Craig on Vilenkin on Cosmic Origins

by Luke Muehlhauser on May 7, 2010 in Kalam Argument

In 2003, Arvind Borde and Alexander Vilenkin and Alan Guth published a paper claiming to prove that the universe cannot be infinitely old. William Lane Craig is fond of quoting Vilenkin in defense of his Kalam Cosmological Argument:

It is said that an argument is what convinces reasonable men and a proof is what it takes to convince even an unreasonable man. With the proof now in place, cosmologists can no longer hide behind the possibility of a past-eternal universe. There is no escape: they have to face the problem of a cosmic beginning.

This comes from page 176 of Vilenkin’s book Many Worlds in One. But literally two paragraphs later, Vilenkin writes:

Theologians have often welcomed any evidence for the beginning of the universe, regarding it as evidence for the existence of God… So what do we make of a proof that the beginning is unavoidable? Is it a proof of the existence of God? This view would be far too simplistic. Anyone who attempts to understand the origin of the universe should be prepared to address its logical paradoxes. In this regard, the theorem that I proved with my colleagues does not give much of an advantage to the theologian over the scientist. As evidenced by Jinasena’s remarks earlier in this chapter, religion is not immune to the paradoxes of Creation.

What were Jinasena’s remarks? In the 9th century, this Jain poet wrote:

The doctrine that the world was created is ill-advised, and should be rejected.

If God created the world, where was he before creation? …

How could God have made the world without any raw material? If you say he made this first, and then the world, you are faced with an endless regression…

Thus the doctrine that the world was created by God makes no sense at all…1

Divine creation has its own set of a paradoxes. Jinasena mentioned only one.

  1. Jinasena, Mahapurana, in A.T. Embree (ed.), Sources of Indian Tradition (1988). []

Previous post:

Next post:

{ 39 comments… read them below or add one }

LBC May 7, 2010 at 6:19 am

No, there’s no paradox. It’s quite simple really. You see, God is a timeless, immaterial, disembodied mind with enormous power. You don’t need to explain where he was before creation, because he exists outside of time. Also, he is so powerful that he doesn’t need any raw material from which to create. He simply breathed everything into existence himself, with his timeless, immaterial breath! If you still need more proof, just read Psalm 33:6.



MauricXe May 7, 2010 at 7:02 am

The best part about what you said is that an immaterial being has a mouth :)

I wonder if anyone has ever quoted that second paragraph from Vilenkin during a debate with Craig. That would make an interesting response. Although I’m sure Craig has something prepared.


Taranu May 7, 2010 at 7:12 am

“There is no escape: they have to face the problem of a cosmic beginning.” Vilenkin makes such a radical statement. Does the 2003 paper provide definitive proof that the Universe cannot be infinitely old?


Josh May 7, 2010 at 8:02 am


See for yourself


Ajay May 7, 2010 at 8:45 am

The most convincing argument I’ve heard against Kalam deals with the idea of causality itself. How can the universe be “caused” when the very idea of causation requires time? Time is an inherent property OF the universe. Before the universe there was no time. Therefore, there could be no action undertaken by a purported God ‘prior’ to the creation of the universe. The idea, as Hawking stated in ABHOT, simply makes no sense.


ayer May 7, 2010 at 9:09 am

Vilenkin’s philosophical speculations are somewhat amateurish; isn’t the real issue instead that his scientific conclusion serves to buttress the relevant premise of the kalam cosmological argument?


Erika May 7, 2010 at 9:11 am

The key take away here is that any time you talk about the origin of the universe, you start running up against paradoxes.

This is what bugs me most about Craig’s defense of the premise that the universe must have a cause. His defense seemed to rest mainly on observation that a causeless universe seems paradoxical. However, a caused universe also seems paradoxical.


lukeprog May 7, 2010 at 9:15 am


Craig is always prepared… MUAHAHAHAHA!


lukeprog May 7, 2010 at 9:15 am


Actually, that same chapter contains a convenient summary of the argument presented in the 2003 paper; definitely worth a read!


lukeprog May 7, 2010 at 9:26 am


Yes, and that is a key part of my response to the kalam in my opening statement in my hypothetical debate with Craig that I wrote a long time ago but never find the time to record. :)


aigbusted May 7, 2010 at 11:14 am

In the third line of your quotation of Vilenkin, it says “What does we makes of this”

What DOES we make of this? Tsk, tsk, tsk. J/K. Does need correction though.

