William Lane Craig CV


Existence of God (part 17)

Transcript of William Lane Craig's Defenders 2 class.

Excursus: Natural Theology
ยง III. Teleological Argument
Lecture 4

We have been talking about the fine tuning of the universe for intelligent life, and we saw that there are three possible explanations for the fine tuning of the cosmos: either physical necessity, chance or design.

We saw that physical necessity is highly improbable because the values of these constants and arbitrary physical quantities are independent of the laws of nature. The laws of nature are consistent with a wide range of values for these constants and quantities. There is not any physical necessity that the universe should exhibit this kind of fine tuning.

Then we began to look at the hypothesis of chance. We argued that the odds of all of these constants and quantities’ falling, by chance alone, into this incomprehensibly narrow life-permitting range are so infinitesimal that these odds cannot be reasonably faced. In response to that, you will remember, we encountered a couple of objections to try to save the chance hypothesis. For example, the appeal to a lottery in which all of the tickets are sold and yet someone has to win, no matter how improbable that might be. We saw why that analogy was flawed. The proper analogy would be a lottery in which billions and billions of white balls are mixed together with one black ball, and you are told that unless the black ball is picked, you are going to be shot and killed. But then, lo and behold, a random ball is picked, and it turns out to be the black one, and you are allowed to live. You ought to think that the lottery was rigged. Why? – not because the black ball is any more improbable than any other particular white ball. Each ball is equally improbable. – but because it is overwhelmingly, incomprehensibly more probable that, whichever ball is picked, it will be white rather than black. Similarly, our universe is no more improbable than any other universe. But it is overwhelmingly more probable that whichever universe exists, it will be life-prohibiting rather than life-permitting. We still need some sort of explanation as to why a life-permitting universe exists against all probability to the contrary.

Anthropic Principle

Some people have argued that no explanation is needed for why we observe a life-permitting universe such as we do. They say that we can observe only a life-permitting universe because if the values of the constants were not life-permitting, we couldn’t be here to observe it! We can observe only life-permitting universes because the ones that aren’t life-permitting are inconsistent with our existence. This is the Anthropic Principle. The Anthropic Principle says that we can observe only properties of the universe that are consistent with our existence. That is obviously true. It is necessarily true – if the fundamental properties of the universe were inconsistent or incompatible with our existence, obviously we couldn’t be here to observe it! So the only kind of universe we can observe is a life-permitting universe. That is obvious. That is the only kind of universe we can observe. From this obvious truth, they infer that therefore no explanation is needed for why a life-permitting universe exists.

This reasoning is logically fallacious. From the fact that we can observe only life-permitting universes, it does not follow that no explanation is needed for why there is a life-permitting universe – why it does exist.

Again, an illustration can help us begin to see the point. Imagine that you are traveling abroad in some third world country and you are arrested on trumped-up drug charges and convicted, and you are dragged in front of a firing squad of 100 trained marksmen, all with rifles aimed at your heart to be executed. And you hear the command given, “Ready! Aim! Fire!” You hear the roar of the guns! And then – you observe that you are still alive! All of the 100 marksmen missed! Now what would you conclude?1 Would you say to yourself, “I guess I really shouldn’t be surprised that they all missed. After all, if they hadn’t all missed, I wouldn’t be here to be surprised about it! Given that I am here, I should have expected them all to miss”? Of course not! You would immediately suspect that they all missed on purpose and that the whole thing was a setup, engineered by someone for some reason. From the fact that you shouldn’t be surprised that you don’t observe that you are dead, it doesn’t follow that you shouldn’t be surprised that you do observe that you are alive. Do you see the point? You should not be surprised that you don’t observe that you are dead. You can’t observe that. You should not be surprised that you do not observe you are dead. But it doesn’t follow from that that you shouldn’t be surprised that you do observe that you are alive, in light of the enormous improbability of their all missing by accident. Given that enormous improbability, you should be very surprised that you are alive. In fact, you would probably think that the whole thing was designed – that the reason the marksmen all missed was because it was intentional, and this was brought about not by chance, but by design.


