05 / 06

St. John's College

Timothy Hull interviews William Lane Craig

Time : 00:50:43

Timothy Hull interviews William Lane Craig


MR. HULL: In his book The God Delusion, Richard Dawkins says the design hypothesis immediately raises the larger problem of who designed the designer. How do you respond to that?

DR. CRAIG: Dawkins’ central argument – in his words – of his book The God Delusion is this claim that you cannot infer God as the designer of the universe because you are left with the question “Who designed the designer?” In fact, this argument is a very bad argument for a number of reasons. For one thing, it does not prove atheism as Dawkins supposes. Even if the argument were entirely successful it doesn’t go to prove that God does not exist. At most all the argument would prove is that you should not infer the existence of God on the basis of design in the universe. That would be a mistaken inference. So if you believe in God, you should believe in him on some other basis: the cosmological argument perhaps, the ontological argument, the moral argument, maybe on the basis of religious experience, or revelation. The only thing his argument would prove, even if it worked, is that you should not infer God as an explanatory entity for the appearance of design in the universe.

But, that having been said, does the argument work? Well, not at all. What he mistakenly thinks is that in order to recognize that an explanation is the best you need to have an explanation of the explanation. And that is clearly mistaken. For example, if archeologists were digging in the earth and came across implements shaped like hatchet heads, tomahawk heads, arrowheads, pottery shards, and so forth, they would instantly recognize that these were the remains of intelligent design – that there was some unknown group of people who had made these artifacts and they happened upon them. They may not be able to explain at all the origin of those people, who they were, how they came to be there, but that wouldn’t matter to the inference that these artifacts are the products of intelligent design. You don’t need to be able to explain the explanation in order to recognize that it is the best. Or, again, if astronauts were to find a pile of machinery on the back side of the moon and we realized that neither we nor the Russians have launched such machinery into space, they would infer that this is the product of some extraterrestrial intelligent designer. Now we wouldn’t have any idea who they were, what they were like, how they got there, but nevertheless we would recognize that is the best explanation. You don’t need to have an explanation of the explanation in order to recognize that explanation as the best.

In fact, if you did have to have an explanation of the explanation to see that the explanation is the best, that immediately leads to an infinite regress so that nothing could ever be explained. It would completely undermine science. So the argument is simply mistaken. In order to recognize that the best explanation for biological and physical complexity in the universe is an intelligent designer, you needn’t need to explain the explanation or to explain the designer. That is one flaw with the argument.

The other flaw is this. He thinks that the inference to a divine designer doesn’t advance your explanatory power because the designer is just as complex as the artifact to be explained. Therefore, there is no advance being made. I think that is completely mistaken. That confuses a mind with a mind’s ideas. A mind’s ideas may indeed by very complex; this mind may be thinking of the infinitesimal calculus or Zermelo–Fraenkel set theory or something of that sort – enormously complex ideas. But a mind as an entity is a remarkably simple thing. A mind is a non-physical, mental or spiritual entity that has no parts. It is not composed. Therefore, it has no complexity to speak of. It is probably the most simple thing that we can conceive of. So the postulation of an unembodied intelligent mind behind the universe is definitely an advance in simplicity over the variegated and contingent constants and quantities inexplicably present in the universe.

So for those two reasons I think this argument, which Dawkins seems very proud of, is really quite sophomoric and quite hopeless.

MR. HULL: Richard Dawkins says that no testament is sufficient to establish a miracle unless this testament is of such a kind that the force of it would be more miraculous than the original event. It seems he is popularizing an argument which has been around since Hume. What is an effective response to that? [1]

DR. CRAIG: Well, I think the best response to David Hume’s argument against miracles has been by John Earman, the University of Pittsburgh philosopher of science who is himself an agnostic but who has written an excellent book entitled Hume’s Abject Failure[2] That is his characterization of Hume’s argument against miracles. What he points out is that Hume did not have a full grasp of what probability theorists call the probability calculus. So he doesn’t take account of all of the probabilities that are involved in computing the probability, for example, of the resurrection of Jesus given our background information of the world and the specific evidence for Jesus’ resurrection like the empty tomb, the postmortem appearances of Jesus, and the origin of the disciples’ belief in his resurrection. All Hume considers is the probability of the resurrection relative to our background knowledge of the world. And he says, relative to the background knowledge of the world, resurrection is highly, highly improbable. It goes against the laws of nature and therefore this is an intrinsically improbable event.

