Our universe’s initial state was incredibly unlikely, and if it were even slightly different, nothing would exist except for black holes. This essay discusses physicist Paul Steinhardt’s theory of Bouncing Cosmology which attempts to explain this unlikely beginning without requiring either an infinite varied multiverse or an intelligent cause. This theory is so wild and imaginative that we really had to struggle to come up with analogies to make it understandable to everyone. But we did. And these might just be our best analogies yet.
Highlights of this essay:
Below is an essay version of the ideas presented in Episode 15 of Season 2 of the Physics to God podcast. You can hear the audio version above.
Introduction
Other than the multiverse, there are two scientific theories that attempt to explain our universe’s highly ordered, low entropy initial conditions without an intelligent cause. One by Paul Steinhardt, a founder of inflationary cosmology, and the other by Roger Penrose which we’ll discuss in the next essay.
There are three reasons why it’s important to discuss these theories. The first is intellectual honesty. In essays 8 and 9 of series one, we presented our third argument for an intelligent cause based on the universe’s highly ordered initial conditions. We mentioned that the weakness of our argument from the initial conditions, as opposed to the arguments from the universe’s fine tuned constants and designed laws, is that there are two scientific attempts to explain the ordered initial conditions without invoking an infinite varied multiverse or an intelligent cause.
Even though mainstream physicists don’t accept either of these two scientific theories, they’re legitimate attempts by two of the best thinkers in modern physics to explain the highly ordered initial conditions. This and the next essay will justify why we think the inference to an intelligent cause is more compelling than either of these two theories.
Our second reason for doing these two essays is that these theories provide alternatives to the theory of inflation. If you recall from essay 3, multiverse scientists use eternal inflation to support the existence of an infinite number of universes. By presenting these two alternate theories, we want you to see that even though there’s scientific evidence to support the basic theory of inflation, it’s not smoking-gun evidence. Since there are other scientific theories that can explain the same observations without any inflation, this shows that eternal inflation isn’t a necessary consequence of current scientific observations.
Our third reason is that from even our brief discussion of these two theories, you’ll see just how speculative and imaginative they are. This will help convince you that while it may be possible to explain the initial conditions without an intelligent cause, it isn’t possible to do so in any straightforward manner. The fact that two of the best modern-day physicists put forth theories like these to explain our universe’s low entropy initial conditions will further illustrate the significant problem of explaining it away without recourse to an intelligent cause.
Because explaining these theories succinctly is one of the most difficult parts of this series, we’ll try to keep things as basic as possible. Despite this, the material involves some technical points that we couldn’t avoid or simplify. Nevertheless, we'll do our best to help the less advanced listener focus on the main line of reasoning without getting bogged down by the more complex details.
Cyclic Cosmology
Steinhardt and Penrose’s theories are particular models within the broader field of cyclic cosmology. To get a flavor of how these two modern cyclic theories work, it will be helpful to introduce the original Oscillating Universe theory proposed by physicist Alexander Friedmann in 1922.
After Edwin Hubble’s 1929 observation that our universe is expanding, the Oscillating Universe theory gathered more adherents, including Albert Einstein. The expanding universe posed a problem for many physicists. This is because by extrapolating backward in time, the natural implication of an expanding universe is that the universe had a beginning. To avoid this conclusion and its clear implication of a creator, some physicists embraced the idea of a cyclic universe which posited that the universe goes through an infinite series of cycles of expansion and contraction.
According to Friedmann’s proposal, each cycle goes through multiple stages: 1) a big bang; 2) an expansion phase; 3) a reversal; 4) a contraction phase; 5) a big crunch; and then the cycle starts again with: 1) a bounce, which is a new big bang…
The rough mechanism proposed for maintaining the cycling was as follows: the energy from the big bang causes a period of expansion that continues until the force of gravity slows it down and ultimately reverses the expansion, leading to a period of contraction. The contraction continues until it ends in a big crunch. After the crunch follows a bounce, or another big bang, and the cycle starts all over again.
According to the cyclic model, our apparent beginning is merely one “beginning” in an infinite series of cycles that have no ultimate beginning or end. For a few years, physicists thought that this solution might work.
Then, in 1934, physicist Richard Tolman exposed the major problem with this idea. Since the second law of thermodynamics states that for any closed system entropy always increases as time goes on, a cyclic universe leads to a build-up of entropy over the successive cycles, with each big bang having higher entropy than the previous cycle.
