Is the Multiverse Real?

by Elie Feder and Aaron Zimmer, cohosts of the Physics to God podcast

What is the Multiverse?

Imagine an infinite number of unobservable parallel universes, each with different laws of nature. While this idea has captured imaginations for millennia—the latest example being Marvel movies—the theory of the multiverse has recently skyrocketed in popularity, even among serious scientists. Many top physicists truly believe in an infinite multiverse and consider it the best scientific alternative for explaining the fine tuned constants of nature without an intelligent cause (or, as it is more commonly stated, without God).

The idea of the multiverse is not new. Over 2,000 years ago, the Roman philosopher, Lucretius, theorized that over an infinite expanse of time, atoms randomly assemble themselves into all possible arrangements—including a universe like our own. This idea was revived over a hundred years ago in 1895 by physicist Ludwig Boltzmann; only this time it was described in the more modern language of random fluctuations in the state of thermal equilibrium.

Despite its various iterations, few philosophers or scientists have taken the multiverse seriously—until 1998. Cosmologists discovered that the expansion rate of our universe is accelerating due to the extremely fine tuned cosmological constant. Because of the clear theological implications of fine tuning, this discovery forced modern scientists to reconsider the ancient theory of the multiverse.

After introducing the idea of fine tuning, this essay lays out the three basic features needed for a multiverse theory to successfully explain fine tuning. It then analyzes and evaluates the scientific supports that multiverse scientists offer for each of its three premises. Finally, it shows why proper scientific methodology leads to the devastating measure problem and necessitates rejecting the multiverse as a viable scientific theory.

This essay discusses:

For all these points in more depth, watch or listen to Season 2 of the Physics to God podcast, cohosted by Elie Feder PhD and Aaron Zimmer, on YouTube, Spotify, or Apple Podcasts.

Fine Tuning

The fine tuning of the constants of nature is one of the main problems the multiverse attempts to solve. Before we get to fine tuning, let’s introduce the constants of nature—25 mysterious numbers that determine the quantities of our universe’s fundamental constituents. One such number is the mass of an electron, 9.109×10^−31 kg, which expresses the size of every electron.

Explaining these numbers presents a huge challenge to physicists. In his 1985 book, QED (p. 127), famed physicist Richard Feynman said that explaining the values of the constants is “one of the greatest damn mysteries in physics.” Feynman’s mystery has two sides that depend upon how one tries to explain the constants.

On the one hand, one may suggest that these 25 numbers are fundamental, brute facts of reality that must be accepted with no deeper explanation. But this goes against physicists' experience and intuition that our universe’s fundamental components are few, simple, and beautiful—not 25 complex, arbitrary numbers.

On the other hand, one may suggest that the constants aren’t fundamental but result from some deeper theory that explains their precise values. But this also seems implausible—what kind of qualitative theory could possibly explain such seemingly arbitrary numbers?

Over time, scientists have taken a major step toward solving Feynman’s great mystery. They slowly realized that these 25 numbers weren’t as arbitrary and random as they first appeared. In fact, scientists began to appreciate that the opposite is the case: the constants are fine tuned. In other words, if the values of the constants were even a little bit different, then our universe wouldn’t contain any complex structures like atoms, molecules, planets, stars, galaxies, or life. Instead, our universe would be a chaotic sea of fundamental particles with no higher structures or complexity.

In his book, The Grand Design, atheist physicist Stephen Hawking described fine tuning as follows:

Most of the fundamental constants in our theories appear fine tuned, in the sense that if they were altered by only modest amounts, the universe would be qualitatively different, and in many cases unsuitable for the development of life. The laws of nature form a system that is extremely fine tuned, and very little in physical law can be altered without destroying the possibility of the development of life as we know it.

While the discovery of fine tuning provided a significant clue towards unraveling the mystery of the constants, for many physicists, it presented a new problem. It seemed to indicate that the values of the constants have the purpose of producing a complex, ordered, and structured universe. This strongly implies that an intelligent cause intentionally chose the specific values of the constants in order to bring about a complex universe.

Not so fast, argued atheist scientists. Perhaps there’s a way to explain the fine tuned values of the constants without an intelligent cause. Enter the multiverse.

Three Premises of Multiverse

A multiverse must establish three premises in order to explain fine tuning without an intelligent cause:

Infinite Multiverse Premise: There are an infinite number of parallel universes.
Varied Multiverse Premise: The values of the constants vary from universe to universe.
Typical Universe Premise: In the infinite varied multiverse, our universe is a typical universe with intelligent observers.

Assuming these three premises are all true, the multiverse can explain fine tuning by chance alone. Here’s how. If there really are an infinite number of parallel universes (premise 1) and the constants vary between these universes (premise 2), then every possible combination of values must occur in some universe in the multiverse. This means that, by chance alone, there will be at least one universe whose constants will have the fine tuned values we’ve observed in our universe.