But what he doesn’t take into account is the probability of no resurrection given not only the background evidence but the specific evidence like the empty tomb, the appearances, and the origin of the Christian faith. If the probability of no resurrection, given the specific evidence and the background information, is low then that can counterbalance any intrinsic improbability in the resurrection itself relative to the background information alone. So Hume’s argument is simply incomplete. It doesn’t take into consideration the full range of the probabilities that you have to calculate in figuring out the probability of the resurrection on our background information plus the specific evidence.

Moreover, I want to say that I don’t see any reason to think that the resurrection is improbable on our background information. What is improbable relative to our background information about the world is the hypothesis that Jesus rose naturally from the dead. That is enormously improbable. I think almost any theory would be more probable than to say all of the cells in Jesus’ body spontaneously came back to life again and he rose naturally from the dead. That would be contrary to all of the bio-medical and scientific evidence. But that isn’t, of course, the Christian hypothesis. The Christian believes that it is naturally impossible for people to rise from the dead as wholeheartedly as he believes the hypothesis “God raised Jesus from the dead.” So what we want to know is: how improbable is the hypothesis, not that Jesus rose naturally from the dead, we want to know how improbable is the hypothesis that “God raised Jesus from the dead?” It is very difficult for me to see why that should be thought improbable. Particularly if part of your background information includes things like the evidence for the beginning of the universe, the fine-tuning of the universe, the existence of moral values, and the other arguments of natural theology which make the existence of God probable. If God’s existence is probable then this is going to increase the probability of the resurrection on the background information because you have already got a being in place who could pull it off.

So I don’t agree with Hume at all that the resurrection is intrinsically greatly improbable. I think at worst we’d have to say this probability is just unknown – that it is just a question mark – and therefore we don’t know what this probability is. But I do not think we can say with any confidence that the probability of the resurrection on our background information is terribly low. Certainly it is not so low that it couldn’t be counterbalanced by the improbability of no resurrection given the evidence that we have. I think that is enough to subvert Hume’s argument.

MR. HULL: The history in N. T. Wright’s book is great. I almost felt that if he and Richard Swinburne got together – because Swinburne is stronger in philosophy – if they were together they could be really powerful.

DR. CRAIG: I think that N. T. Wright’s argument for the resurrection has been misunderstood by some of its critics. Wright does not, in fact, make some sort of grandiose or very strong probability claim for the resurrection of Jesus. What he argues is that the facts of the empty tomb and the appearances of Jesus are established to such a high degree of historical probability as to be “virtually certain.” He compares them to the death of Caesar Augustus in AD 14 or the fall of Jerusalem in AD 70. [3] Notice that is not an argument for the resurrection. [4] That is simply an argument for the data to be explained. The inference to the resurrection is a second step that comes later. Here we are simply talking about how well established are the empirical facts to be explained. And Wright’s claim is that we have very good grounds for believing that the tomb of Jesus of Nazareth was found empty by a group of his women disciples on the Sunday morning following the crucifixion, and that thereafter various individuals and groups of people under a variety of circumstances and on multiple occasions experienced appearances of Jesus alive after his death. Now those are not themselves miraculous claims or conclusions. Those are quite ordinary facts that can be accessed by any secular historian.

Just to give an illustration. After the assassination of President Abraham Lincoln, as his body was being transported from Washington back to Springfield, Illinois for interment, there was actually a plot to steal Lincoln’s body from the train by some of his detractors. Now any historian of the American Civil War will want to know – was this plot successful? Was Lincoln’s body missing from the train? Or was it successfully interred in the tomb in Springfield? These are facts that any historian can look into and investigate. Similarly, whether the tomb of Jesus of Nazareth was found empty on Sunday morning, whether or not his disciples claimed to see appearances of him alive from the dead. These are facts that are available for any historian. It involves no inference to supernatural activity or miraculous claims whatsoever. And N. T. Wright claims that those facts are very firmly established. And I don’t think that that conclusion is way out of line. 75% of contemporary New Testament critics would agree with the facts of the empty tomb and probably 99 point something percent would agree with the fact that the disciples experienced postmortem appearances of Jesus. So he is well within the consensus of scholarship on those claims.