All you really need to understand for this essay is a very basic idea of entropy. You can think of entropy as a number that describes the amount of disorder in a system. The higher the number, the greater the disorder. Most importantly, you should know that entropy always increases as time goes on. For a more in-depth discussion about entropy, read or listen to essay 8 of series one.
Tolman demonstrated that, according to general relativity, the build-up of entropy over each cycle would cause each cycle to have a larger and longer-lasting bounce than the previous cycle. Extrapolating back into the past, each prior cycle would get successively smaller and shorter until we reach an original big bang of very low entropy – a true beginning.
This is obviously a problem for the cyclic universe, as its entire purpose was to avoid a beginning for the universe.
Tolman’s argument prevailed for more than 70 years, resulting in cyclic universes being essentially left for dead - until recently.
Entropy Density
In the 21st century, Paul Steinhardt and Neil Turok reintroduced cyclic cosmology in their theory of Bouncing Cosmology, presented in their 2007 book, Endless Universe. Unlike multiverse theories, Bouncing Cosmology is capable of making a scientific prediction (which is as yet unverified), and is therefore not subject to the devastating measure problem.
To appreciate Steinhardt’s solution to Tolman’s problem that a cyclic universe model inevitably produces a rise in entropy throughout its many cycles, we need to differentiate between the overall entropy of the entire universe and the entropy density of a particular place in the universe. The overall entropy tells us the total entropy of everything, while the entropy density is the amount of entropy per unit volume of space.
Once again, entropy measures the amount of disorder in a system. To understand the difference between overall entropy and the entropy density, let’s consider an analogy to the wealth in the world. One can speak of the world’s total wealth or of the average wealth per person. The total wealth in the world is like the total entropy in the universe, while the wealth per person is like the entropy density.
Of crucial importance is the fact that observers within the universe can’t see the overall entropy of the entire universe, but only the entropy density that is a characteristic of the part of the universe they can observe, or more technically the entropy density within their Hubble radius.
The Hubble radius means the observable universe, beyond which is the rest of the universe that we can’t see - in other words, our local neighborhood in the universe. Let's explain this with our simple analogy. While it may be difficult for a person to determine the overall wealth in the world, one can much more easily measure the average wealth of all people in their neighborhood.
Let’s now consider the relationship between the overall entropy of the universe and the entropy density in a limited region. Even though the overall entropy of the entire universe always increases (based upon the second law of thermodynamics), if the volume of space increases as well, then the concentration of entropy will be diluted and the entropy density can remain the same. If so, any observer would see the same low entropy environment around them.
Going back to our analogy, if the overall wealth in the world doubles but the number of people in the world also doubles, then the average wealth per person remains the same. If so, the increase in the world’s wealth won’t be noticeable in any particular neighborhood.
Based upon this distinction, Steinhardt worked out a way to get around Tolman’s problem that the build-up in entropy leads to cycles of longer and longer durations. Steinhardt noticed that it was not the increasing overall entropy that led to this problem, but the increasing entropy density. Therefore, if the volume of space continually increases as the entropy increases, then the entropy density and the duration of each cycle can remain the same. If so, extrapolating back in time could lead to an infinite series of cycles without a beginning, satisfying a key motivation for a cyclic universe.
Steinhardt explains this idea (on page 192) as follows:
As for avoiding Tolman's entropy problem, the solution depends upon branes, extra dimensions, and gravity, combined with a more refined understanding of Tolman's argument. As Tolman maintained, the entropy must increase from cycle to cycle according to the second law of thermodynamics, and this remains true for the new cyclic model. But increasing the total entropy is not what led to Tolman's ever longer cycles. Closer examination of his argument reveals that the problem was increasing the entropy density. In the particular model that Tolman considered, all the dimensions of space contract as the universe approaches the crunch. As a result, the new entropy plus any old entropy become highly concentrated at the crunch, resulting in a higher entropy density than in the cycle before. According to Einstein's equations of general relativity, it is the greater entropy density that produces the bigger bounce. If the entropy had somehow remained spread out, each bounce could have been identical to the one before.
Of course, there’s one obvious problem Steinhardt has to deal with. His idea depends upon the volume of space continually increasing as the entropy increases. But in a cyclic universe consisting of expansions and contractions, the overall volume of space is not continually increasing; during the periods of contraction, the overall volume of space decreases while the overall entropy increases. This should cause an increase in the entropy density and leads right back to Tolman’s problem.
