation. So the cause of B hitting E is:

B firing off at t0 and hitting A at t1, the causal circumstance meaning it rebounds off A to hit D at t2, meaning the causal circumstance rendering it inevitable that it hits A again at t3…. And then it hits E at t5 at 35 degrees.

The cause is the casual circumstance at t5. This is identical to the causal circumstance in free will discussions – that determinism entails the cause of an action to the first cause of the Big Bang. The causal circumstance is everything up until the moment t5 as well as all the factors at the moment just prior to t5 (at t4). Craig is incorrect, in my opinion, in saying that the cause of is the immediate isolated efficient cause just before t5 (t4).

Furthermore, Craig claims this[6]:


 …the atoms of a watch have existed as long as the universe itself has existed, and that this is the same with everything in the universe. Never mind that atoms have not, as a matter of fact, existed as long as the universe; the more fundamental confusion is obviously the conflation of a thing with the material out of which the thing is made. Because the atoms currently composing my body have always existed, have I always existed? Did I exist during the Jurassic Age and the era of galaxy formation? If such a conclusion is not evidently absurd, reflect: I have certain essential properties, properties without which I could not exist. 


He says too:


"In order for something to come into existence, there must be a time such that the thing exists at t and there is no time t* earlier than t at which the thing exists," or more simply, "In order for anything to come into existence, there has to be a first moment of its existence."


What Craig is doing, then, to attempt to get out of this problem is to appeal to essential properties – abstracts -  as beginning to exist. But as we mentioned earlier, this is entirely subjective, and the nature of its existence questioned, as well as (as many philosophers would say) being causally inert. Essential properties, such as what identifies William Lane Craig as William Lane Craig, are:


1)                           arguably not things

2)                           arguably causally inert

3)                           as mentioned above, compositionally different from the main subject of the argument: the universe, thus allowing for a category error


So even if you could argue that properties aren’t just subjective concepts, that they exist in some dimension, you have a lot of leg work to do before arriving at the conclusion that their behaviour is causal in the same way that matter is.


Essential Properties


So the nub of Craig’s defence seems to be relying on essential properties. He adheres, obviously, to the philosophical notion of essentialism, the belief that objects have at least some essential properties (and one which attracts and has attracted notable criticism from Heidegger to Sartre). What is an essential property? An essential property (as opposed to an accidental property) is a property that an object must have in order to be defined as that object. For example, an accidental property of mine is that I love pizza, whilst an essential one is being human. In other words, I must be human in order to be me, but loving pizza is not so necessary. Of course, even this is open to debate, since being me, and not someone else, entails having properties that separate me from everyone else, such that loving pizza is essentially part of what makes me, me. It also seems like the definition of essential properties might be rather subjective. These ideas are further confused when understanding that there are essential relational properties, such that in every possible world where Peter and Mary exist, Peter is Mary’s father. Thus they have an essential relational property, but that property is not essential to Peter. Some argue that existence is an essential property, and this then refutes the notion that only God has necessary existence.  Properties, as you can see, are a major headache in philosophy. And yet, it seems that Craig has simply assumed these entities as having characteristics (causality) without defining them, or explaining how. Is it the property of something that causes something, or the thing itself? It certainly seems odd to suppose that a property itself can enact a certain causality.

There are issues about properties. One main one, in my books, is known as The Third Man Argument (and a similar relational problem known as Bradley’s regress). This illustrates that there is an infinite regress of property. It would be unwise to bog down this essay in getting into the finer details, but suffice to say that in order to accept this as a by-product of properties, Craig would have to accept infinities of properties – real infinities, in the sense that he gives properties real existence and causal powers.

Properties are a confusing subject. Some philosophers claim them to be abstract, casually inert; others to be some kind of ‘concrete’ entity; and others that they hover between abstract and concrete.

Now if Craig believes that properties are causal, then it is “difficult to see how our minds could make epistemic contact (and how our words could make semantic contact) with entities lying outside the spatio-temporal, causal order.”[7]

Some philosophers, still, claim that properties are causal powers when instantiated in the object. Of course, without the object existing, there is no causal power, so the property seems to supervene, depend upon the matter / energy for its existence. Thus we get back to the argument positing the Big Bang as the one and only cause.

Many naturalistic philosophers favour the Trope Theory to explain property; that each particular property is instantiated in each object so that objects that look red have individual tropes which resemble each other, but that the abstract universal ‘redness’ does not exist (out there in the abstract ether). Again, a confusing and in-depth theory.

My main point here is that in the KCA Craig fails to even tip his hat to the existence, and difficult debate surrounding, these deeply complex subjects. With one fell swoop, he assumes massive swathes of philosophy, disbelieves other massive swathes, and doesn’t even communicate the fact that he has done so in order to be able to assert the first premise of the KCA.


Uncaused, eternal God vs uncaused eternal universe


Craig posits an uncaused cause to explain, as a prime mover, the creation of the universe, in order to halt an infinite regression of causes. However, he does little to explore the notion of an eternally existing universe by putting all of his eggs into the Borde Guth Vilenkin basket. Though he has spent much time denigrating physics which supports these sorts of theories, he has been accused of cherry picking his cosmology, the quotes he employs from individual cosmologists, being outdated and so on. An eternally existent universe (energy etc) is no more improbable than an eternally existing God. Actually, employing Ockham’s Razor means that it is actually more probable, since one does not have to introduce further assumptions.

There really isn’t much more to say about this since one cannot apply a probability to two things for which we cannot remotely gauge a probability: the universe (matter / energy) and God.



The second premise – The universe began to exist


This is not the time and place to look at the cosmology required to either agree with or refute premise 2 of the KCA. Cosmology is a nascent discipline and most cosmologists cannot agree on what the origin of the universe is, or its basic ontology (is it necessary or not). I, personally, prefer to remain agnostic. I have a hunch, an intuition, that the universe must, in some way, be eternal (and thereby not invalidating the Law of Conservation of Energy). It is strange that scientists such as Borde, Guth and Vilenkin are not theists. Their own theories, that Craig has hijacked, do not lead them to the same conclusion, which begs the question about how Craig interprets their work.

There are many prevalent alternate theories to the idea of the singularity that Craig adheres to. Loop Quantum Gravity does a good job, as it stands, to argue a cyclical universe. But this is merely one amongst many.





So where does this leave us? It doesn't mean that just because the KCA is invalid, that the universe is definitely uncaused. It merely means that you cannot prove the universe is caused by using the deductive or inductive logic of the KCA. It could be that matter has been eternal, or that the universe was self-caused, or that God did indeed cause the universe. You just can't prove it from the KCA. This is what I have hopefully exhibited in this short essay.

[3] Jonathan Schaffer, The Metaphysics of Causation, http://plato.stanford.edu/entries/causation-metaphysics/ (11/11)

[4] Grünbaum, A. (1990) "The Pseudo-Problem Of Creation In Physical Cosmology" in Leslie, J. (ed.) Physical Cosmology And Philosophy. New York: MacMillan, pp.92-112

[5] The Origin And Creation Of The Universe: A Response to Adolf Grünbaum

Dr. William Lane Craig, http://www.leaderu.com/offices/billcraig/docs/origin.html