This essay discusses something many people fail to mention about the multiverse - that it actually makes a prediction.
Well, sort of. You can probably guess that a theory about an infinite number of unobservable parallel universes can’t make a conventional prediction in the standard scientific sense. But with this “prediction”, you’ll see how multiverse scientists can avoid the three problems that plague a naive multiverse.
While this essay is incredibly important for understanding the multiverse, it’s more complicated than the last one. But don’t worry - we have a marble analogy that makes everything crystal clear.
Highlights of this essay:
Reviewing the Three Problems with a Naive Multiverse
Let’s begin by reviewing what a naive multiverse is and how it superficially solves the problems of design and fine tuning of the laws and constants of nature.
A naive multiverse theory attempts to explain fine tuning and design using only the first two premises of multiverse: the infinite multiverse premise and the varied multiverse premise.
The reasoning goes something like this. If there are an infinite number of universes - each with different laws, constants, and initial conditions - then every possible universe must exist somewhere in the multiverse - including a universe that’s as fine tuned, designed, and ordered as our own. And, of course, we shouldn’t be surprised to find ourselves in a universe hospitable to intelligent life, because there simply aren’t any observers in universes that aren’t hospitable to life.
In the previous essay, we showed that naive multiverse falls prey to three serious related 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.
Notice how all three problems are based on the fact that a naive multiverse makes no specific predictions. It merely explains all possible observations by recourse to the claim that “everything happens somewhere in an infinite varied multiverse.”
If, on the other hand, it predicted and explained certain things and not others - in other words, it had only limited explanatory power - then none of these problems would apply. It wouldn’t be a theory of the gaps, it could be contradicted by a false prediction, and it wouldn’t undermine science because it would allow predictions to be compared to observations.
We’ll come back to answering these three problems later in this essay. First, let’s introduce the typical universe premise and show how it leads to the one prediction that an infinite varied multiverse can actually make.
The Typical Universe Premise
The third and final premise of a complete multiverse theory is the claim that from the set of all the universes that have intelligent observers, our universe is typical.
This last premise is far more complicated than the other two, so let’s take it slowly. We said two things. First, we mentioned the set of all universes with intelligent observers. Second, we said that from all universes in this set, our universe is typical. Let’s take these one at a time. First, let's explain what we mean by the set of all universes with intelligent observers.
Consider the following: Out of the vast number of universes in the multiverse, only a small subset of them have fine tuned constants that allow for the existence of intelligent observers. While all these universes with intelligent observers are different from one another, they share in common the property of having intelligent observers. Let’s consider some examples of universes in this subset: one universe has just one intelligent observer and nothing else; another has one planet with intelligent observers and nothing else; another has one entire galaxy and intelligent observers; and yet another universe (like ours) has over 100 billion galaxies with over 100 billion stars each, in addition to intelligent observers. The list of different universes with intelligent observers goes on and on.
Now that it's clear what we mean by the set of all universes with intelligent observers, let's try to understand the claim that our universe is typical in this set. Let's first explain what we mean by typical. On a basic level, we just mean a common type of universe. More specifically, if we analyze the set of all universes with intelligent observers, we can study their properties and divide them into groups. For example, we can form one group consisting of all the universes with intelligent observers and fire-breathing dragons and another group consisting of all the universes with intelligent observers but no fire-breathing dragons. Then we can compare the size of these two groups and see which type of universe is more common - one with or without fire-breathing dragons. It’s intuitively reasonable to say that those without fire-breathing dragons are far more common. In this sense, we say that the typical universe with intelligent observers doesn’t have fire-breathing dragons.
While we grouped universes based on whether or not they have fire-breathing dragons, there are many different ways to group universes. No matter how we group them, some groups will have many members while others will only have a few. A typical universe is one that belongs to a group with many members relative to the other groups. On the other hand, universes from groups with only a few members are considered atypical.
It’s actually a bit more complicated than this. While it might seem obvious that universes that don’t have fire-breathing dragons are far more common than those that do, it isn’t so simple. Since there are an infinite number of universes in the multiverse, there are actually an infinite number of universes in both groups - those with fire-breathing dragons and those without them. This is a major problem for an infinite multiverse. Comparing sizes of infinite groups is highly problematic and leads directly to something called the measure problem.