Anyway, are you going to publish something about Vilenkin’s model for the origin of the universe “from nothing”?


lukeprog May 7, 2010 at 11:23 am


Currently, I have no plans to write much about cosmological models of the origins of the universe, since I don’t have the requisite training to understand them!


oarobin May 7, 2010 at 11:46 am

The issue of the origin of the universe has four possibilities as i see it

a) Infinite regress of causes
b) A terminator to regress of causes
c) Multiple terminators to the regression
d) A causal loop

and ALL of the four are in play given current understanding.
yes infinite regress is not ruled out by saying that that since we experience a flow of time from the past to the future on a “Moment by Moment” basis then there cannot be infinite moments in light of the fact that physicist use a block view of time, general relativity says there can be closed, bounded universes with nothing outside the universe and particles whose rest mass is zero experience no time at all.
that is our intuition about time are not a good guide to describe the universe and one needs to look on formal arguments to decide if an infinite regress of causes is actually ruled out.


Shane Steinhauser May 7, 2010 at 2:35 pm

Most scientists that “accept” the begining of the universe simply assume that General Relativity remains true down all the way down to a singularity.


Muto May 7, 2010 at 5:00 pm

That’s funny.
I read the paper today then I looked on this blog and Luke wrote about it. Even in the paper there is a section in the discussion where the authors say that:

‘Whatever the possibilities for the boundary, it is clear
that unless the averaged expansion condition can somehow
be avoided for all past-directed geodesics, inflation
alone is not sufficient to provide a complete description of
the Universe, and some new physics is necessary in order
to determine the correct conditions at the boundary’

Furthermore I found an iteresting paper by one of the coauthors called ‘Eternal inflation and its implications’. In it he also mentions other papers that describe universes where the necessary conditions for the theorems are not met. I haven’t read them by now though.


Gavin Brown May 7, 2010 at 8:42 pm


Craig doesn’t appeal to Vilenkin in explicit support of God’s existence. He only does so to support premise two, “The universe began to exist.” It is merely a proof of a premise in an argument for God’s existence.

And in my view, Vilenkin’s refusal to agree that his (and Borde’s and Guth’s) theorem supports God’s existence actually gives more credibility to Craig’s appeal, since Vilenkin’s motives exonerate Craig from the charge of special pleading.

Thus, the second paragraph you cited by Vilenkin is interesting, but does little if anything to weaken Craig’s appeal to Vilenkin in support of premise two.


Muto May 8, 2010 at 2:31 am

Gavin Brown,

As Vilenkin points out, Craigs view of the theorem is terribly simplistic in nature. He uses this theorem quite out of context of the inflationary modell and I think his notions of causation and time and ‘coming into being’ are so naive that most physicists will just facepalm if they hear the Kalam argument. Its like some extremely stupid creationist arguments: There are so many wrong views hidden in Craigs language and the logic becomes so convoluted that it is very difficult to effectively point out the mistakes.


lukeprog May 8, 2010 at 5:03 am


You’re correct that Craig only appeals to Vilenking in support of premise 2. I’m just agreeing with Vilenkin’s point that ‘God did it’ has its own set of paradoxes.


Gavin Brown May 8, 2010 at 7:03 am


But all Craig is attempting to demonstrate is that the universe began at some point in the past. And one way he attempts to show this is to appeal to a legitimate authority (as we all do with respect to technical matters of which we have only cursory knowledge).

I’m not trying to be a Craig apologist here, but I don’t see how Craig’s appeal to the Borde-Guth-Vilenkin theorem in support of premise two is illegitimate.

Of course, Vilenkin is correct that his theorem is not a “proof” for God’s existence, but Craig does not claim this either.

