pbsspacetime
patreon
Will the Big Bang Happen Again (And Again)? (Script)
🕑 Added 2025-02-28 18:00:06 +0000 UTCHow did the universe begin? How can something come from nothing? One way to “solve” this most difficult of philosophical conundrums is to avoid it altogether. Maybe the universe didn’t begin at all. Maybe the Big Bang was just one in an endless cycle.
On February 11th, 2015 a new show called PBS Space Time appeared on YouTube. In the 10 years since, together we've explored the insides of black holes and ventured across the edge of the universe and seen the beginning and end of time and the peeked at the underlying clockwork of nature. It's been brain-breaking and existentially humbling journey. And we're so happy you have been on this ride with us. Well, that ride continues. We have exciting new plans for year 11 that will be revealed in due course. For now, two things you can do to help us get the anniversary celebration started. First, post in the comments to say how long you've been watching ... and if you joined more recently how much of the back catalog of over 400 videos did you actually manage to get through. And second, merch! We’ve got a limited edition 10 year anniversary design as well as some classic logo merch. Having this gear doesn't just make you incredibly cool, you also helps us keep going for another decade.
No one wants the end of the universe. Maybe that’s why the idea of a cyclic universe is so appealing. So much so that it appears in mythical cosmologies of Hinduism, Buddhism, Zoroastrianism, and the cosmologies of the Norse, Mayans, Egyptians, Greeks, and no doubt more. It makes sense. We tend to extrapolate from the patterns we see in nature. We see recurring cycles of day and night, of the seasons, of life and death. Why shouldn’t the entire universe go through cycles?
Of course, the evidence has to support our extrapolations, and we now know that the evidence does not support cyclic cosmologies. Right? The universe and time itself started at the Big Bang and space will expand forever and that’s it. Or so says mainstream cosmology. But apparently cyclic cosmology has not lost its appeal, because scientists have found a way to make it work for our universe.
The original Big Bang model has all of space originating in an infinitesimal point at the beginning of time, and expanding from there. This fits a lot of observations of our universe.
The recession of the galaxies reveals the expansion of space, and the cosmic microwave background is pretty clearly the afterglow of an early hot, dense state. But some observations aren’t so easily explained. For example, in our universe matter and energy are very evenly spread out, but in the basic Big Bang model there wasn’t time for this smoothing to happen before the expansion threw distant regions beyond causal contact. This is the horizon problem. There’s also the fact that a basic-Big-Bang universe isn’t expected to be so perfectly flat, which requires an uncannily perfect balance between matter and dark energy. And if the universe that began in an extremely hot, dense state, than certain relics of this phase should persist—so-called magnetic monopoles.
Of course we’ve talked about the horizon, flatness, and magnetic monopole problems, as we have the most popular solution—cosmic inflation. This proposes an extreme, exponential expansion phase in the extremely early universe. Inflation becomes the bang in the big bang, and it allows an initially smooth universe to be expanded beyond causal contact, as well as being nicely flattened space and scattering those pesky monopoles far enough apart that they're unlikely to ever be seen.
To top it all off, inflation explains how the universe got its large-scale structure. It predicts quantum fluctuations in the inflaton field, which became the gravitational seeds that grew into galaxies and galaxy clusters. Inflation goes further, predicting the those fluctuations should lead to the same level of lumpiness at all size scales—so-called scale invariance. And that’s exactly what we see in the lumps of the cosmic microwave background.
Inflation does such a good job that it’s practically mainstream, but there are some downsides. Modern versions of the idea predict that if inflation happened, then it never ended. Sure our little patch quit with that extreme growth, spawning our much-slower expanding universe. But as long as inflation didn’t stop everywhere all at once, then out there, somewhere, this eternally inflating greater universe is blowing up forever, constantly spawning bubble universes. Some find this idea a little extravagant.
The other issue with inflation is that it doesn’t avoid a beginning of time, nor a point of infinite density—a singularity at that beginning. You can push that singulary as far back as you like, but in inflationary models it has to be there, just like in the regular Big Bang. And any theory that predicts a singularity is automatically suspect.
