The Boltzmann Brain problem - the multiverse's prediction that you are a randomly fluctuated brain with false memories - sounds like a ridiculous problem. In a sense, it is. But it's nevertheless a major challenge for the multiverse which claims we are typical observers.
The Grand Universe problem - the sheer splendor and grandeur of our immense universe - likewise presents a significant question for the multiverse which claims that our universe is typical.
These problems cast into serious doubt the third premise of the multiverse - that we are typical observers in a typical universe with intelligent life.
Highlights of this essay:
Summary of Last Two Essays
In essay 6, we showed how a naive multiverse that explains the laws, constants, and initial conditions based upon the idea that everything must exist somewhere in an infinite varied multiverse, falls prey to three serious problems.
We showed how these three problems are not merely problems for a naive multiverse attempting to explain fine tuning but are intrinsic problems with a naive multiverse trying to explain anything. Simply positing that in an infinite varied multiverse, everything happens somewhere and we just happen to be in the universe where fire-breathing dragons exist is bad science. That’s why no serious scientist actually believes in a naive multiverse.
And in essay 7, we showed that when we think more deeply about what an infinite varied multiverse actually predicts, we can see that it’s more sophisticated than just “everything has to happen somewhere”. Instead, if we really live in an infinite varied multiverse, then we would expect to observe ourselves in a fairly typical universe with intelligent observers. After all, if we’re trying to explain something by chance, then the odds are that we’d be in one of the most common universes with observers. This follows naturally from the logic of the multiverse as a theory that uses chance and probabilities to explain our observations.
By limiting the multiverse’s explanatory power to a typical universe, we explained last time how all three problems were naturally avoided. This showed why a complete multiverse theory capable of explaining fine tuning must establish the third premise that we live in a typical universe.
Is our Universe Typical?
In this essay, we’ll evaluate multiverse’s prediction that we are typical observers in a typical universe. The first serious, but admittedly funny-sounding problem is called the Boltzmann Brain problem. This problem is based on the fact that it’s not that hard to figure out what the typical observer in the multiverse looks like. The problem for the multiverse is that it doesn’t look at all like us - or at least what we think we look like.
After that, we’ll evaluate multiverse’s prediction in light of the second problem called the Grand Universe problem. This problem is based on the consideration of whether our massive complex universe with over 100 billion galaxies is a typical universe in which an intelligent observer would be expected to exist.
Boltzmann Brain Problem
Before we delve into the challenge that the Boltzmann Brain problem poses to modern-day multiverse theories, let’s begin with a brief history about it. This problem was first put forth in 1931 by physicist Arthur Eddington, as an attack against Ludwig Boltzmann’s multiverse theory that he first proposed in 1895.
Boltzmann was trying to explain the exceptionally high degree of order in our universe by chance alone. He proposed that our universe is akin to a large box containing trillions of particles that bounce around, repeatedly assembling and disassembling into different random arrangements. He suggested that the various arrangements over time can be viewed as different universes, and the set of all the universes together can be viewed as one big multiverse.
Boltzmann argued that given an infinite amount of time, the particles in the universe will eventually randomly arrange themselves into the highly improbable universe we observe. As such, Boltzmann suggested that we can explain our universe’s great order by chance alone.
While this may sound reasonable at first, Eddington saw a critical flaw in Boltzmann’s thinking. While it’s certainly true that every possible arrangement of particles will eventually occur, some arrangements will occur much more frequently than others.
To appreciate this point consider the following analogy. Let’s say that we continually shake a box with a hundred coins. If we periodically open the box and count the number of heads and tails, the most likely state we’ll observe will have around 50 heads and 50 tails. But if we continue this experiment for a long enough time, we’ll eventually observe all unlikely states as well. Nevertheless, not all unlikely states are equal. The state that has a hundred heads and zero tails will occur much less frequently than the state with ninety heads and ten tails.