The contraction phase would be analogous to a time when the amount of money in the world was increasing while the number of people was decreasing. This would clearly cause an increase in the average wealth per person.
This might seem like a dead end for Steinhardt’s solution to focus only on entropy density. How could the volume of space continually increase even during the universe’s contraction phases? Doesn’t contraction mean a decrease in space - not an increase in space? Steinhardt’s Bouncing Cosmology answers this apparent contradiction.
Bouncing Cosmology
Steinhardt’s theory was originally formulated within the framework of string theory and we’ll present it that way to stay consistent with how he presents it in his book. But even if you don’t like string theory, you should know that Bouncing Cosmology isn’t conceptually dependent on string theory. In fact, in an article called Bouncing Cosmology Made Simple (2018), Steinhardt presents a version that doesn’t rely on string theory at all. As our objective in this essay is to convey a basic idea about how Bouncing Cosmology solves the entropy problem, it will be counterproductive to go into the details and subtleties of the various ways Bouncing Cosmology can be formulated.
Instead, we'll try to stick to the essential elements, and explain it according to the best of our understanding. In Steinhardt’s proposed vision of reality, there exist two infinitely large parallel universes that are separated from each other by an unobservable extra dimension of space. The parallel universes are proposed to undergo an infinite series of collisions. Each time the universes collide, there’s what appears like a big bang within each universe. After each big bang, the two universes move away from each other along the direction of the hidden dimension. After a trillion or so years, the universes begin to move towards each other until they once again collide and bounce. The collision creates new hot matter and radiation, which appears like a new big bang.
With this picture in mind, Steinhardt makes a key move to help solve the entropy problem. If you recall, the problem was how can the volume of space be ever-increasing, even during the contraction phases. To answer this question, he suggests that the three regular, observable dimensions of space continually expand in both universes, while the cyclic contraction and expansion only occur in the unobservable extra dimension of space.
This resolves the aforementioned contradiction and allows the volume of both universes to keep increasing despite the contraction phase. The ever-increasing volume of the universe allows the entropy density to remain the same despite the fact that the universe’s overall entropy increases, thus avoiding Tolman’s entropy problem.
Before seeing Steinhardt's quote that explains this point, keep in mind a few things. First, Steinhardt’s theory is presented within string theory and the two colliding universes are called “branes”. So each time Steinhardt says “brane”, you can just think “universe”. Second, we’re going to explain what dark energy is soon, so don't worry if you don’t know what he means by that. Third, and most importantly, just try to get the main gist of what Steinhardt is saying without worrying about the details. It’s essentially what we just explained but some more of the physics details are spelled out.
Steinhardt says as follows:
In the new cyclic model, only the extra dimension contracts. Entropy is created on the branes - for example, at the brane collision or when galaxies and stars form - and then spread thinly during the rest of the expansion epoch. The accelerated expansion due to dark energy is especially effective in enabling gravity to make plenty of space to accommodate the new entropy and keep its concentration low. Then the branes continue to expand during the contraction phase - only the extra dimension contracts - so the entropy on the brane is never concentrated. When the branes collide and hot new matter and radiation are created, any preexisting entropy is exponentially diluted. Only the new matter and radiation are concentrated enough to affect the expansion. And since the amount is the same as in the cycle before, the duration of the next cycle is the same, as well.
Because that was a really complicated quote, let's try our best to give an analogy for what’s going on. Picture two huge ever-expanding rubber balls, connected by an elastic rubber band, that smash into each other, and then bounce away for a while until they once again start approaching each other for another smash. This happens over and over again, with each smash marking a new beginning in each ball, accompanied by the creation of hot new rubber on each ball. This is analogous to the universes continually increasing in size while the extra dimension that attaches them goes through repeated cycles of expansion and contraction.
You may be wondering how we’ll fit entropy density into this one.
Well, here it is. Imagine that as the balls expand, the money in them is increasing while the number of inhabitants is also increasing, such that the overall wealth per person always stays the same. That’s like entropy always increasing in each universe but the entropy density always stays the same.
Changing Dark Energy
While Steinhardt’s solution to the entropy problem is certainly creative, it still requires further assumptions about the mechanism that drives the cycles of contraction and expansion. To understand this, we’ll need to introduce the concept of dark energy.