Of course, 99.9999% of the universes won’t have fine tuned constants and will be bereft of any structures or complexity (that is, no atoms, molecules, planets, stars, galaxies, or life). That might lead you to ask: Isn’t it incredibly unlikely that we would end up in this amazing universe with so much order and complexity?

The straightforward answer is that we shouldn't be surprised to find ourselves in one of the few universes with constants that can support life. This is because the overwhelming majority of universes with the “wrong” constants don’t have any intelligent observers to ponder the values of their constants. In other words, since all intelligent observers capable of wondering why their universe looks fine tuned will, by definition, be in a fine tuned universe, it’s perfectly clear why we’re in a fine tuned universe. (The appeal to an observer bias to justify the observation of fine tuned constants is called the anthropic principle.)

While this may seem like a satisfactory explanation for fine tuning, it’s fully dependent on multiverse’s three premises. With this in mind, let’s examine these three premises more closely:

How do multiverse scientists justify their first premise—the claim that there are an infinite number of parallel universes? Since we can’t ever observe or verify them experimentally, how can positing them be considered science?
How do multiverse scientists justify their second premise—the claim that the constants are different in the other universes? Since we can’t observe them, isn’t it simpler to assume that the constants remain constant throughout all universes?
What’s the significance of the third premise—that in an infinite varied multiverse, our universe is a typical universe with intelligent observers? What role does it play in multiverse theory and how do we know if it’s true or false?

To answer these questions, let’s examine each premise of the multiverse and the evidence that supports it.

Infinite Multiverse Premise

Multiverse scientists attempt to establish the existence of an infinite number of parallel universes in two primary ways. The first is the many worlds interpretation of quantum mechanics and the second is the theory of eternal inflation. While these are both very complicated topics, we’ll describe them briefly and explain how they both only support the first premise but don’t provide any support for the second premise. Don’t worry if you don’t follow all the details; just try to get the main idea.

Many Worlds Interpretation of Quantum Mechanics

One of the strange features of quantum mechanics is the idea that for any event that permits multiple possible outcomes, all the different possibilities are in some sense realized. For example, before quantum mechanics, it was known that if a single indivisible particle is sent through a double slit, it could either go through the left slit or the right slit. But according to quantum mechanics, it somehow goes through both slits. How does this make any sense?

The conventional (Copenhagen) interpretation explains that while all possible outcomes of an experiment somehow really exist in a potential state (thereby contributing to the many strange features of quantum mechanics), a physical measurement causes only one possibility to be actualized (through a process called wave function collapse).

However, there are other interpretations of what happens to the different possible outcomes. The many worlds interpretation, proposed by physicist Hugh Everett in 1957, suggests that every possible outcome is realized in a separate universe. This means that there’s a universe where the particle goes through the right slit and a different universe where it goes through the left slit. And likewise for every single event with multiple outcomes. As you can imagine, this generates lots and lots of universes—a type of multiverse.

Whether or not you subscribe to the many worlds interpretation of quantum mechanics, it has nothing to do with explaining fine tuning without an intelligent cause. Here’s why. Even according to the many worlds interpretation, the laws and constants of nature stay the same in all the worlds (or universes). The only difference between the different worlds is which possibility is realized (that is, whether the particle goes left or right). Therefore, while the many worlds interpretation may justify the first premise that there are an infinite number of parallel universes, it fails to support the second premise that the constants of nature change between universes.

If fact, the opposite is true. Because the many worlds are all generated by the same laws of quantum mechanics, it stands to reason that the laws and their constants remain fixed in all the worlds—in clear opposition to the multiverse’s second premise. And if there really do exist infinitely many worlds that all have the same laws and fine tuned constants as our universe does, this would indicate that an intelligent cause fine tuned them all. In other words, the many worlds interpretation of quantum mechanics, on its own, does nothing to undermine the evidence for the existence of an intelligent cause of our fine tuned universe.

Eternal Inflation

That brings us to the primary way multiverse scientists establish the existence of an infinite number of universes—eternal inflation. To understand eternal inflation, we’ll first need to start with the basic theory of inflation.

The theory of inflation is a modification of classical big bang cosmology which asserts that the universe started from an infinitely small and dense beginning and has been expanding ever since. The theory of inflation accepts that claim but adds that in the first second after the big bang, the universe underwent a period of extremely rapid expansion. Because the theory of inflation solves various problems in cosmology and makes predictions that are supported by observations, it’s a fairly well-accepted cosmological theory.