Analogously, Eddington reasoned that in an infinite multiverse, two very different types of improbable arrangements would satisfactorily explain all of our observations of an ordered universe. The first, more intuitive type, is as Boltzmann suggested: all the particles in our universe are actually arranged into our entire ordered universe. In this case, we perceive order because it actually exists.
However, there’s a second, less intuitive arrangement that would satisfactorily explain all of our observations but would involve far less actual order, and is therefore much more likely to occur. In this arrangement, most of the particles in the universe are entirely disordered, but a “few” particles in a small region of the universe are coincidentally arranged into a single conscious brain. This brain’s neurons could be configured exactly as your brain is right now, leading to the perception and belief of an entire ordered universe, including a full population of other intelligent minds - but in truth, the brain would be surrounded by total chaos that it would have no method of observing.
While the brain wouldn’t last very long in an otherwise chaotic universe, it isn’t necessary to have a long-lasting brain to explain all your observations. Since the reason you believe you’ve existed for more than a moment is that your brain is configured in a special way that leads you to believe that you have memories of having lived for many years, even a short-lived fluctuation of an entire brain could have the exact same “false” memories.
While this might seem like a bizarre idea (and it is), it leads to a genuine problem, because it’s much more probable for a fluctuation to produce a single brain than an entire ordered universe, the multiverse would have many more fluctuations of single brains than ordered universes. This is analogous to the box of coins, where we would expect by chance to get 90 heads much more frequently than 100 heads. As astrophysicists, Luke Barnes and Geraint Lewis explain, “It is much more likely for a fluctuation to produce a fully formed brain, complete with memories of a body and life that never happened, than it is to fluctuate into a huge, life-bearing universe.”
Based upon this realization, the belief in an infinite varied multiverse that predicts we are typical observers necessarily leads to a clear, yet disturbing, conclusion: that the typical intelligent observer is a Boltzmann brain. That is, the most probable fluctuation that explains your current thoughts, memories, emotions, and so on, is that you’re an ephemeral fluctuation of a bodiless brain. There’s no need to posit a far more unlikely fluctuation that produced an ordered universe with billions of real human beings when all you really need to explain your observations is a single brain.
Why Does It Matter If You’re a Boltzmann Brain?
So maybe we are Boltzmann brains? Who cares? Maybe we just don’t like to feel that way but it’s true.
The conclusion that you’re likely a Boltzmann brain is not just emotionally disturbing and depressing – it indicates a devastating flaw in the theory that led to that conclusion.
There are two reasons why you have to reject any theory that leads to the conclusion that you’re a Boltzmann brain. First, on a most basic level, the idea that you are an ephemeral fluctuation of a brain with false memories is self-evidently absurd, and any theory that reduces to something absurd is clearly wrong. In Time Reborn, physicist Lee Smolin takes this approach. He says as follows:
This is called the Boltzmann brain problem: it implies that over an eternity of time there are vastly more brains in the universe which are formed from small fluctuations than brains arising in the slow process of evolution, requiring a fluctuation that lasts billions of years. So, as conscious beings, it is overwhelmingly probable that we are Boltzmann brains. But we know that we are not such spontaneous brains…So Boltzmann’s scenario turns out to be a classic reductio ad absurdum.
While Smolin says that, you may be wondering: is it really self-evident or is it just emotionally disturbing? Let’s say someone wanted to be stubborn and argue that you don’t really have any evidence against being a Boltzmann brain. How would you respond to that?
While we agree with Smolin that it’s self-evidently absurd, we know not everyone will be satisfied with that. For those people, there’s a second more compelling reason to reject any theory that leads to the conclusion that you’re merely a random fluctuation of one single brain with no other order or structure around it.
In his article Why Boltzmann Brains are Bad, physicist Sean Carroll writes as follows:
The right strategy is to reject cosmological models that would be dominated by Boltzmann Brains (or at least Boltzmann Observers among those who have data just like ours), not because we have empirical evidence against them, but because they are cognitively unstable and therefore self-undermining and unworthy of serious consideration.