Dark energy refers to a currently unknown form of energy, hence the term “dark”, that permeates all of space and causes the expansion rate of the universe to accelerate. It is generally accepted that some form of dark energy exists and is responsible for the accelerated expansion of the universe which was first measured by cosmologists in 1998. The simplest explanation for this dark energy is general relativity’s cosmological constant, a notion first introduced by Albert Einstein.
You may be bothered by the following: The cosmological constant and its associated dark energy is a fixed constant that quantifies the universe’s expansion rate. So what does this have anything to do with Bouncing Cosmology's cycles of expansion and contraction?
Well, despite current measurements of a constant expansion, Steinhardt speculates that the dark energy associated with the cosmological constant isn’t constant after all; rather, he suggests that it will change in the distant future. Not only will it change, but he conjectures that the dark energy will undergo repeated cycles of slow increase followed by slow decay. These proposed oscillations are the mechanism that drives Bouncing Cosmology’s expansion and contraction cycles.
Steinhardt (pg. 63) explains this point as follows:
In the cyclic model, accelerated expansion does not continue forever; if it did, a cycle would never end. A key assumption in the cyclic model is that the dark energy can decay: after a period of perhaps a trillion years, the physical properties of dark energy undergo a transformation that causes the expansion to slow down and eventually halt, leading to a phase of very gentle contraction. Once one accepts that the dark energy can slowly and smoothly decay, many interesting consequences follow.
The rate at which the dark energy changes doesn’t naturally emerge from any fundamental principle. The speculation that it changes at all is just an assumption, and the specific rate must be carefully chosen so that the contraction phase is very slow and gentle, allowing for the repeated production of universes like our own in each cycle. Therefore, Bouncing Cosmology requires some degree of fine tuning to work.
Returning to our rubber band analogy, as the rubber band is stretched, it stores more and more energy that can be converted into motion when it’s released. And, the way the energy changes depends on the specific properties of the chosen rubber band: for example, how thick and stretchy it is. So, in a sense, for Bouncing Cosmology to work correctly, its rubber band must be carefully chosen from a really big jar of all different types of rubber bands.
To be fair, while we didn’t focus on this point, the theory of inflation also requires a degree of fine tuning to work. In fact, Steinhardt argues that inflation requires even more fine tuning than Bouncing Cosmology. Our main point is that Bouncing Cosmology has certain key premises, such as the very fact that dark energy changes and its specifically chosen rate of change, that must simply be assumed without any supporting evidence.
Evaluating Bouncing Cosmology
Let’s review the three unsupported and highly imaginative claims that Bouncing Cosmology must make in order to explain our universe’s low entropy initial conditions without an intelligent cause.
There exists another unobservable universe parallel to our own, for which there’s no evidence.
There’s an unobservable hidden dimension connecting the two universes, for which there’s also no evidence.
The cosmological constant, or dark energy, is not as constant as we’ve observed it to be until now but will change in a specific way over the course of a trillion years, causing the universes that are now moving apart to slowly reverse course and eventually collide.
Even if all these unsupported assumptions are true, and Bouncing Cosmology works out, it can still only explain our universe’s highly ordered initial conditions. However, it doesn't begin to address series one’s primary argument for an intelligent cause - the fine tuned constants from quantum mechanics, such as the masses of the fundamental particles or the strengths of the fundamental forces. Also, it doesn’t address the design of the qualitative laws of nature, which was our second argument for an intelligent cause.
All in all, while Bouncing Cosmology is a valiant attempt to scientifically explain our universe’s low entropy initial conditions without an intelligent cause or a multiverse, it’s a highly imaginative theory that posits a tremendous amount with very little evidence. For this reason, we think that the incredibly unlikely initial conditions of the big bang remain a valid argument for an intelligent cause, as the far simpler, more straightforward, explanation for our universe’s initial state is that the same intelligent cause that fine tuned the constants and designed the laws of nature also arranged the proper initial state that would eventuate in our complex, structured, and ordered universe.
In the next essay, we’ll discuss Roger Penrose’s theory of Conformal Cyclic Cosmology which also attempts to explain our universe’s low entropy initial conditions without a multiverse or an intelligent cause.
Penrose is a genius and his theory is absolutely fascinating. But there’s no way to sugarcoat it: his theory is hard to understand, even for his fellow physicists.
Nevertheless, we’ll make sure to present it clearly so that you can get a basic idea of how it works. At the very least, you’ll see once again just how imaginative you have to be to explain the highly improbable big bang without an intelligent cause. So stay tuned!
Comments