Now, the real question is then: how do you best explain these facts? Here Wright is very modest. He says almost nothing to justify the claim that the resurrection is the best explanation. He says simply that the Enlightenment naturalist who is against miracles and the supernatural needs to re-explore his worldview to see whether or not perhaps the universe is much more open than he previously thought; that perhaps there could be a God who caused this event called the resurrection of Jesus. He very tentatively offers this as an explanation that he believes in and that he invites the Enlightenment naturalist to investigate and look at again.

So I think it is entirely wrong to say that he is dogmatic or overly insistent on his conclusion. I think that he really is quite tentative. And I think you could mount a much more robust case for saying that the resurrection is the best explanation. In my own work, I try to use the traditional criteria that historians use for assessing the best explanation. These would include things like the explanatory power of the hypothesis, its explanatory scope, its plausibility, its degree of ad hoc-ness or contriveness, whether or not it contradicts very many accepted beliefs, and the degree to which it outstrips rival hypotheses in meeting these criteria. What I try to do is to argue that the resurrection hypothesis has greater explanatory power, greater explanatory scope, greater plausibility than many of its competitors like conspiracy theory, hallucination theory, wrong tomb theory, and things of this sort. N. T. Wright doesn’t do any of that. He simply punts to Gary Habermas at that point in his book and says for an examination of naturalistic alternatives to the resurrection, see the work of Gary Habermas. [5] Well, Gary has done a good job on those theories but N. T. Wright, I think, could have done much more in justifying why we think that the resurrection hypothesis is the preferred hypothesis. So I would strongly disagree with those who say Wright goes too far in his book in claiming to have established the resurrection of Jesus.

MR. HULL: Your debate with Bart Ehrman caused quite a stir. [6] Can you provide some background to that debate and also comment on his approach to the historicity of the resurrection? [7]

DR. CRAIG: I was delighted to have the opportunity to debate Bart Ehrman because his journey and my journey are so similar in certain respects. We both became Christians about the same time, we both attended Wheaton College, we both learned Greek from Gerald Hawthorne – the same professor at Wheaton -, we both went on to do doctoral work in theology. Then our paths parted. As I studied the historicity of the resurrection of Jesus with Wolfhart Pannenberg I became more convinced than ever of the historical credibility of that event. Bart Ehrman, on the other hand, lost his faith while in graduate school and has now become, I think, an agnostic basically as well as a non-Christian. So I was very eager to debate him on whether or not there was historical evidence for the resurrection of Jesus.

The thing that struck me about his central argument that is repeated over and over again in his books is that it is a rehearsal of David Hume’s argument against miracles in a very unsophisticated and clearly fallacious way. So I called these Ehrman’s Egregious Error and Bart’s Blunder for two statements of this error in his writings. It showed that he is just hopelessly confused when it comes to probability theory and the problems involved. He claims that you cannot establish the resurrection because it is so improbable that no amount of evidence could prove it to be true and even claims there is a contradiction in saying the most probable event is the most improbable event. What I show in the debate is just his lack of understanding of confirmation theory and probability theory.

One misunderstanding, I think, that arose as a result of the debate, particularly on his part, is that he seemed to think that I was trying to use the probability calculus to prove either the existence of God or the resurrection of Jesus. And that wasn’t my approach. That is Richard Swinburne’s approach. Swinburne uses the probability calculus to try to prove the resurrection of Jesus. I think that that is probably not the best approach because some of the probabilities involved are inscrutable. For example, what is the probability of the resurrection on the background information alone? Just given our general knowledge of the world, what is the probability that God would raise Jesus from the dead? Well, I don’t know. I don’t know how to assess that. I don’t think it is low but I don’t know what it is. So I think it is best to say this is an inscrutable probability.

And historians generally don’t work with Bayes Theorem or probability calculus. Instead, their approach is the approach that I take in my own work on the resurrection – you use certain criteria for assessing historical hypotheses like explanatory power, explanatory scope, plausibility, etc., etc. And you assess the explanation or the hypothesis in terms of how well it meets those criteria. That is the basis on which I would argue for the resurrection of Jesus.