And even if inflation didn’t have its issues, it’s worth exploring other options. So what about the option where, instead of asking what happened at the beginning of time, we ask what happens if time never had a beginning? Cyclic cosmologies exist in many ancient traditions, but also in modern cosmology. Soon after we noticed that the universe was expanding, scientists came up with models in which the universe eventually slows and starts contracting, then bounces into a new Big Bang, and repeats over and over. But these models didn’t manage to do away with the beginning of time because both entropy and the lifespan of the universe had to increase with each bounce. That means we couldn’t extrapolate the bouncing back in time indefinitely. There had to be a first. And anyway, these cyclic universes didn’t solve the problems that inflation solves.
But it turns out that cyclic cosmologies can give us everything we want. To explain the horizon, flatness and monopole problems without inflation, and at the same time eliminate that pesky beginning of the universe. In fact, it turns out that the same type of field that causes inflation can also be tweaked to give an infinitely regenerating universe. This is the idea of the ekpyrotic universe—named after the cyclic cosmology of the ancient Greek Stoics, in which the universe is rebirthed in fire—ekpyrosis—between unending cycles. By the way, this is a very different idea to the conformal cyclic cosmology of Roger Penrose, and I refer you to our previous episode for that equally awesome proposal.
The ekpyrotic universe was first proposed in 2001 in a paper by Justin Khoury and collaborators.
Now, I’m going to come back to the ideas of this paper, but first I want to give the part of the story told in a followup paper by two of the original authors. In 2002, Paul Steinhardt and Neil Turok showed how the same type of quantum field proposed to cause inflation—the inflaton field—can be tweaked to resurrect the universe rather than blowing it up.
The extreme accelerating expansion of inflation is driven by the same type of quantum field as we think now drives the relatively chill acceleration that we attribute to dark energy. It’s a scalar field—the simplest type of quantum field in that it’s just a simple numerical property—a field strength—in space everywhere. The field also has a potential energy associated with that numerical value. Sometimes the relationship between field value and its potential energy is simple—stronger the field the more the energy, sometimes it's complicated. For example, in some versions of inflation, the field value slowly drops and energy decreases, but then the energy reaches a minimum value and any further change in the field would add more energy. Therefore the field becomes stable and inflation stops.
But there are lots of different ways you can relate the field energy to the field value. And, as shown by Steinhardt and Turok one of those ways gives you a cyclic universe.
We normally think of dark energy as being due to a constant energy density everywhere in space that does not change over time. But maybe dark energy changes only very slowly. For example, if there’s this quantum field that is slowly changing in value—say, decreasing—with barely noticeable changes in the associated energy. Then we have accelerating expansion far into the future even as the field value drops. But eventually the energy in that field fades and becomes negative, and as that happens acceleration slow, then expansion slows and the universe briefly halts.
And then the universe recollapses. The field potential bottoms out in a minimum and rises back to zero. Now, we might expect the field to get stuck in that minimum, however during this contraction, gravitational potential energy is converted into kinetic energy of the field so that the field value blows past this minimum. In the final phase of contraction the field has no potential energy—no “dark energy” equivalent—and the kinetic energy of the field gets converted into radiation.
This liberates the universe from the constraints of this quantum field and it quickly starts expanding again. The radiation spawns matter and then dissipates, the matter dominates for a while, and finally dark energy takes over and we find ourselves back where we started in the roughly the modern era.
So how does this version of a cyclic universe solve all our problems? Well, the magic happens in the contraction phase—what the authors call the ekpyrotic period. When the universe was at its largest, matter was so far-flung that the only meaningful energy in the universe was in its scalar field, and the only meaningful structure in the universe were the quantum fluctuations in that field. Now as the universe collapses very slowly, those fluctuations are amplified. The shape of the scalar field is tuned so that these fluctuations have a scale-invariance—equal frequency for all sizes of lumps, just as is observed in the CMB and as is predicted by inflation.
That contraction phase also smooths out the universe, solving the horizon problem. And the universe does not reach arbitrarily high temperatures, so no magnetic monopoles ever need to be created. There’s no singularity, and there’s no significant difference between one cycle and the next, so this model is consistent with cycles extending back in time forever.