Carroll’s point is that if I truly were a Boltzmann brain, then all my beliefs about physics could no longer be justified. After all, they’d all be based on false memories of observations that never actually occurred. But the rejection of the laws of physics undermines the entire motivation for believing in a multiverse theory in the first place (such as inflation and string theory) and along with that the entire motivation for assuming that I am a Boltzmann brain.
As silly as all this sounds, it presents a major problem for multiverse’s main prediction that we are typical observers in a typical universe. The Boltzmann Brain problem afflicts every theory that seeks to explain the apparent order in our universe as a coincidence that randomly occurred in one universe in a vast multiverse. Once a theory explains our highly ordered, highly improbable universe based upon a random fluctuation, it becomes far more likely that you are a fluctuation of a single brain, than one human amongst many in a vast, genuinely ordered, universe. In other words, in an infinite varied multiverse, the typical observer in the typical universe is not a human being in an ordered universe - but is rather a single Boltzmann brain in an otherwise chaotic universe.
This whole line of reasoning shows that the straightforward prediction of an infinite varied multiverse is that we’re in a typical universe in which the typical intelligent observer is a single Boltzmann brain. This is a patently false prediction because for one, it’s ridiculous and absurd, and two, it contradicts the very scientific theories, inflation and string theory, that allegedly justify the first two premises of multiverse theory.
So while it’s good that multiverse makes a prediction - that we are in a typical universe - it’s bad for multiverse scientists that this prediction is false. We’re clearly not typical observers in a typical universe.
The Grand Universe Problem
Let’s leave Boltzmann brains aside and for the moment assume that the typical observer is not a Boltzmann brain but an actual human being. This brings us to the second major problem that establishing the Typical Universe Premise poses for multiverse scientists.
Granted that human beings are typical observers in an infinite varied multiverse, but is our observable universe the typical universe that intelligent observers would expect to find themselves in? In other words, an infinite multiverse must not only explain how human beings, and not Boltzmann brains, are typical observers, but it must also predict that the rest of our known universe is the typical universe with intelligent observers.
While this isn’t so easy to figure out, let’s see how far we can get by intuitive reasoning. Let’s look around and see if our environment aligns with what we’d expect a typical observer in the multiverse to see. Remember, observer bias can only explain the existence of intelligent observers, but it doesn’t explain everything else besides intelligent observers.
Assuming the existence of an infinite varied multiverse, there will be many universes that contain actual intelligent observers. Some will have only one or a few lonely intelligent observers; others will have a star orbited by a planet that is full of intelligent observers; others will be much larger than that - besides these intelligent observers, they will contain an entire galaxy full of a hundred billion stars; and yet others, like ours, will be even larger than that - they will contain more than a hundred billion galaxies with a hundred billion stars each. While all these universes have intelligent observers, they have varying degrees of order and complexity alongside these intelligent observers.
Considering all these universes with intelligent observers, we can ask: is our grand universe typical? And the straightforward answer is that it certainly isn’t. The known universe is so astronomically large that it’s impossible to even get an intuitive grasp on its size. There doesn’t seem to be any need for all this excess order, structure, and complexity just to get intelligent observers.
But one may argue that this assessment isn't so simple. When we look around our immediate environment, alongside ourselves, we see an amazing planet, filled with water, plants, animals, and about eight billion other intelligent observers. While, at first glance, this may all seem unnecessary for one intelligent observer, a bit of reflection shows that it’s reasonable after all. A strong argument can be made that the most likely way to get one actual human being would be through a process like biological evolution on a planet that orbits a star, has liquid water, and has many other forms of life besides intelligent observers. In fact, it would be hard to imagine an intelligent observer naturally developing without the supporting environment that the Sun and Earth offer. If so, maybe we can explain Planet Earth, and even our entire solar system, as the typical way to get intelligent observers.
While that's true, the problem emerges when we look beyond our solar system and even beyond our galaxy. Biological evolution can only explain the plentitude of order on Earth and the solar system. But what about all the order beyond our solar system? Even if we grant multiverse scientists that for some reason the emergence of intelligent observers like ourselves needs the entire Milky Way galaxy, the rest of the universe seems entirely unnecessary for our own existence.