You may wonder: What does inflation in our one universe have anything to do with an infinite number of parallel universes? The answer is that while scientists have suggested many ways to model the mechanism that caused our early universe to inflate, the simplest models all suggest that once inflation begins, it keeps on going forever and ever—what’s called eternal inflation—with the inflation only slowing down within small isolated pockets. This means that while most of the vast expanse of the universe keeps growing forever, small portions slow down and form distinct bubble universes. Since the inflation continues infinitely, eternal inflation produces infinitely many isolated bubble universes—hence, an infinite multiverse.

Assuming eternal inflation is true, it only justifies the existence of an infinite number of universes. However, since the same laws of physics generate all the bubble universes, it would stand to reason that the values of the constants should always remain the same. So let's see how multiverse scientists attempt to justify the second premise—that the values of the constants differ from universe to universe.

Varied Multiverse Premise

Let’s begin with an analogy. If you bought infinitely many lottery tickets that all had the same numbers, your odds of winning would be just as low as if you only bought one ticket. To increase your odds of winning, you would need to buy tickets with different numbers. Taken to the extreme, if you bought an infinite number of tickets with different numbers, you would be guaranteed to win.

Similarly, as we just mentioned, the existence of an infinite number of universes that all have the same fine tuned values for the constants is insufficient to solve the problem of fine tuning without an intelligent cause. After all, you’d have to ask: Why do all the universes have these same fine tuned constants if not for the purpose of producing many complex, ordered, and structured universes? This scenario would lead directly to an intelligent cause—exactly what multiverse scientists are trying to avoid.

Therefore, to explain fine tuning without an intelligent cause, multiverse scientists must establish that the constants vary from universe to universe. To do this, they take two approaches.

Fine Tuning and the Hidden Premise

As surprising as it may seem, some multiverse scientists argue that fine tuning itself indicates that the constants are different in other universes. To do so, they argue by elimination: Since there’s no reasonable way to explain fine tuning in our one universe other than positing an infinite varied multiverse, the constants must vary from universe to universe.

Of course, this argument is entirely dependent on the unjustified premise that an intelligent cause is impossible. If, on the other hand, one were to examine the evidence without any prior prejudice, fine tuning in our one universe would directly indicate that an intelligent cause set our constants for the purpose of producing a complex universe. Therefore, multiverse scientists’ argument by elimination commits the fallacy of assuming its conclusion.

String Theory

A somewhat stronger support for the second premise comes from string theory, the primary way multiverse scientists attempt to establish that the constants vary between universes. Before explaining how it does that, let’s give a basic introduction. String theory is currently the most popular contender for a theory of everything—one theory that unifies all known physics and explains everything about our universe. Despite string theory’s initial promise, it comes with a few surprises.

One such surprise is that for string theory to be mathematically consistent, it must posit at least six hidden dimensions of space alongside our three familiar dimensions. A hidden dimension is a tiny, unobservable, curled-up dimension of space. A classic analogy for a hidden dimension is an ant on a rope. From far away, it seems that the ant can only crawl back and forth along one dimension of the rope. But zooming in reveals that the ant can also crawl around the circumference of the rope. In this sense, the circumference is a hidden dimension of the rope. Similarly, string theory posits that our universe has six such hidden dimensions.

According to string theory, the precise shapes of the curled-up hidden dimensions impact the properties of the fundamental particles (such as their mass and charge). Therefore, for string theory to explain the various features of our universe, scientists must identify exactly how our hidden dimensions are curled up. But because there’s no way to observe our universe’s hidden dimensions, there’s also no direct way to determine how they’re curled up.

However, physicists harbored another hope for determining the shape of the hidden dimensions. Based on scientists’ long-standing dream of finding one unique final theory of everything, string theorists hoped that theoretical considerations would imply that there was only one unique way to curl up the hidden dimensions.

To their chagrin, string theorists soon realized that this couldn't be further from the truth—they discovered more than 10^500 possible ways for the unobservable hidden dimensions to be curled up. This discovery presented a major challenge to string theory’s hope to provide a final theory for our one universe.

In came the multiverse. Multiverse scientists, led by Leonard Susskind, turned the problem on its head. They argued that if eternal inflation is correct that there really are an infinite number of universes, then maybe it isn’t so bad that string theory has so many different versions. Perhaps the hidden dimensions aren’t fixed in one precise shape but vary from universe to universe. If so, the values of the constants—which follow from the shape of the hidden dimensions—would also vary from universe to universe.

Stepping back, we can see that the desire to rescue string theory from the major disappointment of its not being a unique theory was turned into a positive feature that justified the belief in other possible values for the constants of nature. This conveniently provided multiverse scientists with some justification for the varied multiverse premise’s claim that the values of the constants vary from universe to universe.

Putting it all together, in conjunction with the infinite universes of eternal inflation, string theory allows multiverse scientists to claim that they have scientific justification for an infinite varied multiverse.