In the debate, Ehrman did not give his own views until his final rebuttal in which it was too late for me to respond. What he said was that the disciples began to read the Scriptures and this convinced them that Jesus must have been risen from the dead and therefore they had these visions and so forth. The problem with that is that is the old obsolete theory of the origin of belief in the resurrection that stems from Bultmann and others in the mid-20th century that has now been widely rejected. It has been widely recognized that the disciples’ faith in Jesus would not have led them to think that he was risen from the dead. This is in fact quite contrary to Jewish belief. Moreover, there was no idea in Judaism of a dying Messiah, a Messiah who would be humiliatingly executed by his enemies rather than triumphing over them. So, even if they had continued to regard Jesus as a Jewish martyr and perhaps believed that God would vindicate him someday and maybe they would see visions of him in heaven, that would at best lead them to preserve his tomb as a shrine where his bones could reside until the eschatological resurrection when they and all the righteous dead of Israel would be reunited in the Kingdom of God. But they wouldn’t come up with this outlandish and un-Jewish idea that he was somehow already risen from the dead. So I think that the theory that he suggested is one that is just quite implausible and I don’t think there was enough time for that to happen in any case.

John Dominic Crossan believes that much of the passion narrative is what he called historicized prophesy. That is to say they read these prophesies in the Old Testament and then they made up the details of the passion story to kind of fit what the prophesies say. [8] But Crossan himself says it would take them at least ten years to develop the passion story based on these prophetic motifs. He says beyond that there aren’t any prophetic motifs for developing the resurrection stories; there is nothing in the Old Testament that you could use. Well, even given Crossan’s dating, ten years is already too late because the material from 1 Corinthians 15:3-5 goes back to within the first five years after the crucifixion, maybe even within months of the crucifixion and it already includes the crucifixion, the burial, the resurrection of Jesus, and the postmortem appearances. So by Crossan’s own dating, they have come to believe in the resurrection of Jesus prior to even being able to find the motifs in the Old Testament to manufacture the passion not to speak of the resurrection.

So Ehrman’s theory, I think, is one that simply won’t hold water and I think represents an obsolete sort of scholarship that really is not the current cutting edge in the discussion today.

MR. HULL: Many associated with the Emerging Church think that postmodernity is a great blessing. Do you think it is a blessing or a bind? Is it a problem?

DR. CRAIG: I think that the Emerging Church represents a tiny backwater or back current in a larger tide that is moving in the quite opposite direction of a renaissance of Christian apologetics and Christian philosophy and biblical criticism in our day. I think that they are quite mistaken in thinking that we live today in a postmodern culture in which the traditional arguments and evidence for the existence of God no longer work. That is completely contrary to my experience and also, I think, is quite a misanalysis of our contemporary culture.

I would argue, in fact, that a postmodern culture is an impossibility – it is a myth. No one could live in a postmodern culture. No one is a postmodernist when it comes to reading the label on a box of rat poison and a bottle of aspirin. If you’ve got a headache, you better believe that texts have objective meaning. So there isn’t any such thing as a postmodern culture. We live in a culture which remains deeply modernist in its flavor.

People say, “But people are relativistic and pluralistic today.” They are relativistic and pluralistic about matters of religion and ethics but when it comes to science and technology and engineering people are not relativistic. So that is not postmodernism; what that is is modernism! That is old line verificationism and positivism which says anything you can’t verify by your five senses is just an expression of personal taste. Our culture remains at its core deeply modernist. Science and technology give us the truth about the way the world is; religion and ethics are just expressions of personal taste. That is what modernism says. That is not postmodernism.

In fact, I think that postmodernism is probably the most clever deception that Satan has yet devised. He says to us, “Modernism is dead. You don’t have to worry about it anymore. It is dead and gone, forget about it.” Meanwhile, modernism pretending to be dead comes back around in the guise of a fancy new masquerade called postmodernism. And Satan says, “Your old arguments and apologetic won’t work against this new challenger. Lay them aside. They no longer work. Just share your narrative.” So we voluntarily lay aside our best weapons of logic and argument and evidence and are tricked into succumbing to secularism and modernism in the guise of this postmodernist challenge. If we do that I think the result will be catastrophic for the church. What will happen is that the church’s voice will be reduced to just one voice in a cacophony of competing voices, none of them claiming to have the objective truth. All of them just sharing their narrative while scientific naturalism continues to tell us about the way the world really is.

So I don’t think that the insights of postmodernism are anything that the church should be appropriating in terms of laying aside traditional apologetics and argumentation and, again, I think that the general drift of the Christian movement today is toward a reaffirmation of the importance of apologetics and the life of the mind and intellectual engagement as Christians with our faith. [9]

MR. HULL: In recent times there has been more and more discussion about open theism. Is middle knowledge a way of responding to open theism? Can you elaborate about this for us?