Are we really supposed to believe that the typical way of getting an intelligent observer also produces our entire complex universe that is more than 93 billion light years across? Out of all the universes with intelligent observers, our universe’s plentitude of order and structure beyond our planet certainly seems to render it quite exceptional. Given the enormous size and complexity of our vast universe, it certainly seems much more like the one gold marble than one of the million silver marbles.
Yet, multiverse scientists are stuck trying to justify the astonishing claim that our observed universe with its billions of galaxies is the typical universe containing intelligent observers. They must argue that just like the mechanism of evolution that’s responsible for the existence of people will typically occur in a solar system and will typically also generate other forms of life, so too the mechanism responsible for our solar system will typically occur in a galaxy with 100 billion stars and will typically also occur in a universe with over 100 billion other galaxies. So in a sense, they’re saying that our entire massive universe can be explained by the claim that it’s the most probable way to get human beings. While it’s possible, it’s certainly a highly unintuitive and quite astonishing claim.
A Historical Irony
Before moving on from this pretty shocking and unlikely claim, we’d like to point out that it brings an ironic twist to the historical disagreement between religionists and scientists regarding the importance of humanity in the overall universe. As opposed to religionists, scientists have historically downplayed the role and significance that humans play in the grand cosmos. Physicist Richard Feynman expressed this sentiment as follows:
It doesn't seem to me that this fantastically marvelous universe, this tremendous range of time and space and different kinds of animals, and all the different planets, and all these atoms with all their motions, and so on, all this complicated thing can merely be a stage so that God can watch human beings struggle for good and evil — which is the view that religion has. The stage is too big for the drama. - Gleick pg. 372
Besides Feynman, many scientists and philosophers throughout the ages have legitimately argued that the universe is so much greater and grander than a mere stage for human beings. While there is something special and unique about intelligent humans, it is hard to argue that the entirety of the cosmos exists for our sake alone.
While multiverse scientists obviously don’t say that an intelligent cause made the entire universe for the purpose of the ethical human drama (the position Feynman is rejecting), multiverse theory nevertheless explains fine tuning simply based upon its being needed for yielding intelligent observers. In their framework, the fact that the stage also happens to have galaxies, stars, planets, molecules, and atoms is an accidental consequence of those other objects being included in the typical way to get us humans.
In contrast to scientists’ ordinary historical attitude, multiverse scientists have assigned human beings a central role relative to all other features of our universe. As such, the line of reasoning used by multiverse scientists to explain the fine tuning, design, and order of our universe as a mere illusion is typically referred to as the anthropic principle, borrowed from the Greek ánthrōpos, meaning "human being".
On the other hand, our argument in the first series is based upon the complexity and structure of all the features of our universe (which includes but is not limited to, intelligent life). We agree with Feynman’s intuitive skepticism about the belief “that our whole universe is a stage so that God can watch human beings struggle for good and evil.” Therefore, we’re not arguing that God fine tuned the constants for the exclusive purpose of life.
Instead, we’re arguing that an intelligent cause fine tuned our entire universe for the purpose of bringing about all the vast complexity and structure in our universe including atoms, molecules, planets, stars, galaxies, and life. Because this argument doesn’t demand or imply the exclusive centrality of human life in particular, we don’t employ the “anthropic” terminology that is used by multiverse scientists.
The Need for Justification
While that historical irony was an interesting point, let’s return to our main line of reasoning. In the last essay, we explained that an infinite varied multiverse actually makes a prediction - that we’re typical observers in a typical universe. In this essay, we assessed the truth of this prediction and discovered two major problems.
First, an infinite varied multiverse seems to predict that the typical intelligent observer is nothing like us but is rather a temporary fluctuation of a brain. Second, multiverse scientists claim that the rest of the universe in all its grandeur is the typical way for a universe to have intelligent observers. But, even a cursory glance at the grand universe we occupy would seem to indicate that it’s so much grander and greater than merely being the necessary environment for human beings. Together, these two problems seem to falsify the only prediction an infinite varied multiverse makes.