Naive Multiverse

One might naively assume that the eternal inflation/string theory multiverse can sufficiently explain fine tuning without an intelligent cause. After all, if eternal inflation is correct that there are an infinite number of universes and string theory is correct that they each have different values for the constants, then every possible combination of constants, including our fine tuned values, is realized somewhere in the multiverse. And, obviously, we can only exist in such a fine tuned universe. Fine tuning explained! This is how the multiverse is often (naively) depicted in popular literature.

As compelling as this may sound, no serious multiverse scientist contends that this is sufficient justification for the multiverse. They all agree that for the multiverse to be a viable theory, it needs the third premise that our universe is typical. Before we get to that, let’s explain what’s wrong with a naive multiverse—that is, what’s wrong with explaining fine tuning (or, for that matter, explaining anything at all) based on the first two premises alone. We’ll thereby show the necessity of the all-important third premise.

The basic problem with a naive multiverse begins with the fact that literally everything physically possible happens somewhere in an infinite varied multiverse. In the words of physicist Alan Guth, the originator of the theory of inflation, “In an eternally inflating universe, anything that can happen will happen; in fact, it will happen an infinite number of times.” Therefore, attempting to explain anything merely by saying, “It happens somewhere in an infinitely varied multiverse,” leads to many absurd consequences.

To fully appreciate these absurdities, let’s consider three serious and related problems that emerge from a naive multiverse’s attempt to explain fine tuning solely based on the idea that everything must happen somewhere in an infinite varied multiverse.

1) Multiverse of the Gaps

A naive multiverse can be used to explain anything and everything. Imagine we had no knowledge of the biological theory of evolution and were trying to explain the existence of life. While life seems extremely unlikely to emerge by chance alone, it’s physically possible (though incredibly unlikely) for a quantum fluctuation to randomly and instantaneously produce cats, dogs, and all the other forms of complex life.

Since this is physically possible, an infinite multiverse will have at least one universe where all forms of life randomly emerge in this manner. This being the case, a naive multiverse can explain the diversity of life—without evolution—by positing that there’s an infinite varied multiverse, and we happen to be in an unlikely universe where all forms of life simply “fluctuated into existence.”

By the same exact line of reasoning, any gap in knowledge could be filled in by the claim that in an infinite varied multiverse, it has to happen somewhere. This is a major issue. Since a naive multiverse can explain anything, it’s merely a theory of ignorance. A naive multiverse of the gaps could be invoked to explain any possible observation, irrespective of what our actual universe looks like, and therefore doesn’t explain any specific observation, such as our fine tuned constants.

2) Intrinsically Irrefutable

A naive multiverse is capable of explaining away blatantly contradictory evidence. There is literally no evidence or observation that could possibly disprove a naive multiverse. Imagine a great voice came from heaven and said, “I am God. I fine tuned the constants for the purpose of bringing about our complex universe. The multiverse is false.”

While this would seem to falsify the multiverse and prove that God exists, a naive multiverse theorist could explain away this voice without an intelligent cause. After all, he can argue that perhaps a random quantum fluctuation produced an unlikely sound wave just like that voice. Since such an unlikely sound is physically possible, it must occur somewhere in the infinite varied multiverse—and apparently, we’re in that universe.

It’s not good for a theory to be intrinsically irrefutable. Clear contradicting evidence should be able to disprove a theory. The fact that even a voice from heaven can’t disprove a naive multiverse reveals that it’s no better than the many irrefutable theories that have been imagined throughout the millennia (such as Descartes’ theory of an evil demon or the theory that you’re a brain in a vat). They can’t all be true, because they’re mutually exclusive—so what justifies your arbitrary belief in any one of them?

One may argue that even though a naive multiverse’s ability to explain anything makes it intrinsically irrefutable, it’s different from other irrefutable theories because it’s grounded in solid scientific theories like eternal inflation and string theory. This leads to a third fundamental issue with a naive multiverse.

3) Self-defeating

A naive multiverse undermines the very scientific theories it’s based upon (eternal inflation and string theory). The only justification multiverse scientists have for believing in an infinite varied multiverse is that scientific observations and experiments seem to support the theories of inflation and string theory. But if every possible experimental result will occur somewhere in an infinite varied multiverse, there’s no meaning to the claim that a particular experimental result either confirms or contradicts any theory.

Think about it. In an infinite varied multiverse, since everything happens somewhere, any experimental test for any theory will be confirmed in some universes and contradicted in others. Therefore, a naive multiverse can explain all these results in the exact same manner: It has to occur somewhere in an infinite varied multiverse, and apparently, we’re in that universe. This renders the entire scientific method completely useless, thereby undermining the very scientific theories purported to establish a multiverse in the first place.

For the three reasons just described (and because of other absurd consequences), no serious multiverse scientist actually believes in a naive multiverse. They all realize that for the multiverse to be a viable theory, it must make a specific prediction about what we should expect to observe if we really were in an infinite varied multiverse. That’s why multiverse scientists need to complete multiverse theory with the third and final premise: that our universe is typical.