DR. CRAIG: The theory of divine middle knowledge provides a rapprochement between the strong Reformed Calvinist doctrine of predestination and open theism which has a god who does not know future contingents and takes risks and gambles. What the theory of middle knowledge says is that God knows what any creature would freely do in any freedom permitting circumstances God might place him in. Thereby he can decree just certain circumstances with just those creatures in them knowing what they would freely do in those circumstances. Thereby he has complete knowledge of the future including future contingents. He does not determine what the creatures will do in those circumstances – they determine what they will do – but God simply knows how they would freely act in any circumstances he places them in.

So, on the theory of middle knowledge, God can have complete knowledge of the future including all contingents but without robbing us of human freedom. I find this theory to be a wonderfully fruitful understanding of divine omniscience that forges a middle path between the Reformed view, which seems to annihilate human freedom in the name of divine sovereignty, and the so-called open theist view which sacrifices divine sovereignty on the altar of human freedom.

MR. HULL: Regarding the kalam cosmological argument, it seems to fit very well with modern day cosmology. What are some of the distinctive features of this version of the argument?

DR. CRAIG: The kalam cosmological argument is distinctive in that it argues for a first cause of the universe in a temporal sense; not in the sense of rank as in, for example, Thomas Aquinas’ cosmological argument. Or not in the sense of a sufficient reason for the existence of the world even if the world is eternal as in Leibniz’s version. It is different from the customary cosmological argument that until recently has dominated modern discussions. It argues that the past cannot be infinite. The past is finite, there was a first event, and since something cannot come out of nothing there must have been a transcendent cause which brought the universe into being. So the argument can be very simply formulated as:

1. Whatever begins to exist has a cause.

2. The universe began to exist.

3. Therefore the universe has a cause.

Then you do a conceptual analysis of what it means to be a cause of the universe. A number of striking divine attributes will fall out of such a conceptual analysis.

MR. HULL: Can you comment on the two strands behind it. The idea of talking about infinity seems to be nonsense and the role of contemporary cosmology. These seem to be two supporting planks of argument.

DR. CRAIG: Pre-modern defenders of the kalam cosmological argument – such as John Philoponus, the Alexandrian theologian in whose thought this argument originated, as well as medieval Islamic scholars such as al-Ghazali and al-Kindi, Jewish thinkers like Saadia ben Gaon, and Christian thinkers like Bonaventure – defended the key premise in the argument (that the universe began to exist) on the basis of philological arguments against the possibility of an infinite regress of events. They would typically argue either that the idea of an actually infinite number of things cannot be instantiated in reality; that you cannot have an actually infinite multitude in reality and therefore the number of past events in the history of the universe has to be finite. Or alternatively they would argue that given an infinite past one could never traverse an infinite past to arrive at today. So if an infinite number of events had to elapse before the present moment could arrive, the present moment would never arrive. But obviously here we are, so the past must be only finite. So even prior to the advent of modern cosmology there was a robust philosophical tradition defending the finitude of the past and creation out of nothing that went right up through the medievals to a person like John Locke, for example, the father of British empiricism, and Immanuel Kant who, in The First Antinomy concerning time, argued in the thesis of that Antinomy for the finitude of the past and said that this was a rationally demonstrative argument. [10] So there is a very rich history of this argument up to the time of modern science.

What has happened in the last two centuries is that as a result of modern physics we now have good empirical reasons for thinking that the past is finite and that the universe had an absolute beginning. The first of these reasons began to emerge already in the 19th century with the statement of the second law of thermodynamics. What physicists realized was that because entropy increases irreversibly, that is to say, because the level of disorganization in the universe tends to increase rather than decrease with time, given enough time the universe would come to a state of equilibrium or heat death in which life would no longer exist. Now, if that will happen in a finite amount of time, the question obviously arises: if the universe has existed for infinite time why are we not now in a state of heat death already? This seemed to imply the conclusion that the universe must have begun to exist.

The advent of relativity theory has changed somewhat the eschatological scenario predicted on the basis of the second law but the fundamental question still remains the same. If, given enough time, the universe will become cold, dark, dilute, and dead, why are we not now in such a state if the universe has existed forever. This gives good grounds for thinking empirically that the past is finite and that the universe had an absolute beginning.