But if you argued that the universe began to exist, and someone said, “No, I disagree; the universe did not begin to exist,” would you not then appeal to some other credible information that supports your view? That seems to be what Craig is doing with Vilenkin.


Gavin Brown May 8, 2010 at 7:05 am

Luke, My previous comment was partially in response to Moto’s comment, even though I addressed you. My bad.


Muto May 8, 2010 at 10:11 am

I tink the use of the theorem is illegitimate because Craig mixes up his own intuitive notions of ‘coming into existence’ with ‘any backward-going null geodesic with Hav >
0 must have a finite affine length’ what the theorem states. If one is neither physicist nor mathematician and has no training in these subjects it is very easy to mix up how things should behave with how they actually are. E.g: If I state casually that there are more real numbers than integers, I don’t mean ‘more’ in any commonsensical sense but that there exists no bijection between the set of integers and the set of real numbers. It can be challenging for outsiders to interpret the work of physicists properly even if they try to make is as accesible as possible in their popular books. I think Craig Kalam Argument is a combination of missrepresented physics, naive usage of words and whole bunch of faith holding it togeter.


Muto May 8, 2010 at 10:32 am

To use the example of the last post for a complete nonsensical ‘proof’: The number of integers is infinite. My metaphysical intuition tells me that nothing can be *more* than infinite. There are *more* real numbers than integers. Therefore real numbers don’t exist.


Gavin Brown May 8, 2010 at 12:17 pm


I’m not really sure what you’re getting at. Numbers have no causal relation and are therefore abstract, so to illustrate your point with math is problematic (in the same way that Craig’s use of Hilbert’s Hotel is not without it’s difficulties).

But Craig’s use of Vilenkin is straighforward and simple enough; there’s no need for interpretation. Craig argues that the universe began and Vilenkin agrees, no? Craig’s not being a physicist is irrelevant.


Muto May 8, 2010 at 4:33 pm

Gavin Brown,

You are right that the analogy between mathematical structures and reality doesn’t seem straightforward.But you will have to agree that the physical models cosmologists use are mathematical in nature. What I was trying to point out, that if Vilenkin says something like ‘the universe had a beginning’, he doesn’t necessarily mean the same thing as Craig. If I may quote coauthor Alan Guth:

‘There is of course no conclusion that an eternally inflating model must have a
unique beginning, and no conclusion that there is an upper bound on the length of
all backwards-going geodesics from a given point.’

Both would be inherent properties Craigs ‘beginning’ would have. Maybe he will not need them in his argument, but the problem is that Craig doesn’t seem used to this way of thinking and when he says ‘the universe came into existence’ it doesn’t correspond to what Vilenkin says.

I’m not critiquing Craigs 2nd premise explicitely, but more the convolution of scientific and metaphysical notions in his mind. This convolution takes place exactly because Craig is neither physicist nor mathematician.


Byrom May 9, 2010 at 4:34 am

This isn’t really a very good rebuttal to Dr Craig quoting Alexander Vilenkin. All it does is show Vilenkin’s personal opinion about how his discovery fits into theistic arguments – but it doesn’t change the fact of what he discovered. If Dr Craig has a logically valid and sound deductive argument, into which Vilenkin’s model fits perfectly, then that is what we must challenge. Indeed it’s rather complimentary to the Kalam Cosmological Argument that the objective scientific evidence supports it, and that the subjective opinion of the scientist is what is being used against it. How about some evidence which actually supports an eternal universe, in which case some evidence that Vilenkin’s science is wrong? Anybody got some?


rvkevin May 9, 2010 at 12:58 pm

How about some evidence which actually supports an eternal universe

First law of thermodynamics: Energy can neither be created nor destroyed. It can only change forms.


ayer May 10, 2010 at 6:05 am

First law of thermodynamics: Energy can neither be created nor destroyed. It can only change forms.

Second law of thermodynamics: the universe will reach a state of heat death due to entropy given sufficient time; if it has existed for an infinite amount of time, why is it not already in a state of heat death?