Typical Universe Premise

If we think deeply about the multiverse, we’ll realize that even though everything possible happens somewhere in an infinite varied multiverse, an intelligent observer would expect to find himself in a typical universe with intelligent observers. This expectation follows directly from using chance and observer bias to explain our observations, and most crucially, provides a type of falsifiable prediction for the multiverse.

We know this sounds complicated. Of everything we've discussed here, it's probably the hardest part to understand. So we'll take it slow and explain it carefully.

In an infinite multiverse, there are many different types of universes. Some have intelligent observers, while others do not. Even though 99.9999% of universes have no intelligent observers, you will invariably find yourself in a universe with intelligent observers—after all, you are one. In other words, an observer bias explains very well why you observe a universe with intelligent observers.

Let’s now consider the set of all different universes that contain intelligent observers. These too are very diverse. Some universes (like ours) have intelligent observers and a hundred billion galaxies. But there are other, simpler universes with intelligent observers and only one galaxy. On the crazier side, there are universes with intelligent observers as well as unicorns and fire-breathing dragons. More generally, there are infinitely many different types of universes that all have intelligent observers, but are differentiated in many other regards.

From all the universes in this set, which type of universe would an intelligent observer like you expect to be in? In other words, of the wide variety of universes with intelligent observers, which would you predict to see if you truly lived in an infinite multiverse? In considering this question, the key idea is that an observer bias won't help answer it—after all, all the universes we’re considering have intelligent observers. So what would you expect?

The reasonable prediction is that you would observe a typical, or likely, universe with intelligent observers. This follows from using chance and probabilities in an infinite varied multiverse.

To appreciate this point, consider the following examples: If the vast majority of universes with intelligent observers had unicorns and fire-breathing dragons, then you’d predict that you would see them. However, if the vast majority of universes with intelligent observers did not have unicorns and fire-breathing dragons, then you would predict that you would not see them. The point is that in an infinite varied multiverse, an observer bias will always lead to the prediction that you’ll observe the typical universe with intelligent observers.

In the words of physicist Brian Greene, from The Hidden Reality (p. 207):

Life may be rare in the multiverse; intelligent life might be rarer still. But among all intelligent beings, the anthropic assumption goes, we are so thoroughly typical that our observations should be the average of what intelligent beings inhabiting the multiverse would see. (Alexander Vilenkin has called this the principle of mediocrity.)

The realization that we would predict to find ourselves in a typical universe solves the three aforementioned problems with a naive multiverse:

Since an infinite varied multiverse predicts that we should observe a typical universe, it isn’t a theory of the gaps. After all, it couldn’t explain the observation of an atypical universe (such as a universe with unicorns and fire-breathing dragons).
If we witnessed atypical refuting evidence—like a voice from heaven that rejected the multiverse—we would know that the multiverse is false. That’s crucial because it makes the multiverse falsifiable, and thus a more meaningful scientific theory.
Even though an infinite multiverse contains many atypical phenomena somewhere, our experiments should yield results that would be expected in a typical universe. We are therefore justified in relying on the scientific method.

With this point in mind, we have a way to test the multiverse—we must look at our universe and rigorously assess if it’s truly typical. If we find that it is, then a multiverse combined with an observer bias is a good explanation for fine tuning. However, if we find that our universe is atypical, an observer bias couldn’t explain it; even worse, this would falsify the multiverse’s prediction that our universe is typical. So let’s go and check…

Is Our Grand Universe Typical?

At the outset, it’s worthwhile to note that the claim we live in a typical universe with intelligent observers is highly counterintuitive. Our grand universe has over 100 billion galaxies, each with over 100 billion stars and planets. This seems quite atypical indeed. Wouldn’t it be more likely for us to exist in a much smaller universe with much less order and complexity?

To try to get around this problem, one can plausibly argue that the typical way to get intelligent observers like human beings is through biological evolution—which needs the existence of our entire planet, and perhaps our entire solar system and galaxy. While this can potentially explain everything in our galaxy, there seems to be no reason to think that the other 100 billion galaxies are all somehow part of the typical way to bring about intelligent observers like humans.

To maintain the typical universe premise, a multiverse scientist would have to maintain that the typical universe with intelligent observers has over 100 billion galaxies. This is highly unintuitive, if not obviously wrong; there’s no reason to think that the rest of our grand universe has anything to do with us at all. To quote physicist Richard Feynman, “The stage is too big for the drama.”

Nevertheless, multiverse scientists are compelled to maintain that our grand universe is the typical way to get human beings. So how do they evaluate whether this counterintuitive claim is true or false?