But we’ll get to the measure problem in a later essay. In this essay, we need to ignore the problem infinities pose to the multiverse and assume that it’s possible to make relative comparisons between infinite groups of universes. Otherwise, we’ll never be able to clearly explain the Typical Universe Premise.
To help appreciate the idea of a typical universe, consider the following analogy. Imagine we consider the set of all professional basketball players. Of course, they all have different heights. Some are very short and some are very tall. But it would be fair to say that a typical player from this set is taller than 6 feet. This doesn't mean that all players are this tall, but since the overwhelming majority of them are, we can say that the typical player is more than 6 feet tall.
Now that we understand the meaning of the set of all universes with intelligent observers and the meaning of a typical universe, we can put them together to understand the meaning of the Typical Universe Premise. It claims that from the infinite subset of universes that do have intelligent observers, our universe is typical. In other words, our universe is like a typical basketball player who is taller than 6 feet, and not like one who is shorter than 6 feet.
A Typical Universe is a Prediction of Multiverse
Now that we understand the Typical Universe Premise, let’s see why its claim that our universe is typical is an actual prediction of an infinite varied multiverse.
Since this is a complicated point, we'll keep going slowly and clearly explain it.
If we’re trying to predict something based on chance, then, by definition, the odds are that we’ll observe something probable as opposed to something improbable.
For example, if we were to randomly pick a professional basketball player, we would predict that he would be taller than 6 feet. Though there are a few shorter players, we would never predict that we would pick one of these rare players.
Similarly, if we really lived in an infinite varied multiverse, then we would predict that we would observe ourselves to be in a typical universe. Though there are some atypical universes with intelligent observers, we would never predict that we would exist in one of these rare universes.
Though there are universes with fire-breathing dragons, we’d never predict that we’d be in one of these rare universes. Notice that this point naturally follows from the logic of multiverse as a theory that uses chance and probability to explain our observations.
It’s critical to realize that observer bias in no way affects the prediction of which universe with intelligent observers we would be in because all universes in the set under consideration have intelligent observers. While it is reasonable to invoke observer bias to explain why our universe has some intelligent observers, observer bias can’t explain anything in our universe that doesn’t depend on intelligent observers. Instead, the prediction of which type of universe we would expect to observe is exclusively based on which universes are typical.
This is a subtle point. Let's try to explain it a little bit more. Out of the vast number of universes in the multiverse, only a small subset of them contain intelligent observers. Of course, it’s unreasonable for us to ask why we happen to be in this small subset since there simply are no intelligent observers in the many other universes to even ask this question. Observer bias demands that we’re in a universe with intelligent observers.
But not all universes with intelligent observers are the same. And observer bias can’t tell us in which type of universe with intelligent observers we would expect to find ourselves. All we can say is that probabilities predict that we would expect to be in a typical universe within the subset of universes with intelligent observers.
While many popular presentations of the multiverse fail to mention this critical point, multiverse scientists, some of the smartest people in the world, of course, realize it. When they present the multiverse to popular audiences they sometimes leave it out, but in their more rigorous discussions, they acknowledge the point.
In the words of Brian Greene in The Hidden Reality (pg.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 Marbles Analogy
We’re finally ready for your marble analogy. Imagine a massive container that’s filled with many many marbles: it has a trillion white marbles and one gold marble. The bottom of the container has an opening that can let out one marble at a time. If one marble is let out, obviously we would expect it to be a white marble. Let’s assume that it happens to be the single gold marble. Would chance be a good explanation?
Of course not. The odds of that are one in a trillion. Instead, we would look for some other explanation - maybe it’s rigged or something like that.
Now, let’s assume that upon further analysis of the container and the marbles, we realize that the marbles are not all the same size - the white marbles are all large, and the gold marble is small. Additionally, the opening at the bottom of the container is medium-sized. Now you don’t even need chance to explain the appearance of the one gold marble. After all, the only marble that can possibly fit through the medium-sized opening is the small gold marble. Even though there are a trillion large white marbles, a selection bias prevents them from coming out.
Let's now carry the analogy over to the multiverse. The container full of marbles is analogous to a multiverse full of universes. The trillion large white marbles correspond to the overwhelming majority of universes without any intelligent observers. The gold marble corresponds to a universe with intelligent observers. While the overwhelming majority of universes don’t have any intelligent observers, a selection bias prevents us from ever observing them. Therefore, we shouldn’t be surprised to live in a universe conducive to intelligent life. This is essentially the anthropic principle.