Question: It seems like there are two possibilities. You could say it was designed or there was an infinity of other people in front of firing squads, and one of them was missed by all marksmen.

Answer: Very good! That is exactly right. What you could say instead is, it is not that they all missed on purpose; it is that, unbeknownst to me, there is an infinite number of other people standing in front of firing squads being shot at, and by chance alone somewhere in that ensemble of firing squads, all of the 100 marksmen would miss by accident. And for the person to whom that happens, he shouldn’t be surprised at being alive, it was just by chance. This other explanation is that they all missed by chance because there is an infinite number of other firing squads that I don’t know about and can’t see. That would be the other explanation. That will form a nice segue to the Many Worlds hypothesis to explain away fine tuning. But it is a helpful illustration because I think we see how absurd that would be for someone to conclude such a thing. Clearly, the better explanation would be to think that the marksmen all missed on purpose than to think that you are just one of an infinite ensemble of other victims in front of firing squads.

Question: If you expand your probabilistic resources in this way, then you can explain anything away, and it becomes ridiculous.

Answer: I do think that is a good point. If you cannot get, say, double-sixes on one throw of the dice, then you can imagine 25 throws of the dice, or 100 throws of the dice. And then you are probably going to get double-sixes at some point. You have expanded your probabilistic resources in order to explain the chance result. The point that you are making is, if you expand your probabilistic resources by saying, “This isn’t the only universe there is. There are all these other parallel universes that we cannot see where things are going on,” then you can explain virtually anything away. Somebody playing cards, who gets four aces every time he deals, could reply, “Well, there is an infinite number of poker games going on in this World Ensemble, and in some of them, every time I deal, I get four aces! What are you complaining about?” No protest could be made because you can explain anything by chance. And yet that would be absurd. If you were sitting at the table playing with that guy, would you sit down for another round of cards? I should hope not! You wouldn’t make any life decision based on this kind of reasoning. It would make rational behavior literally impossible because everything could be attributed to chance.

Question: I bring this up because that is what some physicists are saying.

Answer: That is exactly right and forms a nice segue to our next point.2

Anthropic Principle With a World Ensemble

Theorists have come to recognize that the use of the Anthropic Principle cannot eliminate the need for an explanation of the fine tuning unless you posit this World Ensemble of parallel universes which are real and just as actual as this one and which are randomly ordered in the values of their constants and quantities. According to the World Ensemble hypothesis, or Many Worlds hypothesis, our universe is just one member of a World Ensemble, preferably an infinite World Ensemble, of other universes, and these are all randomly ordered in their constants and quantities, so that by chance alone somewhere in the World Ensemble one or more life-permitting, finely tuned worlds will exist. Since only the finely tuned universes have observers in them (the other ones are life-prohibiting, so it is only the finely tuned worlds that have observers in them), the observers in the World Ensemble will naturally observe their universe to be finely tuned. So there would not need to be any explanation of this. You would not need to appeal to design to explain fine tuning; it is purely due to chance.


Question: The first thing that comes to my mind is the skeptic apparently does not present any mechanism for having all of these universes. There is one popular idea where you have multiple universes because every time a choice is made, both ways actually cause universes to split off.

Answer: You are raising a question on the mechanisms. I will comment on that in a moment.

Question: Can you distinguish between many worlds and multiverse, or are they the same?