Steinhardt and Turok claim that the scalar field needed to achieve all of this magic is no more finely tuned than the field needed to achieve inflation, and so this cyclic model is just as plausible as the inflationary model because they both result in the same observables—at least as far as current observational sensitivity allows. There are ..
But is there any more motivation to believe that a field of the needed variety actually exists? Well, these guys say yes, and the mechanism was proposed the year prior by a team including these authors and led by Justin Khoury.
This does require a little more than a modification of the inflaton field. It requires an entire new dimension of space. 6 new dimensions really. One motivation for the type of scalar field needed, with its particular potential energy curve, lies within M-theory. This is an encompassing framework for string theory, in which our universe, with its 3 dimensions of space and 1 of time, exists as a single slice in a greater object with 4 large spatial dimensions. And with 6 coiled, compact dimensions, but we don’t need to worry about those for this description.
Our universe would be something called a brane—short for membrane—living within the higher dimensional space—itself called the bulk. In this picture, our universe is one of the boundary layers of the bulk. We call it the visible brane. Normally this brane is just chillin—it’s pretty empty and is not changing in size. Things get interesting when another free-floating brane within the bulk—called hidden brane—smashes into us. Which, apparently, is something that can happen. Try not to let it keep you up at night. This collision dumps a bunch of energy into the visible brane, sparking a big bang.
This connects to the description of the scalar field because we can interpret the value of that field as the distance between the visible brane and this incoming hidden brane. So on the graph we saw earlier, movement to the left—decreasing field value—corresponds to decreasing distance between the branes.
Khoury paper assumes a simpler form of the potential than Steinhardt and Turok, with energy decreasing exponentially as the branes approach, which is like a purely attractive force between the branes. In the more complex version where potential energy decreases then increases again, we have attraction then repulsion of the branes. Either way, when the branes collide, the energy is dumped into the visible brane causing space there to start expanding. The hidden brane recoils, propelled back the way it came with the energy of the bounce until eventually it’s pulled inwards again.
In this interpretation, the quantum fluctuations of matter manifest as wiggles in the incoming hidden brane. These result in different parts of that brane arriving at different times, and so there are variations in the start time of expansion in the visible brane. These ultimately result in density and temperature fluctuations in the resulting cosmic microwave background, which, again, have scale invariance. And we can interpret the solutions to the horizon, flatness, and magnetic monopole problems in the context of colliding branes. Both the visible and hidden branes exist long before the collision, and so they can reach thermal equilibrium over a large enough region to explain the smoothness of the CMB. The branes can be very flat over the range that eventually become the observable universe, so there’s flatness achieved. And this type of Big Bang doesn’t start as a singularity—there’s a limit to how hot it gets—and so we don’t need to create magnetic monopoles here either.
So, did we just manage to save the universe from ever ending, or save it from ever starting for that matter? And at the same time, did we save ourselves from having to be part of an eternally inflating multiverse? Let’s not get ahead of ourselves. For one thing, if this ekypyrotic behavior is due to our universe colliding with others within a higher dimensional space, that’s hardly less extravagant than the eternal inflation. This M-theory stuff may not be the cause of the peculiar potential, but even so, we’re going to want ways to test this against the also-untested inflationary model.
Although the ekpyrotic model predicts almost exactly the same observables as inflation, there are potential differences. There may be slight differences in the spectrum of density fluctuations, but more concretely we would expect differences in the gravitational waves produced in the inflationary versus ekpyrotic Big Bangs. In either case, the extremely energetic early universe would have generated gigantic gravitational waves, weighted towards lower frequency in the ekpyrotic case compared to inflation. No currently planned detector will be able to see these waves, but it’s conceivable that one day we’ll build a detector that can sense they now extremely faint buzz of ancient spacetime ripples, and read from the m the nature of the beginning of this universe. Those waves may also have left a signature on the matter that formed soon after, and we may one day be readable in the polarisation of the CMB.
Honestly, it’s crazy to even imagine that we may one day be able to test ideas like this, and actually have a good idea, one way or another, whether there’s an infinite multiverse extending through inflating space, or if we’re just one universe in an endless temporal chain of expanding and contracting spacetime.