While these two problems give the strong impression that the multiverse’s only prediction is false, they don’t quite prove it. After all, we can’t actually observe other universes to empirically determine what a typical universe looks like. So you might think that since we can’t prove their prediction wrong, maybe multiverse scientists can just claim that we are, in fact, typical observers in a typical universe.
The critical question is: Do multiverse scientists need to provide real concrete justification for the Typical Universe Premise, or is it enough for them to postulate, without any evidence, that we’re typical observers in a typical universe?
As you might have expected, the answer is that they must justify it. We have two arguments for this.
The first follows from our discussion at the end of the previous series and the beginning of this series. Since fine tuning indicates an intelligent cause of our universe, the burden of proof is shifted to multiverse scientists to support all three premises of a complete multiverse. Just as they needed to provide evidence - like eternal inflation and string theory - that there exist an infinite number of universes and that the constants vary from universe to universe, so too they must provide evidence, not just speculate, that our universe is a typical universe with intelligent observers.
Besides this point, there’s an additional reason why multiverse scientists need to justify the Typical Universe Premise in particular. To appreciate this point, let’s recall the last few essays. We considered a naive multiverse theory that attempts to explain fine tuning by an infinite varied multiverse without the Typical Universe Premise. We showed that such a theory fails because of the following three serious problems:
1) It’s a theory of the gaps which can be used to explain anything, and therefore explains nothing at all;
2) It’s an intrinsically irrefutable theory that is even capable of explaining away blatantly contradictory evidence;
3) It’s a self-defeating theory that undermines the very scientific theories that it’s based upon.
Ultimately, we showed that adding the third premise, that we’re typical observers in a typical universe, would save multiverse and enable it to explain fine tuning.
Now here’s the critical point. If multiverse scientists simply make the unintuitive claim that we’re typical observers in a typical universe, without any justification, then they fall right back to the three problems with a naive multiverse. Here’s why:
If one is simply willing to posit, without any further justification, that our fine tuned universe is typical, then they could just as well posit the same for any gap in knowledge. For instance, if one were trying to explain the diversity and origin of life, there would once again be no need for the theory of evolution. Just posit, without any justification, that we’re in an infinite varied multiverse in which the typical universe with intelligent observers has all the diverse forms of life we observe on our planet.
Likewise, if one were trying to explain something like a heavenly voice, which was an example we offered in essay 6, just posit that we’re in an infinite varied multiverse in which the typical universe with intelligent observers contains a heavenly voice.
Finally, if one formulates a scientific theory that’s contradicted by an experimental result, just posit that we’re in an infinite varied multiverse in which the typical universe with intelligent observers has this experimental result. Since there’s nothing that can’t be explained by this strategy, it becomes equivalent to naive multiverse and therefore explains nothing at all.
The only way multiverse scientists can avoid the three problems of a naive multiverse is by making a genuine prediction. Of course, the problem is that when they attempt to actually compare this prediction to reality they run into the Boltzmann Brain and Grand Universe problems that seem to show their prediction is false.
And, again, if multiverse scientists want to suggest that since we can’t absolutely prove their prediction is false, maybe it’s true, then they’ve undermined everything they would have gained with a real prediction. Simply positing, without any justification whatsoever, that we’re typical observers in a typical universe is tantamount to admitting that the multiverse is incapable of making any concrete genuine predictions. Such an approach reduces the theory to a fatally flawed naive multiverse that can be used to explain anything at all, and therefore completely fails to explain fine tuning.
Because multiverse scientists acknowledge this point, they’ve worked hard to justify their shocking prediction that we’re typical observers in a typical universe using something called measures. In the next essay, we’ll explain what measures are and assess the results of this hard work. We’ll see how multiverse scientists’ attempt to make a prediction necessitates the use of measures - and that will lead directly to the devastating measure problem. So stay tuned!