Answer: I am using them as synonyms. The Many Worlds hypothesis takes different forms. This one that the first question just discussed is the Many Worlds Interpretation of quantum physics, which says that every time a quantum measurement is made by a classical apparatus, the universe splits in two and the correlated values of the measurement both exist, and the universe sort of forks. It is a fantastic interpretation of reality to think that, as one person put it, a mouse, by observing something, could split the universe. That would be one. Another interpretation would be inflationary theories of the universe where our universe is just a bubble of true vacuum in a broader false vacuum that is expanding at fantastic speed, and within this false vacuum there are lots of other bubbles, too. That is another way the multiverse scenario plays itself out. That is probably the most popular today. But then there are others as well, as to how you might have many worlds. Dawkins, in The God Delusion, suggests an oscillating theory of the universe, where the universe expands and contracts repeatedly from eternity past, and each time the constants and quantities are recycled and get new values. That would be another form of the Many Worlds Hypothesis. I am not distinguishing between multiverse and other World Ensemble theories. They are all basically trying, as was said, to multiply our probabilistic resources. If we cannot get it in one throw of the dice, we will posit an infinite number of throws, and then you are sure to get your chance result.3

Question: We talk about having a possible universe where it is life-permitting. Are we being gracious that that is necessarily a certainty that life would develop, or is there a conditional probability within that where you have life-permitting, but not necessarily the development of life?

Answer: The latter. We are saying that these constants and quantities which are finely tuned are merely prerequisites for life, but they are not sufficient conditions for life. You would still need to have a planet that is situated at a certain distance from its star, it would need to have a moon to affect the tides, you would need a Jupiter in the solar system to serve as a vacuum cleaner to suck in asteroids and comets that would otherwise bombard the planet and destroy life. There are all kinds of conditions that would need to be fulfilled as well in order to have a life-permitting planet. But here we are talking just about these bare initial conditions that would be prerequisites for life. That is why theorists use the language of “life-permitting” and “life-prohibiting.” They are not, by any means, “life-guaranteeing.”

I want to pause to reflect on what is happening in this debate. The current debate over the fine tuning of the universe has become a debate over the Many Worlds Hypothesis. This is, in fact, the cutting edge of the discussion today. This is at the heart of the discussion of fine tuning. In order to explain the fine tuning of the universe, we are being asked to believe not merely that there are other universes, unobservable by us, but that there is an infinite number of other universes and, moreover, that they are all randomly ordered in their constants and quantities, so as to guarantee by chance alone that some of them will be finely tuned. All of this is needed to guarantee that a life-permitting world will appear somewhere by chance in the ensemble.

When you think about it, this is really extraordinary. Otherwise hard-headed, sober scientists would not be flocking to embrace so metaphysical a hypothesis unless they felt absolutely compelled to do so. This recourse to the World Ensemble is a kind of backhanded compliment to the power of the design argument. It shows the lengths that people will go to in order to avoid a Designer of the universe. The odds against a life-permitting universe are so incomprehensibly great that in order to explain it away, you have got to recur to metaphysics and do the postulation of this World Ensemble in order to save the chance hypothesis. If somebody says to you, “Well, it could have happened by chance!” or “It was just dumb luck!”, then what you should say to them in response is, “If that is the case, then why is it that scientists feel compelled to embrace the Many Worlds Hypothesis in order to explain it away?” They would not be embracing so extravagant a metaphysical hypothesis unless they felt compelled to do so. So this very alternative is itself a testament to the fact that chance alone won’t explain the fine tuning. The odds against it are just too great unless you appeal to the World Ensemble in order to guarantee that the chances will be fulfilled somewhere.

How might one respond to the Many Worlds Hypothesis? We might just leave it there and say, “Take your pick: a divine Designer or else a Many Worlds Hypothesis. It is up to you to think which is more plausible.” But we can actually say more than that. Let me add three comments by way of response to the Many Worlds Hypothesis.

