String Theory, the Multiverse, and God

In one episode of the television show, “The Big Bang Theory,” Cal Tech string theorist Sheldon Cooper is visited by his mother Mary, who invites him to come on a Christian cruise called the “born-again boat ride.” “Uh, well, Mom, if I did, it would be conclusive proof that your God can work miracles,” Sheldon replies.

Sheldon’s rejection of theism is perhaps indicative of popular culture’s attitude toward science and faith. The concept of a divine legislator, which motivated so many of the pioneers of science to expect law and regularity in nature, has largely been forgotten. Instead, most modern western people see theism as an outdated model that will eventually collapse in the face of scientific progress. And string theory, in particular, is sometimes seen as the weapon that will deal God his deathblow. As the best candidate for a “theory of everything,” string theory promises a framework for physics at all energy scales (removing the role of God, perhaps?). And the string theory landscape suggests the possibility of many, many universes besides our own (explaining the apparent design of the universe, perhaps?). It might seem that God and string theory, like Harry and Voldemort, cannot live while the other survives.

In this article, we will focus on this last theological issue—the possibility that many universes, or a “multiverse,” could exist. But before we do, we must first understand why the issue is relevant to theology in the first place. And for this, we need to understand the so-called “fine-tuning” of the universe for intelligent life. When one looks carefully at the values of the quantities and constants that appear in the laws of physics that govern our universe, it becomes clear that life-permitting universes are incredibly rare in the sea of possible universes. If some of these quantities were changed by as little as 1 part in 10^5, 1 part in 10^11, 1 part in 10^120, etc., our universe would have been unable to produce life, and may very well have self-destructed long ago.1 When we look at those odds, it is safe to conclude that the lifepermitting universe in which we live was not produced by random chance.

Physicist Victor Stenger recently published a book called The Fallacy of Fine-Tuning: Why the Universe is Not Designed for Us.2 In it, Stenger claims that “the most commonly cited examples of apparent fine-tuning can be readily explained by the application of a little well-established physics and cosmology.” This, unfortunately for Stenger, is false. Space will prevent us from offering a sufficient rebuttal here, so suffice it to say that the overwhelming consensus of top particle theorists and cosmologists from all theological persuasions, both here at Harvard and globally, is that fine-tuning is a serious problem that will not easily be explained away.3 It is clear that the known laws of physics cannot explain fine-tuning.

It is possible in principle that future advances in physics could solve the problem, but this solution is starting to look extremely unlikely. If our universe is not actually fine-tuned for life, it is highly probable that we should have started to see the effects of the correct solutions already at experimentally observable scales. The most reasonable solutions to the hierarchy problem, for instance, have been already ruled out by the recent data coming from the Large Hadron Collider (LHC), and the most reasonable solutions to the cosmological constant problem were ruled out even earlier. When we look at some of the less famous fine-tuning problems, the situation becomes even worse. If the universe is somehow not fine-tuned, it was assembled in a way to fool us into thinking that it was.

So, how does the non-theist go about explain this fine-tuning without resorting to any sort of cosmic designer? The answer given by just about all of the non-theistic physicists I know is the “multiverse hypothesis.” The multiverse hypothesis holds that there exist an enormous, possibly infinite, number of universes like ours, each of which has slightly different values for the constants and quantities that appear fine-tuned in our own universe. In this vast ensemble of universes, the reasoning goes, certainly some universe’s constants must fall into a life-permitting range. And so, because we can only observe a universe if we are alive in it, we necessarily find ourselves living in this life-permitting universe rather than one of the life-prohibiting ones.

If this sounds more like science fiction to you than science, you are not alone. Perhaps the largest problem facing the multiverse is that it is completely untestable. And, if the other universes are really out of causal contact with our own universe, then the hypothesis will remain forever untestable. There will be no way for us to detect these universes by any experiment. And so, for the non-theist who wants to deny the existence of God because of a perceived scarcity of experimental evidence, the multiverse has to be very tough to swallow. Belief in the multiverse requires a leap of faith beyond science and into the realm of metaphysics.

Aside from experimental evidence, the next best thing a multiverse proponent could hope for would be a multiverse creating mechanism that pops out fairly naturally from known laws of physics. However, contrary to what some physicists might be claiming,4 even this is a fantasy at the moment. No testable physical theory predicts a multiverse. Period.

No physical theory that will become testable in the foreseeable future predicts a multiverse. Period. But if we want to get into untestable physical theories, then there is indeed a chance. And this is where string theory comes in.