The other empirical confirmation came with the enunciation of Albert Einstein’s General Theory of Relativity and his application of that theory to the universe as a whole – to cosmology. What it was discovered is that when Einstein’s gravitational theory is applied to the universe as a whole, the universe is radically unstable. On the basis of this theory, Alexander Friedmann, a Russian mathematician, and George Lemaître, a Belgian astronomer, were both able to formulate, independently, models of the universe which predicted an expanding universe – that the universe is literally growing apart. The radical implication of these theories is that as you extrapolate the expansion back in time everything becomes denser and denser until finally at a point in the finite past the universe is contracted down to a point of infinite density which marks the boundary of space and time and beyond which nothing existed. So that it marks the literal inception of the universe out of nothing.

Well, subsequent to those purely theoretical formulations of the expanding model, the so-called Friedmann-Lemaître model of the universe, Edwin Hubble, the American astronomer, discovered the cosmic redshift of the galaxies which was the recessional expansion of the universe predicted by Friedmann and Lemaître on the basis of Einstein’s equations. Subsequent discoveries have continued to confirm this so-called Big Bang model of the universe.

So the empirical evidence for the expansion of the universe also indicates that the universe is finite in the past.

This second key premise of the kalam cosmological argument – that the universe began to exist – has both philosophical and scientific justification which I think render it more plausible than its negation, which is all you need for a good premise in an argument.

MR. HULL: I notice in a lot of your debating, one of the first things you talk about is why there is something rather than nothing.

DR. CRAIG: One wonders when one propounds an argument like this how effective it will be in public debate. I have been stunned, frankly, by how weak the responses are to this argument when I typically defend it in a public forum or in professional journals when I publish this argument. Philosophers seem pretty much at a loss as to what to say about the philosophical arguments for the beginning of the past and I think that the responses that have been offered are quite weak and ineffective. [11] As for the scientific evidence, well, although people will sometimes make gestures toward alternative cosmologies which don’t involve a beginning, what they usually fail to realize is that these cosmologies are largely discredited. In fact, the history of 20th century cosmology could be regarded as one failed attempt after another to avert the absolute beginning predicted by the standard Big Bang model. We’ve seen the steady state theory, oscillating theories, vacuum fluctuation models, inflationary theories, quantum gravity theories – they come and go and what is inevitably discovered is that either these models are mathematically inconsistent, or incomplete, or in contradiction to observation, or else they imply the very beginning of the universe that its proponents sought to avoid. So I have found that these arguments hold up extremely well in public exchange with opponents of natural theology.

MR. HULL: Russell Stannard, who is a physicist, has said – hold on a minute, if space and time start in a Big Bang, how can you say there can be a cause before the Big Bang because cause and effect is something that happens in a space-time universe.

DR. CRAIG: I don’t think that there is a difficulty in saying that there is a transcendent cause of the Big Bang that exists timelessly and spacelessly without the universe. What we must not equate is causal priority with chronological priority. There is nothing chronologically prior to the Big Bang because it is the beginning of time. But that doesn’t mean you can’t have something that is causally prior to the Big Bang so long as you don’t think of it as something that is prior to it in time. I would say that even theists who think that God exists, say, timelessly and the universe is a four-dimensional space-time block that co-exists with God as, for example, Russell Stannard does, recognize that there is a kind of causation that doesn’t involve temporal priority because on that view God exists timelessly and the universe exists in ontological dependence upon God in a kind of timeless, ontologically dependent state. So, if that is possible, there is no reason to think that God could not cause the first moment of the Big Bang as well as the universe as a whole.

Now my take on this actually is rather unusual. I am of the inclination to think that at the moment of the Big Bang, God enters into time. He exists timelessly without the universe but temporally at the moment of creation and subsequent to that. So the decision to create a universe on God’s part represents the decision to enter into time and be really related to the temporal world that he has made. So on the view I would defend, I would say that God’s causing the Big Bang is simultaneous with the Big Bang. The two occur at the same instant. I think this is intuitively evident. If God creates the universe at the moment of the Big Bang, then there couldn’t be any delay in the effect appearing later. It would appear at the same moment that he has created it. So the Big Bang, or the origin of the universe, just is God’s creating time and space and they are simultaneous, they occur co-incidentally.

MR. HULL: In his book, The Goldilocks Enigma, Paul Davies says you don’t even need a miracle to begin the universe in the first place. He appeals to two things. One is quantum cosmology and the second is inflation. Can you say a bit about these alternative models to First Cause Big Bang Cosmology?