One way to respond to the Many Worlds hypothesis would be to show that the multiverse itself also involves fine tuning. In order to be scientifically credible, the Many Worlds Hypothesis has to posit some sort of mechanism for generating these many worlds. But this mechanism had better not be fine tuned itself, or all you have done is kicked the problem upstairs! You still have the same problem. Like a stubborn bump in the carpet, you depress it in one point only to have it pop up somewhere else.4 If the fine tuning is to be explained away by the Many Worlds Hypothesis, the mechanism that generates the many worlds had better not be fine tuned. If it is, the problem still persists and arises all over again. The proposed mechanisms for generating the World Ensemble are, frankly, so vague that it is far from evident that the physics governing the multiverse will be free of fine tuning. These explanations pretty much amount to hand waving at this point, and so there is no way of knowing that they are going to be free from fine tuning. For example, if M-Theory is the physics of the multiverse, then it remains unexplained why there exists only and exactly 11 dimensions. The theory itself does not explain why that particular number of dimensions exist. So fine tuning still exists in the theory of the multiverse. The mechanisms that actualize all of the various possibilities permitted by M-Theory in the cosmic landscape – that 10500 possible universes – also may well involve fine tuning. There are some pretty specific conditions that have to be met in order for this to work. Merely postulating a World Ensemble is not in and of itself enough to explain away fine tuning because it is not at all clear that the World Ensemble itself doesn’t involve fine tuning, and then you haven’t solved anything.

Here is the second response. A good many theorists today are quite skeptical of the Many Worlds Hypothesis itself. Why, after all, think that a World Ensemble of universes actually exists? Is there any independent evidence, apart from fine tuning, for the existence of such a thing? Remember when we talked about the Kalam Cosmological Argument, we saw that the theorem developed by Borde-Guth-Vilenkin in 2003 applies to the multiverse as a whole. This theorem shows that even a multiverse of expanding bubble universes cannot be infinite in the past but must have a beginning. If that is the case, that means that the mechanisms that generate the many worlds has been chugging away for only a finite amount of time. Therefore, it may well have only generated a finite number of universes in the World Ensemble, in which case there is no guarantee at all that a finely tuned universe will have come to exist yet in the World Ensemble. There is really no evidence that an Ensemble, which is required by the Many Worlds Hypothesis, actually exists. By contrast with that, we do have independent reasons for thinking that a Designer of the cosmos exists. Remember al-Ghazali’s argument for the beginning of the universe and the need for a transcendent Creator who brought the universe into being, including all of the laws of nature. Remember Leibniz’s argument for a Sufficient Reason for the existence of the universe, a reason which, again, must be found in a transcendent, personal Mind who is the ground of being for why there is something rather than nothing. So we have independent reasons already to think that there is a Designer of the universe but no independent reasons for thinking that a World Ensemble exists.

Finally, third, the Many Worlds Hypothesis faces what may well be a devastating objection. You remember that when we talked about the thermodynamic properties of the universe, we talked about Boltzmann’s Many Worlds Hypothesis? Boltzmann thought that the universe as a whole exists in a state of equilibrium, but throughout the universe there will be little pockets of disequilibrium that occur by chance alone, and we are just one of those pockets of disequilibrium in a sea of equilibrium.5 And remember what sank Boltzmann’s hypothesis was that if our world is just a random member of a World Ensemble, then it is vastly more probable that we should be observing a much tinier region of order than we do? A tiny region of disequilibrium is unfathomably more probable then the vast universe we see, and yet it would be sufficient for our existence. Therefore, Boltzmann’s hypothesis has been rejected by contemporary physics. Well, it turns out that an exactly parallel problem to Boltzmann’s Many Worlds Hypothesis also faces the modern Many Worlds Hypothesis as an explanation of cosmic fine tuning.

This objection has been very forcefully stated by Roger Penrose of Oxford University, for example, in his book, The Road To Reality. Penrose points out that the odds of our universe’s initial low entropy condition are just incomprehensibly small – one chance out of 1010(123). This is an incomprehensible number. By contrast, he says the odds of our solar system’s just falling together by accident by a random collision of particles (imagine particles swarming around, and all of a sudden they just fall together to form our solar system!) would be about one chance out of 1010(60). That is a huge number, obviously, but nevertheless, Penrose says this is “utter chicken feed” in comparison with 1010(123). What does that mean? It means that if we are just a random member of a World Ensemble, then we should be observing an [orderly] universe no larger than our solar system. Why? Because this sort of universe is vastly, vastly more probable than one that is finely tuned for our existence. There are simply many, many more observable universes in the World Ensemble which are small, like the size of our solar system, than which are large like the finely tuned universe that we observe. If we are just a random member of the World Ensemble, we should be observing an orderly world no larger than, say, our solar system. That is unfathomably more probable than a finely tuned universe because most of the observable universes in the World Ensemble are these small ones.