Mathematical consistency of string theory requires a universe that has nine dimensions of space and one of time. Our universe, however, appears to have only three dimensions of space and one of time. Where did the other six spatial dimensions go? The answer is that they must be so tiny we cannot see them. For the sake of analogy, consider a garden hose. From far away, it looks one-dimensional, like a line. But to an ant on its surface, the hose looks two dimensional, as the ant can walk either along the hose or around its circumference. String theory suggests that the six extra dimensions should be curled up just like the circular dimension of the garden hose.

This is where the twist comes in: there is no unique way to curl up the extra six dimensions of string theory. Rather there are something like 10^500 different ways to get a 4-spacetime-dimensional theory out of a 10-spacetime-dimensional one. And so, assuming that string theory is true, it appears that there are at least 10^500 possible universes like ours, with slightly different physics in each one. This collection of universes is often referred to as the “string landscape.”

Does this mean that string theory predicts a multiverse? Here, we must be careful. The many different universes of the string landscape are, as far as our study of string theory has told us, mere possibilities. They need not be realized in any physical way. However, there are a few ways in which the string landscape suggests a multiverse. First of all, the landscape indicates strongly that there are mathematically consistent and logically possible universes that are not life permitting. Throughout the years, a handful of atheistic philosophers have tried to get around the apparent design of the universe by arguing that the laws of nature that describe it are necessarily fixed to be what they are. In other words, they would argue, the physics of our universe could not have possibly been different. String theory has shown this claim to be false beyond a reasonable doubt. The landscape indicates that there are plenty of logically possible and mathematically consistent universes besides our own.

Secondly, the string landscape seems like exactly the sort of thing we would expect to find if indeed a multiverse does exist.5 This is not an overwhelming argument in favor of a multiverse, but in the absence of a cosmic designer that prefers a universe with life to one without life, it is difficult to imagine why our universe should be the only member of the landscape that is actually realized in nature.6

Thirdly, and most importantly, there do exist proposed mechanisms for turning the possible universes of the landscape into actual, physical entities. A particular theory of the early universe called eternal inflation predicts a scenario in which some regions of space expand indefinitely, growing into “bubble universes” that would each give life to a member of the landscape.7

So, given both eternal inflation and string theory, we would indeed expect a multiverse to exist. So, there are reasons to believe in a multiverse. It is not as crazy of an idea as it might seem at first. Nevertheless, there are several reasons to be skeptical. Before we get into these, however, I want to mention two commonly made arguments against the multiverse that I do not find compelling.

The first bad argument against the multiverse is, as far as I can tell, a simple misunderstanding of conditional probabilities. The argument proceeds by analogy: suppose you are playing a game of poker, and your opponent deals himself four aces three times in a row. By the third time, you realize that he must be cheating, and draw your pistol. “Hold on a minute,” says your opponent, “In this infinite ensemble of worlds, there is an infinite number of poker games going on right now. And in some of these universes I am bound to deal myself four aces three times in a row. We just happen to live in one in which I do.” The argument, then, is that this reasoning certainly wouldn’t be enough to convince you not to shoot the guy, so therefore you should not follow similar reasoning when it comes to explaining how our universe happens to be the one in a zillion in which life exists.

The problem with this reasoning is that it fails to take into account the fact that we cannot possibly observe a universe that is not life-permitting, for if our universe did not permit life, then there would be no one around to observe it. In other words, the conditional probability that our universe permits life given that we live in it is 1. On the other hand, the probability that your opponent has dealt himself four aces three times in a row given that we happen to live in this universe is still extremely low. The analogy fails.

The other argument against the multiverse that I find unconvincing is an appeal to Occam’s razor: it is absurd, some would argue, to hypothesize an infinite number of other universes just to explain our own. It is simplest to assume that only one universe exists. Incidentally, atheists will often say the same thing about God, claiming that it is simpler to assume that just the natural universe exists rather than postulate a complicated entity like God to explain fine-tuning.8 The problem with both of these arguments is that Occam’s razor does not say that the simplest idea is usually the right one— it says that the simplest explanation is usually the right one. And if we rule out the multiverse and design, then we are stuck without any sort of explanation for fine-tuning. The problem of fine-tuning is not one that can be ignored, and whether we like it or not, the best proposed solutions to it are a) a multiverse or b) a cosmic designer.