DR. CRAIG: Some theorists like the Russian astrophysicist Andre Linde have attempted to avert the absolute beginning of the universe by postulating an inflationary model of the universe in which inflation never stops. Inflation begets inflation and further inflation. So the universe is a kind of self-replicating entity that goes on forever. What Linde tried to do was to extend this into the infinite past as well as into the infinite future – to say that inflation has been going on from eternity past. He did this deliberately in order to avoid the metaphysical problems of the beginning of the universe – how could the universe come to exist if it had an absolute beginning? But that attempt on the part of Linde in fact backfired. In 2003, three cosmologists, Arvind Borde, Alan Guth (the father of inflationary cosmology), and Alexander Vilenkin, were able to prove that any universe which is, on average, in a state of cosmic expansion cannot be infinite in the past but must have a boundary at some point in the finite past. [12] Therefore, inflationary universes cannot be past eternal. [13] They have to have a beginning. So the attempt to drive inflation back eternally into the past ultimately backfired and it turned out that the model implied the very beginning of the universe that Linde at first sought to avoid.

Other cosmologists, like Vilenkin himself, while acknowledging the absolute beginning of the universe have said we don’t need to have a transcendent cause of the universe; we can say that the universe quantum tunneled into being out of nothing. What Vilenkin means by that is that if you take a universe which is a small bubble with a finite radius of what is called false vacuum there is a certain probability within quantum theory that it will tunnel or turn into an inflating universe. As you shrink the radius of this little universe back down to zero there always remains some finite probability, however small, that it would tunnel into an inflating universe. What Vilenkin does is he equates that initial point of zero radius with nothingness and says therefore there is some probability that nothingness will tunnel into an inflating universe and that is how the universe originated.

In fact, cosmologists recognize that this initial point is not nothingness. It is not what we mean philosophically by nothingness. It is definitely something. If it were really nothing then the quantum function that he describes would only have one variable, namely the posterior variable. But quantum tunneling is a two-valued function – from one value to another value. Therefore it is always from something to something. It is not from nothing to something. Another way to see the point is to realize that to have a radius whose measure is zero is not the same thing as having no radius at all. Nothingness has no radius at all. It does not have a radius whose measure is zero. So Vilenkin’s attempt to explain the origin of the universe out of nothing turned out to be maladroit. It is not really an origin from nothing.

What that means, of course, is that therefore that initial state in his model still needs an explanation. You still need to explain: where did that initial state come from? It is not eternal. It hasn’t always existed. It came into being from nothing. Why does the universe exist? So in fact I think it is widely recognized that Vilenkin’s model is philosophically naive in its equation of this initial state with nothingness.

MR. HULL: I remember one physicist saying that there is a lot going on in a quantum vacuum. It is not like it’s nothing.

DR. CRAIG: Right, that is certainly true. That was a mistake made by certain other cosmologists who proposed the universe coming into being out of the vacuum. The vacuum is a sea of fluctuating energy governed by physical laws. It is definitely not what we mean by non-being. So these vacuum fluctuation models as well did not explain how the universe could come into being out of nothing.

MR. HULL: Can you tell us a bit about the idea of a multiverse?

DR. CRAIG: When one thinks about the hypothesis of a multiverse in order to rescue the alternative of chance as an explanation for the fine-tuning of the universe, what we have here is really an attempt to multiply your probabilistic resources without really any evidence for it. It is an attempt to say that our probabilistic resources are much, much greater than we realize so that somewhere by chance alone a finely tuned universe will exist. So the hypothesis is that there is an infinite world ensemble of randomly ordered universes that exists out there and we are just one random universe in this ensemble that happens to be fine-tuned by chance for the existence of intelligent life.