Muto May 10, 2010 at 8:13 am


Several prepetual universe models would violate the second law of thermodynamics by recreating a state of uniform density.


ayer May 10, 2010 at 9:21 am

ayer,Several prepetual universe models would violate the second law of thermodynamics by recreating a state of uniform density.  

So the first law of thermodynamics is inviolate but the second law can be discarded?…interesting


Muto May 11, 2010 at 11:22 am

The second law is statistical in nature and many physicists don’t consider it to be a fundamental law, whereas the conservation of energy is one of the priciples of modern physics. Furthermore I wasn’t defending rvkevin, but objecting to the often heard complaint about perpetual universes. But if the universe is indeed as Vilenkin’s proof assumes, we need not to argue about the second law of thermodynamics in this context.


Lamplighter Jones May 12, 2010 at 10:46 am

The Poincaré Recurrence Theorem: Every state in an autonomous system that occurs with positive probability (i.e. being far from thermodynamic equilibrium) will, with probability one, recur infinitely often. In particular, in a universe with finite mass and finite energy, in which mass-energy is conserved, the probability, with respect to any measure preserved by the underlying dynamics, of permanent heat death is zero (although trajectories resulting in heat death exist).

This is certainly incompatible with Ayer’s description of the Second Law, and is also a theorem of (a fragment of) Peano Arithmetic; in fact only finite additivity, not countable additivity, is assumed. Given the choice between believing arithmetic or believing Ayer’s description of the Second Law, I would have to go with arithmetic.

Of course, the (natural?) assumption of ergodicity implies that with probability one, the universe will experience states resembling heat death, and subsequently return to low-entropy states.


Arizona Atheist May 28, 2010 at 10:32 am

Interesting discussion. I’ve just recently finished trying my hand at refuting Craig’s arguments for God and while doing research on the theorem by Vilenkin, et al. I found this page. I agree with commenter Gavin Brown in that I also didn’t feel the quote in Vilenkin’s book was a very good rebuttal so I did one better. I was able to get a hold of Vilenkin myself via Victor Stenger and he told me that it is possible to get around his theorem so it’s not a definitive argument; it’s not set in stone so it’s not that useful to Craig.

I simply do not know – as does anyone, really – if the universe was “caused” by something, came from “nothing”, as in the example of a vacuum in space, etc. Either way, I think it’s completely illogical for theists to insist it was their god. I find the entire concept just ridiculous. Science is divided on the issue and theology and philosophy obviously aren’t going to give us an answer to this so I think we have to wait for further scientific discoveries.


lukeprog May 28, 2010 at 1:16 pm

Arizona Atheist,

Yeah, that post of yours shall be linked in my next ‘news bits’ post.


Arizona Atheist May 28, 2010 at 3:31 pm

Hi Luke,

Thanks! I’m curious and looking forward to any comments you may have. Take care.


H.S.Pal June 23, 2010 at 4:01 am

A. Circular Reasoning

In his article ‘The other side of time’ (2000) scientist Victor J. Stenger has written:
“Quantum electrodynamics is a fifty-year-old theory of the interactions of electrons and photons that has made successful predictions to accuracies as great as twelve significant figures. Fundamental to that theory is the spontaneous appearance of electron-positron (anti-electron) pairs for brief periods of time, literally out of “nothing.””
From here he has concluded that our universe may also come literally out of nothing due to quantum fluctuation in the void, and therefore we need not have to imagine that God has done this job.
But is it true that electron-positron (anti-electron) pairs are appearing literally out of “nothing”? Are scientists absolutely certain that the so-called void is a true void indeed? Because here there is a counter-claim also: God is there, and that God is everywhere. So actually nothing is coming out of “nothing”, only something is coming out of something. Here they will perhaps say: as there is no proof for God’s existence so far, so why should one have to believe that the void here is not a true void? But even if there is no proof for God’s existence, still then it can be shown that scientists’ claim that the universe has literally come out of nothing is a pure case of circular reasoning. If believers say that the void is not a true void at all, and if scientists still then hold that it is nothing but a void, then this is only because they are absolutely certain that God does not exist, and also because they think that God’s non-existence is so well-established a fact that it needs no further proof for substantiation. But if they are absolutely certain that God does not exist, then they are also absolutely certain that God is not the architect, designer, creator of our universe, because it is quite obvious that a non-existent God cannot be architect, designer, etc. So their starting premise is this: God does not exist, and therefore our universe is definitely not the creation of a God. But if they start from the above premise, then will it be very difficult to reach to the same conclusion?
But their approach here could have been somehow different. They could have said: well, regarding void, it is found that there is some controversy. Therefore we will not assume that it is a void, rather we will prove that it is such. Then they could have proceeded to give an alternate explanation for the origin of the universe, in which there will be neither any quantum fluctuation in the void, nor any hand of God to be seen anywhere. And their success here could have settled the matter for all time to come.