Thus, the multiverse is not as bad of a solution to fine-tuning as some of its opponents have claimed. Nevertheless, there are at least five reasons why a multiverse is not a great solution. Firstly, all versions of the multiverse rely on speculative theories. Inflation is a reasonably wellestablished theory, but not all forms of it will go on forever and create bubble universes. Even more importantly, the mechanism for eternal inflation will not by itself give rise to different physics in each of the bubble universes it creates.9 This means that, without an additional speculative theory, eternal inflation cannot explain why the laws of physics should be tuned to permit life, and hence it cannot account for fine-tuning. To quote Bernard Carr and George Ellis, “We are being told that what we have is ‘known physics →multiverse.’ But the real situation is ‘known physics → hypothetical physics → multiverse.’”10

Secondly, even given a highly speculative theory (like string theory) in addition to eternal inflation, it is not clear that the set of allowable universes should contain a life-permitting one. In the string landscape, for instance, it appears that only certain values of the cosmological constant should be allowed.11 Considering how tiny the cosmological constant must be to permit life, roughly 10^-120, it is conceivable that not even a single one of the ~10^500 vacua of the string landscape will work! To solve the finetuning problem with a multiverse, we need sufficient variation in the laws of physics between the universes to ensure that some universes will be lifepermitting, and this is not certain. Third, the universe we observe is atypical even for a universe that permits life. According to Roger Penrose, the probability that an eternal inflation mechanism should produce universe just 10 times smaller than our own is 10^10^123.12 I have often heard people try to argue against Christianity by asking, “How could God care about us tiny little insignificant human beings when the universe is so vast?” Aside from the fact that significance is not determined by size and the possibility that God made the universe so large to tell us something about his own character, Penrose’s argument actually shows that the size of the universe works against the multiverse hypothesis and in favor of the design hypothesis. For if our universe is simply the result of a blind multiverse-producing mechanism, we should overwhelming expect to find ourselves living in the smallest possible life-permitting universe. And it is clear that our observable universe could be far smaller than it is and still permit life.

The fourth objection is like the third, and it raises the infamous problem of Boltzmann brains. In any universe that expands indefinitely at an increasing rate (such as our own evidently will), quantum fluctuations will spontaneously generate isolated, self-aware entities known as “Boltzmann brains.” And Boltzmann brains, unlike biologically evolved life, do not require a finely tuned universe—they arise naturally in the sea of possible universes.13 Thus, if a multiverse really does exist, it appears far more likely that I should be a Boltzmann brain rather than a truly evolved life form. The type of selfaware beings we have here on Earth is highly atypical, even for a multiverse.

The last objection is a fun one put forth by physicist Paul Davies. Davies argues that if there exists a multiverse capable of producing universes with intelligent beings, then surely some of these universes must hold intelligent beings that are smart enough to simulate worlds like ours on a computer or similar contraption. In this case, what justification do we have for believing that our universe is real, rather than a computer simulation of some ultra-intelligent species in another (real) universe (as in “The Matrix”)? And in light of this, what justification do we have for trusting the laws of nature in our universe at all, in particular the ones that give rise to a multiverse in the first place?14 Yep, you should have taken the blue pill.

Thus, the multiverse is not a bad idea. Reasonable physical theories have been proposed that will give rise to one. And if a cosmic designer does not exist, I am willing to say that a multiverse does. But on the other hand, a multiverse is far from a sure thing, and additional issues plague even the best models. As Barnes writes, “The multiverse may yet solve the fine-tuning of the universe for intelligent life, but it will not be an easy solution. ‘Multiverse’ is not a magic word that will make all the fine-tuning go away.”15

Beyond this, it should be noted that a multiverse is not in any way incompatible with Christian theism.16 Nowhere does the Bible suggest that God has created only one universe or one set of beings in his image. A successful multiverse scenario, at most, could help refute the teleological argument for God’s existence. But as atheistic philosopher Kai Nielsen pointed out, “To show that an argument is invalid or unsound is not to show that the conclusion of the argument is false…. All the proofs of God’s existence may fail, but it still may be the case that God exists.” God could perfectly well have created an infinite number of universes, even an infinite number of universes with intelligent life. After all, if you were God, would you really create just one?

In conclusion, modern physics, and string theory in particular, is not at odds with theism. Cosmology in the last 100 years has given us strong reasons to believe in a cosmic creator and designer of the universe. Beyond this, I think I can safely say that every serious physicist has at some point been struck by the mathematical coherence and reliability of the laws of nature, a piece of cosmic serendipity that begs for more of an explanation than “sheer coincidence.” It is hard for any scientist to get too far into his or her field without feeling the unmistakable impression of design.17

Nevertheless, a question necessarily arises in the mind of the skeptic: why has God not given us a perfect scientific proof or logical of his existence? My favorite answer to this question lies in John 1:14, which says that when God entered the world in the person of Jesus Christ, “The Word became flesh and dwelt among us.” The Greek word John uses for “Word” here is the Greek word logos, and from it we get the English word, “logic.” In other words, “The logic became flesh”: God has not given us a perfect proof for his existence, but rather he has given us a perfect person in Jesus Christ.18 Jesus Christ lived the life we should have lived and died the death we were owed for our sin. And with his Resurrection, he proved that he conquered death once and for all, granting all who would receive him the undeserved gift of adoption into God’s family. God has primarily chosen to reveal himself not by philosophical argument, but by personal relationship. And that relationship, unlike string theory, is something you can test for yourself.