The very fact that otherwise sober scientists are driven to adopt a metaphysical hypothesis like an infinite, randomly ordered world ensemble is, I think, a kind of backhanded compliment to the power of the design argument. Scientists would not be opting for metaphysical hypotheses like this unless they felt absolutely forced to. So the very fact that they have to recur to the multiverse hypothesis (and this is where the debate lies today) I think is testimony to the strength of the design argument. [14] 

The multiverse hypothesis, however, faces two very formidable objections that I think make it less plausible than the hypothesis of intelligent design. The first one is that there is no evidence that a world ensemble exists. There is no evidence that there is an ensemble of an infinite number of randomly ordered worlds out there. This is a metaphysical hypothesis. The best possibility for generating a world ensemble would be inflationary cosmology. According to inflationary cosmology, the universe is a sea of expanding false vacuum and there form within this false vacuum bubbles of true vacuum which are, as it were, mini-universes. These can be ordered according to different constants and quantities by chance. So you could get a multiverse in that way if you have such a correctly arranged inflationary scenario. No one knows if this is true or not but that would provide some hope for thinking that you could get a multiverse. The problem here is, however, that the Borde-Guth-Vilenkin theorem that I described earlier also applies to the multiverse as well as to the universe. It cannot be extended infinitely into the past. It can only be finite in the past. What that means is that there has only been a finite time since the origin of the multiverse and therefore we have no guarantee whatsoever that enough bubble universes have been generated by now to produce, by chance alone, enough finely tuned universes such as ours to make our existence possible simply by chance. There is no guarantee of that at all. There is no guarantee or evidence that a world ensemble of an infinite number of randomly ordered universes exist. In fact, they cannot be infinite given the finitude of the past of the multiverse.

The second problem is that if we are just a random member of a world ensemble of randomly ordered worlds, then it is incomprehensibly more probable that we should be observing a much different universe than we do. This point has been made very forcefully by Roger Penrose at Oxford University. Penrose points out that the chances of our solar system forming instantaneously by just the random collision of particles is about 1 chance out of 1060. Now that is an inconceivable number. But it is tiny in comparison to the chances of a finely tuned universe like ours existing by chance alone. For example, the odds against the low entropy condition in the universe existing by chance alone is one chance out of 1010(123). Penrose says the improbability of our solar system just appearing instantaneously by the random collision of particles is “utter chicken feed” in comparison to the improbability of this low entropy condition of the universe obtaining by chance alone. [15]

So what that means is that if we are just a random world in this world ensemble of randomly ordered universes, it is incomprehensibly more probable that we should be observing a much smaller universe – one that just formed instantaneously by the random collision of particles because that is vastly more probable than the chances of a finely tuned universe occurring. One way to see this would be to simply realize that in the world ensemble the range of universes in which life exists in these tiny solar systems is just vastly, vastly greater than the life containing universes that are fine-tuned. So if a random dart were thrown at the cosmic landscape, for example, and it were to hit a life permitting universe, it is incomprehensibly more probable that that life permitting universe would be a little tiny solar system that just formed by the random collision of particles.

Similarly, if we were just one member of a world ensemble, we ought to observe highly extraordinary and improbable events like horses popping into being out of nothing by random collision of particles, or perpetual motion machines, or pink rabbits wearing bow ties popping into being because those kinds of worlds are vastly more probable than a finely tuned universe.

So the conclusion is, I think, that if atheism is true at least then it is highly, highly improbable that we live in a world ensemble. Our observations highly disconfirm the idea of a multiverse. So really the only way, I think, to make the multiverse hypothesis salvageable would be by theism – to say that God has created a multiverse. But on naturalism, at least, it is highly improbable, I think, that we are simply a random member of a world ensemble. [16]

  • [1]


  • [2]

    John Earman, Hume’s Abject Failure: The Argument Against Miracles (Oxford University Press, 2000).

  • [3]

     N. T. Wright, Christian Origins and the Question of God, III: The Resurrection of the Son of God (Minneapolis: Fortress Press, 2003), p. 710.

  • [4]


  • [5]

    In Wright’s book The Resurrection of the Son of God, on page 706 in footnote 61 he says, “For a survey of ‘naturalistic’ theories, and a response, cf. e.g. Habermas 2001 [‘The Late Twentieth-Century Resurgence of Naturalistic Responses to Jesus’ Resurrection,’ Trinity Journal n.s. 22:179-96].” To read a copy of Habermas’ article, see (accessed October 3, 2013).

  • [6]

    To watch a video of this debate, see (accessed October 3, 2013).

  • [7]


  • [8]


  • [9]


  • [10]


  • [11]


  • [12]

    See (accessed October 4, 2013).

  • [13]


  • [14]


  • [15]

    Roger Penrose, The Road to Reality: A Complete Guide to the Laws of the Universe (New York: Alfred A. Knopf, 2005), pp. 762-765.

  • [16]

    Total Running Time: 50:43 (Copyright 2007 © William Lane Craig)