In fact, we can carry the argument farther. We actually wind up with the same sort of illusionism that plagued Boltzmann’s hypothesis. You see, a small [orderly] universe with the illusion of a wider universe is vastly more probable than a real, wider universe. Carried to its logical extreme, this has led to what some theorists have called, in language reminiscent of grade-B horror movies of the 1950s, “the invasion of the Boltzmann Brains.” What is a Boltzmann Brain? A Boltzmann Brain is a brain that just pops into being through a thermal fluctuation in the equilibrium. Like the solar system’s falling together by chance, it is vastly more probable than 1: 1010(60) that just a single brain would fall together by chance with an illusion of the external world that we observe. That is vastly more probable! The vast majority of observable universes in the World Ensemble will be worlds that just have a Boltzmann Brain in them. If you are going to accept the Many Worlds Hypothesis to explain fine tuning, you are rationally obligated to believe that you are a Boltzmann Brain and that your friends, this room, the external world, are all illusions of your projected consciousness. Well, no sane person believes that he is a Boltzmann Brain. Therefore, on atheism, it is highly improbable that there exists a randomly ordered World Ensemble. Given atheism, it is vastly more probable that there is no World Ensemble. The fact that we observe this wide, orderly universe disconfirms the Many Worlds hypothesis.

Ironically, the best hope for the partisans of a multiverse theory is theism.6 This is because, if God exists, he could create many worlds and plan them to have life in them according to his design. They would not be randomly ordered – they would be designed. God could give preference to worlds which are cosmically fine tuned. So in order to be rationally acceptable, it turns out that the Many Worlds Hypothesis requires God. I think that if you are going to be a Many Worlds theorist, you need to be a theist. Otherwise the evidence heavily disconfirms the multiverse hypothesis.

With the failure of the Many World Hypothesis, the last ring of defense for the hypothesis of chance falls away. Neither physical necessity nor chance seems to give a good, plausible explanation for the fine tuning of the universe. The question will be, does design fare any better? Is the design hypothesis equally implausible or is it superior as an explanation? That will be covered next.


Question: That last sub-point about it being a better theistic worldview. Can you explain that further?

Answer: The idea there is that the finely tuned universes would not appear in the Ensemble by chance, but it would rather be that they are designed by God. So God makes a World Ensemble in which he creates worlds that would not result by chance, and therefore you could have these finely tuned worlds among them, and it wouldn’t simply be a matter of randomness.

Question: Would the multiverse theory fall into the category of the gambler’s fallacy?

Answer: The question here is whether or not this commits what is called an “inverse gambler’s fallacy.” [According to the gambler’s fallacy], because I have won at roulette, I therefore conclude that there must be a lot of other roulette wheels spinning in other rooms, and that is why I can explain my winning by chance alone. The inverse gambler’s fallacy is concluding that because there are many other roulette wheels spinning, that explains why the roulette wheel came up with my number. I do not think that it commits an inverse gambler’s fallacy because of what I said last week. We are not trying to explain why this universe exists, or why this particular constellation of constants and quantities exist. What we are just trying to explain is why any life-permitting universe exists and not why this particular one does. This one is just as improbable as any other one; but what is more probable is that a life-prohibiting world rather than a life-permitting world would exist. So I do not think it commits that fallacy. But that is a very good question you are raising.7


1 5:04

2 10:11

3 15:07

4 20:05

5 24:59

6 30:00

7 Total Running Time: 33:35