1 This phenomenon is discussed most famously in cosmologist Martin Rees’ book, Just Six Numbers. Rees, Martin J. Just Six Numbers: The Deep Forces That Shape the Universe. New York: Basic Books, 2000.

2 Stenger, Victor J. The Fallacy of Fine-Tuning: Why the Universe is Not Designed for Us. Amherst, N.Y.: Prometheus Books, 2011. p. 22-24.

3 For an excellent, though snarky, rebuttal, as well as a list of such physicists who agree, see Barnes, Luke A. “The Fine Tuning of the Universe for Intelligent Life.” 2012. arXiv: physics. hist-ph/1112.4647.

4 See, for instance, Stephen Hawking’s recent book, The Grand Design. Hawking, Stephen, and Leonard Mlodinow. The Grand Design. United States and Canada: Bantam, 2010. p. 164.

5 Susskind, Leonard. “The Anthropic Landscape of String Theory.” arXiv: hep-th/0302219.

6 String theorists are searching for a “vacuum selection principle” that would select our universe to be the one that is actually produced, but this is looking less and less likely.

7 It should be noted that these “Wbubble universes” might sometimes collide with one another, producing a signal that we could observe in the cosmic microwave background (CMB). (See Feeney, Stephen M. et al, “First Observational Tests of Eternal Inflation: Analysis Methods and WMAP 7-Year Results.” arXiv: astro-ph/1012.3667.) However, the expected number of such collisions is small, and if a sufficient amount of expansion has taken place in our universe, then such a signal will remain beyond the realm of experiment. We have not seen compelling evidence of such a collision so far, and we would have to be very lucky to ever see any. (See Bousso, Raphael, “The Cosmological Constant Problem, Dark Energy, and the Landscape of String Theory.” arXiv: astro-ph/1203.0307.)

8 Richard Dawkins even calls this the “central argument” of his book, The God Delusion. Dawkins, Richard. The God Delusion. New York: Mariner Books, 2007, p. 187-189.

9 Ellis, Bernard. “Does the Multiverse Really Exist?” Scientific American, August 2011, p. 38- 43.

10 Carr Bernard J., and George Ellis. “Universe or Multiverse?” Astronomy & Geophysics. 49, 2.29, 2008.

11 J. Polchinski. “The Cosmological Constant and String Landscape.” 2006. arXiv:hepth/ 0603249.

12 Penrose is actually talking more specifically about a certain, more potent type of eternal inflation called “chaotic inflation” which will actually produce universes with different laws of nature. (See: The Road to Reality: A Complete Guide to the Laws of the Universe. Vintage, London, 2004, p. 763ff.) However, his argument is broad enough to apply to most plausible multiverse scenarios. Barnes, Luke A. “The Fine Tuning.”

13 Barnes, Luke A. “The Fine Tuning.” 14 P. Davies. Cosmic Jackpot: Why Our Universe is Just Right for Life. New York: Houghton-Mi_in Co., p. 179-185, 2007.

15 P. Davies. Cosmic Jackpot: Why Our Universe is Just Right for Life. New York: Houghton-Mi_in Co., p. 179-185, 2007.

16 C.S. Lewis handled the issue of life on other planets well in his essay “Religion and Rocketry,” and his arguments apply equally well to the possibility of life in other universes. Lewis, C.S. “Religion and Rocketry.” The World’s Last Night, and Other Essays. New York: Harcourt, Brace, 1960.

17 To quote biologist Francis Crick, certainly no theist, “Biologists must constantly keep in mind that what they see was not designed, but rather evolved.” Crick, Francis. What Mad Pursuit: a Personal View of Scientific Discovery. New York: Basic, 1988. p. 138.

18 Keller, Timothy. “The Word Made Flesh.” Sermon. December 13, 1990.


Tom Rudelius is a second-year Ph.D candidate in Physics at Harvard and is a guest writer for the Ichthus.

Tags: , , , , , , , , , , , , , , , , , , , ,