Infinities and Measures

When scientists try to carefully evaluate the multiverse’s sole prediction—that our universe is typical—they immediately face a serious problem: an infinite multiverse is incapable of making any prediction whatsoever!

To see why, remember physicist Alan Guth’s statement that in an infinite multiverse, “anything that can happen will happen; in fact, it will happen an infinite number of times.” This implies that every single possible universe will happen over and over again. In other words, there will be an infinite number of identical copies of every possible universe. This presents a major problem for computing probabilities to determine the typical universe.

To appreciate why, consider the following example. If the multiverse contained a large, but finite number of universes, and for every universe with unicorns there were a million without unicorns, then we could legitimately predict that the typical universe wouldn’t have unicorns. But if there are an infinite number of universes with unicorns and an infinite number of universes without unicorns, it’s impossible to say which universe is more likely. Both types are infinite!

Physicists' only solution for this problem is to introduce a measure. A measure is an externally imposed rule, or a meta-law, that tells you to order the infinite set in a specific manner that allows you to compute probabilities for different parts of the set. The measure allows probabilities to be meaningful even though the set is infinite.

To get a better sense of how a measure helps determine the typical item in an infinite set, consider the following example. You randomly select a marble from a row containing an infinite number of both gold and silver marbles. Because there are an infinite number of both colors, there’s no way to predict which color you’re likely to select. This is the problem of taking probabilities in an infinite set.

To get around this problem, let’s alter the example. When you set up the infinite row of marbles, you impose a rule that every gold marble will be followed by ten silver marbles. Now, even if you have an infinite number of both colors, it would be reasonable to predict that a random selection would yield a (typical) silver marble. The point is that once you choose your measure, you can take probabilities even though your set is infinite.

This is just what multiverse scientists attempt to do. In order to take probabilities in an infinite multiverse, they posit a measure—a rule for weighting different types of universes in the multiverse. Multiverse scientists are currently searching for just the right measure that would yield the result that our universe is typical.

Ad Hoc Measures

While the introduction of a measure seems like it may salvage multiverse theory, it leads directly to the three layers of the devastating measure problem. We’ll explain this problem and develop its three layers over the next few sections.

The first layer is that the whole idea of introducing an ad hoc measure is artificial. This is because a measure doesn’t naturally emerge from any fundamental law of physics, like inflation or string theory, but is a superadded meta-law that’s tacked onto the laws of nature for the sole purpose of making our universe typical.

To see this problem more clearly, consider the gold and silver marbles. Since there are infinitely many gold marbles and infinitely many silver marbles, the rule (or measure) doesn’t have to be 10 silver marbles for every 1 gold marble; it could just as easily have been to place 37 silver marbles for every 1 gold marble, or 23 gold marbles for every 7 silver marbles. Since the measure is an externally imposed rule, nothing inherent to the situation dictates what it should be.

Likewise, the whole idea of introducing measures to rescue the multiverse is entirely artificial. Multiverse scientists’ subjective need to posit a measure to prevent their theory from reducing to a flawed naive multiverse doesn’t imply the objective existence of a meta-law that governs probabilities in the multiverse.

Intuitive Measures Don’t Work

Since measures are ad hoc additions that don’t emerge from any fundamental law of physics, multiverse scientists face a challenge in selecting the right measure. To do so, they search for some intuitive reason to select a specific measure that could, in some sense, be described as self-justifying. For example, a volume-weighted measure asserts that universes that take up more volume in the multiverse’s overall spacetime get assigned a higher probability.

The problem with the volume-weighted measure—and every other intuitive measure that multiverse scientists have tried over the past few decades—is that none of them work. In other words, when multiverse scientists use these measures to check if our universe is typical, they find the exact opposite—due to one reason or another, our universe is very special and atypical. That is, a multiverse governed by any of these measures makes a false prediction. This is the second layer of the measure problem.

The Boltzmann Brain Problem

One of the most common ways intuitive measures make a false prediction is due to the Boltzmann brain problem. This problem was first formulated in 1931 by physicist Arthur Eddington as an attack against Ludwig Boltzmann’s 1895 multiverse theory. We’ll introduce the problem in an intuitive way (similar to how it was first formulated) and then show how it applies to measures.

Let’s consider for a moment what the typical observer in an infinite multiverse looks like. Broadly speaking, in an infinite multiverse, there are two possible types of intelligent observers that are generated in two entirely different ways:

(1) The first type, called a normal observer, is a regular person who finds himself in an ordered universe and is generated through the ordinary unfolding of the laws of nature.

(2) The second type, known in the scientific literature as a Boltzmann brain or a freaky observer, is a single brain—with false memories and beliefs—that’s surrounded by total chaos and is generated by a random fluctuation.

So, which type of observer is typical? Or, to say it differently, which type of observer would you expect to be if you truly lived in an infinite multiverse?