B. “Circular Reasoning” Case Reexamined


There can be basically two types of universe: (1) universe created by God, supposing that there is a God; (2) universe not created by God, supposing that there is no God. Again universe created by God can also be of three types:
(1a) Universe in which God need not have to intervene at all after its creation. This is the best type of universe that can be created by God.
(1b) Universe in which God has actually intervened from time to time, but his intervention is a bare minimum.
(1c) Universe that cannot function at all without God’s very frequent intervention. This is the worst type of universe that can be created by God.
Therefore we see that there can be four distinct types of universes, and our universe may be any one of the above four types: (1a), (1b), (1c), (2). In case of (1a), scientists will be able to give natural explanation for each and every physical event that has happened in the universe after its origin, because after its creation there is no intervention by God at any moment of its functioning. Only giving natural explanation for its coming into existence will be problematic. In case of (1b) also, most of the events will be easily explained away, without imagining that there is any hand of God behind these events. But for those events where God had actually intervened, scientists will never be able to give any natural explanation. Also explaining origin of the universe will be equally problematic. But in case of (1c), most of the events will remain unexplained, as in this case God had to intervene very frequently. This type of universe will be just like the one as envisaged by Newton: “Gravity explains the motions of the planets, but it cannot explain who set the planets in motion. God governs all things and knows all that is or can be done.” So we can with confidence say that our universe is not of this type, otherwise scientists could not have found natural explanation for most of the physical events. In case of type (2) universe, here also there will be natural explanation for each and every physical event, and there will be natural explanation for the origin of the universe also. So from the mere fact that scientists have so far been able to give natural explanation for each and every physical event, it cannot be concluded that our universe is a type (2) universe, because this can be a type (1a) universe as well. The only difference between type (1a) and type (2) universe is this: whereas in case of (1a) no natural explanation will ever be possible for the origin of the universe, it will not be so in case of (2). Therefore until and unless scientists can give a natural explanation for the origin of the universe, they cannot claim that it is a type (2) universe. And so, until and unless scientists can give this explanation, they can neither claim that the so-called void is a true void. So scientists cannot proceed to give a natural explanation for the origin of the universe with an a priori assumption that the void is a real void, because their failure or success in giving this explanation will only determine as to whether this is a real void or not.


Scientists want to prove that God does not exist. Since they want to prove it, therefore they cannot claim that it is already an established fact. So the statement “God does not exist” can be given the status of a theory only and nothing more than that. Therefore its fate will be determined like any other theory of the scientific world. Like any other scientific theory it will have to prove its validity afresh at each and every new instance. So, not by assuming that the void is a real void, and thus not by assuming that there is no God, but by any other means, scientists will have to show that there is no hand of God behind the origin/birth/creation of this universe, and therefore their no-God theory is again validated here. So the scientific community all over the world should realize that the story of the origin of our universe from a vacuum fluctuation is a myth only, not a scientific truth.