Let's now adjust the analogy a little bit. Besides the trillion large white marbles and one small gold marble, there are also a million small silver marbles. Now, what would we expect if we let one marble out of the medium-sized opening? Once again, since the large white marbles can’t fit out of the medium-sized opening, a selection bias determines that we can only get a small gold or silver marble. However, since there are a million silver marbles and only one gold marble, we would obviously predict a silver marble. Since the selection bias doesn’t prevent us from getting silver marbles, chance would almost definitely lead to us getting a typical precious marble - a silver one. If, on the other hand, the single gold marble emerged, chance would be a very poor explanation. Since the odds against it are a million to one, we’d seek an alternative explanation.
Once again, let's carry the analogy over to the multiverse. The trillion large white marbles correspond to the overwhelming majority of universes with no intelligent observers - universes that by definition can never be observed. The million small silver marbles correspond to all the typical universes that have intelligent observers but don’t have any rare improbable phenomena such as fire-breathing dragons, unicorns, or voices coming from heaven. The small gold marble corresponds to a special universe containing intelligent observers and also one of these rare improbable phenomena.
Even though there are many more universes without intelligent observers, an observer bias prevents us from seeing any of them. Our options are therefore limited to universes with intelligent observers: either a special universe with fire-breathing dragons or one of the less special universes that have intelligent observers but no fire-breathing dragons. According to the laws of probability, we would certainly expect to find ourselves in a universe with intelligent observers but no fire-breathing dragons.
Now, if we observe our universe and find it to be typical - one without fire-breathing dragons - then chance alone could be a good explanation. However, if we find ourselves in a rare atypical universe – one with fire-breathing dragons - then positing an infinite varied multiverse and chance is a poor explanation. It defies the odds! This problem can’t be circumvented by an appeal to the anthropic principle and an observer bias. This is because the observer bias doesn't demand fire-breathing dragons.
To appreciate why, let’s consider the type of question that can be answered by invoking an observer bias. Imagine we were faced with the following question: given that, relatively few universes in the multiverse have intelligent observers, what are the odds that we happen to be in such a universe and not one of the many more universes without intelligent observers? The answer to this question is that observer bias ensures that we must be in a universe with intelligent observers.
This was akin to the medium-sized hole at the bottom of the container that ensured we picked a small gold or silver marble, but not a large white marble.
But if we see a fire-breathing dragon in our universe, we face an entirely different question: given that there are numerous different universes that all have intelligent observers, what are the odds that we find ourselves in a rare atypical universe with fire-breathing dragons and not a typical universe that has no fire-breathing dragons? This can’t be answered by simply invoking an observer bias because all universes we’re considering have intelligent observers who are capable of observing them. While an observer bias restricts us to a universe with intelligent observers, it still predicts that we should find ourselves in a typical universe – not one with fire-breathing dragons. If we instead observed ourselves to be in a rare atypical universe, we would discard the infinite varied multiverse theory for having made a false prediction. Our main point is that the explanatory power of an infinite varied multiverse is limited to explaining typical universes.
Similarly, since chance predicts that it’s a million times more likely to get a silver marble than a gold marble, if we did get the gold marble, our prediction would be falsified and we would pretty much discard chance as a viable explanation. And even if we wouldn't entirely reject chance because one in a million events sometimes do occur, if the odds of attaining the gold marble by pure chance were instead one in a trillion trillion trillion trillion trillion, then getting the one gold marble would certainly outrule chance as a viable explanation.
Avoiding the Problems
The main takeaway of the analogy is that an infinite varied multiverse actually predicts something - that the universe we observe should be like the typical silver marble and not like the atypical gold marble. Since this is the case, multiverse’s explanatory power is limited - it can only explain a typical universe. Let’s see how this limitation solves the three problems that afflict a naive multiverse.
The first problem was that a naive multiverse is a theory of the gaps that can explain anything and everything. However, once you limit a multiverse’s explanatory power to typical universes, this is no longer the case. This is because an infinite varied multiverse predicts that we’ll only observe a typical universe with intelligent observers. As such, a multiverse can’t fill all gaps and explain anything, but only those observations that can be shown to be expected in a typical universe.
For instance, a multiverse can’t naively explain the diversity of life on Earth by merely claiming that it must occur through a quantum fluctuation in at least one universe in the infinite varied multiverse. Rather, it must show that the diversity of life can be expected in the typical universe with intelligent observers.