While your observations and memories of a grand, ordered universe may lead you to believe that you are a normal observer, in truth, if your entire brain fluctuated into existence a second ago, everything would appear exactly the same. In other words, your observations are incapable of distinguishing between the two scenarios. So which is more likely?

On the surface, you may think it’s much easier for the laws of nature to produce a normal observer in an ordered universe than for a random fluctuation to produce an entire ordered brain. However, if you think about it (or make some calculations), you’ll realize that the opposite is the case. This is because the only way an infinite multiverse explains the existence of our highly ordered universe in the first place is through a random fluctuation that produced the big bang. So, while a fluctuation that produces a single brain is indeed very unlikely, it’s still far more likely than a fluctuation that produces our entire ordered universe. (Physicist Roger Penrose, in his book The Emperor's New Mind, calculated the odds of such a fluctuation to be 1 out of 10^10^123!)

This leads to a startling conclusion: An infinite multiverse predicts that the typical observer in the typical universe is a Boltzmann brain—a freaky observer that recently fluctuated into existence and will just as quickly fluctuate out of existence. This conclusion was true in 1895 regarding Boltzmann's multiverse and is still true about the modern-day multiverse governed by an intuitive measure.

“So what?” you might say. “Maybe it is true. Maybe I do live in an infinite multiverse and maybe I am a Boltzmann brain?”

The problem with accepting this conclusion (beyond the fact that it’s clearly absurd and ridiculous) is that if you really are a Boltzmann brain, then all your memories are false, including your memories of learning physics. They never really happened—your brain is just configured in a way that makes you think they’re real. If so, your belief in a multiverse, which is based on theories in physics like eternal inflation and string theory, would be based on false memories of experiments that never occurred. But this would invalidate your motivation for believing in a multiverse in the first place!

This is a serious problem. An infinite multiverse that predicts you’re a Boltzmann brain ultimately undermines the very reason for believing in the multiverse in the first place. In other words, an infinite multiverse leads to the self-defeating conclusion that you’re a Boltzmann brain with false scientific knowledge that the multiverse is real—not a real human with true scientific knowledge.

Hence, the second layer of the measure problem can be summarized by saying that the multiverse’s sole prediction—that we are typical observers in a typical universe—seems patently false. To get around this major problem, multiverse scientists have one last attempt that leads directly to the third layer of the measure problem.

The Fine Tuned Measure Problem

Given that all semi-plausible measures don’t work, multiverse scientists attempt to start backward by assuming their desired result—that our universe is typical—and searching for a measure that would make it happen. Even with this contrived approach and even after decades of searching, they have still been unable to find a measure that would make our special universe typical.

While this is a serious issue in its own right, the third layer of the measure problem runs even deeper than multiverse scientists’ inability to find the “right” measure. Even if, at some point in the future, they were to find some contrived measure that happens to make our universe typical, this measure wouldn’t be intuitive or self-justifying. After all, there are dozens of more intuitive measures that could’ve theoretically existed but have, over the years, demonstrably failed to make our universe typical. Therefore, this handpicked measure would be subject to the all-important question: What fine tuned and designed the special measure that makes our complex universe typical, as opposed to all the intuitive measures that make our universe atypical?

Physicist Paul Steinhart in his Scientific American article, The Inflation Debate, gets to the essence of this problem:

Measure enthusiasts take a trial-and-error approach in which they invent and test measures until, they hope, one will produce the desired answer -- that our universe is highly probable. Even if they someday succeed, they will need another principle to justify using that measure instead of the others, yet another principle to choose that principle, and so on.

This argument clearly demonstrates that even if a contrived measure were discovered, it would still point right to an intelligent cause that selected this particular measure for the purpose of making our complex universe typical. Therefore, the whole approach of using measures to rescue multiverse’s typical universe premise and explain fine tuning without God is doomed to failure.

Three Layers of the Measure Problem

Before moving on, let’s summarize the three layers of the measure problem. While multiverse scientists attempt to use measures to show that our fine tuned universe is a typical universe in an infinite varied multiverse, the devastating measure problem totally undermines this approach. Its three layers can be succinctly stated as follows: (i) ad hoc measures are bad ideas to begin with; (ii) all intuitive measures don’t work; (iii) even if multiverse scientists were to find a contrived measure that did work, it would beg the question of what fine tuned and designed it.

Since the measure problem completely undermines the approach of introducing measures to determine the typical universe, we are left with the fact that an infinite varied multiverse is nothing more than a flawed naive multiverse theory. Since it’s a theory of the gaps that can explain anything and everything, in truth, the multiverse explains nothing at all.

Is the Multiverse Science?