The question is absolutely irrelevant here as to whether scientists have done anything wrong by treating the void as a real void. Also the question is absolutely irrelevant here as to whether God has got any chance to revive. The real question is: the premise from which scientists begin here (that the void is a real void, that particle-antiparticle pairs are appearing practically from nowhere) already contains the conclusion to which they want to reach (that God was not needed to create this universe). From purely logical point of view this is untenable. And therefore, these scientists should rethink again before they declare that our universe is a free lunch.



Zeb June 23, 2010 at 9:11 pm

HS Pal, that’s great. I would love to see the responses you’d have got if you’d posted this sooner.


phil July 30, 2010 at 12:09 am

isuggest anyone debating Craig should read the Vilenkin’s book andGuths lectures too. There whole point is if inflation is eternal into the future can it be eternal into the past?”They answer no, but there are a number of important points to note:
1) eternal inflation is thhe process of creating an endless ensemble of pocket universes, so this is referring to the origin fo the mutliverse, not the universe. The diffculty is that some people refer to the ensemble of pocket universes as the universe.
2) Aguire and Gratton published a paper showing a way round Thhe theoreom as Guth himslef states in hiw recent lectures
3) Vilenkin also shows us how its possible for the universe to “come from nothing”.

If Craig wnts to use Vilenkin et al theoroems he should admit that thhey imply
a) a mutliverse , he wont want to do this becuase it will destroy his fine tuning argument
b) creation of the universe/multiverse from nothing, this is why Vilnekin probably rejetcs the Kalam argument. its not just opinion, its implied by the physics.


J.A. Kraulis September 7, 2010 at 1:21 am

H. S. Pal,

There are many more types of universes – or possible realities may be a better way of putting it – than are dreamt of in your philosophy.

For example, there could be multiple universes, all of them created by God or none, or some created by God, and some not. There could be a universe that was not created by God, but into which God came into existence and sometimes intervened. Or not. There could be two gods that created the universe. Or four, one governing each of the four fundamental interactions. Or a vast number of gods could have created the universe. Or there is a supergod that created a number of subgods who then each created their own universe (ours with its imperfections being the product of one of the less competent subgods). Or the universe and God are one. Or God became the universe, transforming into it and ceasing to exist at that point. Or there may be no universe at all, everything is simply an illusion of your own mind, outside of which nothing exists.

The above is hardly a complete list. And no matter how long a list it were to be, it could only contain the possibilities we could imagine. There are possible kinds of universes or realities which we could not even conceptualize, which simply would be beyond our comprehension. For example, perhaps time is multidimensional, implying kinds of causation so weird that all the possibilities we can think of are in fact meaningless, like speculations on what happens when we get to the ends of the earth.

There will always be possibilities that neither science nor philosophy can prove or disprove. Not even within the realm of pure math can everything be proved or disproved, if I understand it correctly. (I don’t understand Godel’s incompleteness theorem, but I believe it concludes something of the kind.)

You write “Scientists want to prove that God does not exist.” That’s a nonsensical assertion. Science has no such agenda. Many scientists do of course respond when religion seeks to undermine scientific knowledge and tries to replace this with superstition, or when it tries to use science to make unjustifiable claims.

Whether your void is a “real void” or not is entirely irrelevant to science. Science will pursue the understanding of reality as far as it can, and there will be a point beyond which neither science nor philosophy will be able to go. Religion may attempt to there make some assertions, but these will be arbitrary and basically empty speculations, as they have always proven to be.


Udaybhanu Chitrakar August 27, 2011 at 7:19 pm

In olden-golden days the saying was: When there was nothing, there was God. When there will be nothing again, there will still be God.
But then came the scientists and changed everything. The above saying also changed to this: When there was nothing, there were quantum laws. When there will be nothing again, there will still be quantum laws.
These quantum laws are spaceless, timeless, changeless, eternal, all-pervading, unborn, uncreated and immaterial. Only that these laws lack consciousness. In every other respect they are just like God.
These quantum laws are spaceless, timeless and immaterial, because when there was no space, no time and no matter, there were still these quantum laws. (Vilenkin’s model)
These quantum laws are all-pervading, because these laws act equally everywhere.
These quantum laws are scientists’ God.


Leave a Comment