The biological theory of evolution can try to establish this by arguing that the most likely way to get intelligent observers is through a process of natural selection that won’t only bring about intelligent observers like us but will also produce a large diversity of living organisms throughout our planet.
The second problem was that a naive multiverse is intrinsically irrefutable, capable of explaining away even contradictory evidence. Once again, after limiting multiverse’s explanatory power to a typical universe, this is no longer the case. While it’s still true that a small percentage of the infinite universes will have intelligent observers who hear heavenly voices denouncing the existence of a multiverse, it’s much more likely that an observer will find himself in a universe without such a voice. Thus, if one were to hear this heavenly voice, a phenomenon that wouldn’t be predicted to exist in a typical universe, this would be considered refuting evidence. As such, a complete multiverse that includes the Typical Universe Premise is at least theoretically refutable.
While the possibility of a heavenly voice refuting the multiverse is still a far cry from scientists’ ordinary demand that a scientific theory be falsifiable, nevertheless, a complete multiverse theory avoids naive multiverse’s even greater problem of being intrinsically irrefutable.
The third problem was that a naive multiverse is self-defeating, in that it undermines the very scientific theories upon which its first two premises are based. This problem emerged from the claim that since we are in an infinite varied multiverse, every possible experimental outcome could be expected – a result that makes it impossible to test any particular theory. However, once we limit multiverse’s explanatory power to typical universes, this problem is also removed.
This is because probabilities would predict that we should only observe phenomena that would be expected in the typical universe. As such, a theory can be tested by predicting what would be expected in the typical universe and comparing its predictions with experimental results. If this theory is thereby confirmed, its supporting evidence would no longer be undermined by belief in an infinite varied multiverse.
An Unconventional Prediction
Even though a complete multiverse solves the three problems and makes the prediction that we live in a typical universe, it’s important to notice that this prediction isn’t a conventional scientific prediction.
In order to confirm a theory, scientists usually use the theory to make an unexpected prediction that they can go out and test for, or at the very least to find a type of prediction that explains something that scientists had already observed but didn’t previously understand.
For example, Einstein’s theory of relativity predicted that gravity would bend light rays. That was an unexpected surprising prediction that scientists would have never guessed without relativity. Therefore, when physicists measured the bending of light during a solar eclipse, that was considered strong evidence in support of relativity. Additionally, Einstein’s relativity predicted certain anomalous features about the orbit of Mercury that were already known to exist but were unexplained by any other theory. Since general relativity naturally explained Mercury’s anomalous orbit, that too counted as strong confirmation of Einstein’s theory.
On the other hand, a multiverse merely predicts that we live in the typical universe with intelligent observers. If that’s right, it would amount to little more than saying that multiverse predicts we should observe everything we have already observed and expect to keep observing. So, while this unconventional prediction makes a complete multiverse theory better than a naive multiverse theory, it’s not quite up to the standard of prediction that science has conventionally demanded in order to confirm and verify a theory.
However, multiverse’s unconventional prediction that we live in a typical universe does have value insofar as it allows multiverse to be falsified. That is, if we could show that our universe is not typical and therefore not explainable by chance, then we would know that the multiverse is false because it made a false prediction.
Of course, all this is only the case because we are ignoring the problem of infinities and the measure problem. Once we address that, we’ll see that the multiverse doesn’t even make this extremely limited type of prediction. More on that in a later essay.
Conclusion
This essay has shown that the multiverse makes the falsifiable prediction that we live in a typical universe with intelligent observers. Therefore, the natural next step is to evaluate whether the Typical Universe Premise is true. In other words, we must evaluate multiverse’s claim that our universe is like one of the typical silver marbles.
If multiverse scientists can establish that our universe is typical, then an infinite varied multiverse plus chance would provide a good explanation for our fine tuned universe. But if it turns out that our universe is atypical and special - more like the gold marble - then the theory of an infinite varied multiverse has made a false prediction and should therefore be rejected.
We now face the critical question: Is our universe typical? You might think this is impossible to determine - after all, we can’t possibly observe what all the other universes look like. But, in the next essay, you’ll be surprised to hear two compelling arguments that indicate our universe is not at all typical. It’s much more special than any run-of-the-mill universe with intelligent observers. So stay tuned!
Comments