To some degree, multiverse scientists know that an infinite number of unobservable universes where everything possible happens sounds like a wild philosophical theory. They are also well aware of the significance of the measure problem. Despite all this, they take refuge in the claim that the multiverse is a scientific theory that’s indicated by evidence. They argue that the multiverse should be no different than quantum mechanics—a theory that initially seems kind of bonkers but is a highly verified and well-accepted scientific theory. Likewise, no matter how crazy the multiverse may seem, its proponents argue for its acceptance as an ordinary part of science. But are they right? Is the multiverse truly science?

Let’s first analyze this question using the conventional definition of the scientific method (that includes the requirements of experimentation and testing). For example, the Oxford English Dictionary defines the scientific method as, “A method or procedure that has characterized natural science since the 17th century, consisting in systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses.”

It’s clear that speculation about an infinite number of unobservable parallel universes is intrinsically untestable. Furthermore, even the theories that give rise to the multiverse—string theory and eternal inflation—have never been experimentally verified. String theory has been unable to make a testable claim for decades, and eternal inflation can’t been tested because it predicts an infinite number of copies of every type of universe. This makes it incapable of making any predictions without introducing an ad hoc measure specifically designed to make "predictions" that align with all our previously known observations. That’s hardly what conventional science considers a legitimate test.

The multiverse’s failure to satisfy the key requirements of the scientific method (experimentation and testing) has moved some multiverse scientists to argue that we need to change the definition of science to accommodate it under the banner of science. However, this is highly controversial, even in the scientific community. Physicist George Ellis articulated the significant harm that would result if the definition of science were broadened to make room for the multiverse:

The foundations must be respected if one is to preserve the core features of science that have led to its phenomenal success: that is the feedback from reality to theory provided by experiment and observational testing. One abandons that at one’s peril...it is dangerous to weaken the grounds of scientific proof in order to include multiverses under the mantle of “tested science”. It is a retrograde step towards the claim that we can establish the nature of the universe by pure thought without having to confirm our theories by observational or experimental tests. This abandons the key principle that has led to the extraordinary success of science. The claim that multiverses exist is a belief rather than an established scientific fact.

The reason we collectively accept the conclusions of science is because they’ve been tested and compared to reality. However, changing the definition of the scientific method would undermine the very foundation of science and erode the essential reason it’s accepted as the gold standard of verified knowledge. Changing science to accommodate the multiverse would be a serious error that would threaten the entire scientific endeavor, and ultimately wouldn’t change the fact that the multiverse is an untenable speculative philosophical theory—no matter what you call it.

Before moving on, we’d like to mention that physicist Lee Smolin has proposed a theory, called Cosmological Natural Selection, that attempts to mimic the success of biological natural selection and explain our fine tuned constants as the result of a slow evolutionary process that is proposed to occur through new universes, with slightly altered constants, being born inside black holes.

While this theory is better than an infinite varied multiverse and has potential as a genuine scientific theory, it too fails to explain our fine tuned constants without an intelligent cause. For more on this theory and where it goes wrong, see our essay on Cosmological Natural Selection.

Conclusion

Over the past few decades, the multiverse has surprisingly risen in popularity among respected scientists. In our opinion, this is largely because it’s seen as the only viable alternative to explaining the fine tuning of the constants without an intelligent cause. For atheist scientists who view God as simply impossible, the multiverse is their only plausible way to explain fine tuning.

The popular version of the multiverse supports its first two premises by combining two theories: 1) eternal inflation, which implies the existence of an infinite number of parallel universes; and 2) string theory, which makes it somewhat plausible that the values of the constants vary from universe to universe.

The initial problem with an infinite varied multiverse in which everything possible happens is that it seems like a naive theory of the gaps that can explain any observation—even a voice from heaven proclaiming the fallacy of multiverse theory. To get around this problem, multiverse scientists need their third premise which claims that in an infinite varied multiverse, our universe is a typical universe with intelligent observers. This allows the multiverse to make its sole prediction that our universe is typical.

While this might seem promising at first, the problem is that in an infinite varied multiverse, not only does every possible universe exist, but there exist infinitely many identical copies of each universe. This makes it impossible to calculate probabilities in any straightforward manner and thereby evaluate if our universe is typical.

To fix this problem, multiverse scientists introduce a measure, an externally imposed rule that generates a “prediction” about the typical universe. However, the measure approach has serious problems—namely, measures are intrinsically ad hoc additions, all intuitive measures don’t work, and any contrived measure begs the question of what fine tuned and designed it.

Finally, because the multiverse deviates from the scientific method which rightly demands prediction and testing, it’s not a good scientific theory. Rather, the multiverse is a poor philosophical theory that’s incapable of explaining the fine tuning of the constants (or, for that matter, anything else). We are therefore left with the direct implication of fine tuning: An intelligent cause—God—fine tuned the constants of nature for the purpose of bringing about our one complex universe.

Comments

Share Your ThoughtsBe the first to write a comment.