00:00:00
Before we get started, just wanted to let you
00:00:02
know we have some new merch at the merch
store celebrating 10 years of Space Time.
00:00:07
How did the universe begin? How can
something come from nothing? One way to
00:00:12
“solve” this most difficult of philosophical
conundrums is to avoid it altogether. Maybe
00:00:18
the universe didn’t begin at all. Maybe the
Big Bang was just one in an endless cycle.
00:00:30
On February 11th, 2015 a new show called PBS Space
Time appeared on YouTube. In the 10 years since,
00:00:39
together we've explored the insides of black holes
and ventured across the edge of the universe and
00:00:44
seen the beginning and end of time and the peeked
at the underlying clockwork of nature. It's been
00:00:51
brain-breaking and existentially humbling journey.
And we're so happy you have been on this ride with
00:00:58
us. Well, that ride continues. We have exciting
new plans for year 11 that will be revealed in
00:01:07
due course. For now, two things you can do to help
us get the anniversary celebration started. First,
00:01:16
post in the comments to say how long you've been
watching ... and if you joined more recently how
00:01:22
much of the back catalog of over 400 videos did
you actually manage to get through. And second,
00:01:30
merch! We’ve got a limited edition 10
year anniversary design as well as some
00:01:36
classic logo merch. Having this gear
doesn't just make you incredibly cool,
00:01:42
you also helps us keep going for another decade.
00:01:46
No one wants the end of the universe. Maybe
that’s why the idea of a cyclic universe is so
00:01:52
appealing. So much so that it appears in mythical
cosmologies of Hinduism, Buddhism, Zoroastrianism,
00:01:59
and the cosmologies of the Norse, Mayans,
Egyptians, Greeks, and no doubt more. It makes
00:02:04
sense. We tend to extrapolate from the patterns
we see in nature. We see recurring cycles of day
00:02:09
and night, of the seasons, of life and death. Why
shouldn’t the entire universe go through cycles?
00:02:16
Of course, the evidence has to support
our extrapolations, and we now know
00:02:20
that the evidence does not support cyclic
cosmologies. Right? The universe and time
00:02:25
itself started at the Big Bang and space
will expand forever and that’s it. Or so
00:02:30
says mainstream cosmology. But apparently
cyclic cosmology has not lost its appeal,
00:02:35
because scientists have found a way
to make it work for our universe.
00:02:39
The original Big Bang model has all of space
originating in an infinitesimal point at the
00:02:44
beginning of time, and expanding from there.
This fits a lot of observations of our universe.
00:02:49
The recession of the galaxies
reveals the expansion of space,
00:02:53
and the cosmic microwave background is
pretty clearly the afterglow of an early
00:02:57
hot, dense state. But some observations
aren’t so easily explained. For example,
00:03:01
in our universe matter and energy are very evenly
spread out, but in the basic Big Bang model there
00:03:06
wasn’t time for this smoothing to happen before
the expansion threw distant regions beyond causal
00:03:11
contact. This is the horizon problem. There’s
also the fact that a basic-Big-Bang universe
00:03:16
isn’t expected to be so perfectly flat, which
requires an uncannily perfect balance between
00:03:22
matter and dark energy. And if the universe
that began in an extremely hot, dense state,
00:03:28
than certain relics of this phase should
persist—so-called magnetic monopoles.
00:03:34
Of course we’ve talked about the horizon,
flatness, and magnetic monopole problems,
00:03:39
as we have the most popular solution—cosmic
inflation. This proposes an extreme,
00:03:45
exponential expansion phase in the extremely
early universe. Inflation becomes the bang in
00:03:50
the big bang, and it allows an initially smooth
universe to be expanded beyond causal contact,
00:03:55
as well as being nicely flattened
space and scattering those pesky
00:04:00
monopoles far enough apart that
they're unlikely to ever be seen.
00:04:04
To top it all off, inflation explains how the
universe got its large-scale structure. It
00:04:09
predicts quantum fluctuations in the inflaton
field, which became the gravitational seeds
00:04:16
that grew into galaxies and galaxy clusters.
Inflation goes further, predicting the those
00:04:21
fluctuations should lead to the same level of
lumpiness at all size scales—so-called scale
00:04:27
invariance. And that’s exactly what we see in
the lumps of the cosmic microwave background.
00:04:33
Inflation does such a good job
that it’s practically mainstream,
00:04:36
but there are some downsides. Modern versions
of the idea predict that if inflation happened,
00:04:41
then it never ended. Sure our little
patch quit with that extreme growth,
00:04:49
spawning our much-slower expanding universe. But
as long as inflation didn’t stop everywhere all
00:04:54
at once, then out there, somewhere, this eternally
inflating greater universe is blowing up forever,
00:05:01
constantly spawning bubble universes.
Some find this idea a little extravagant.
00:05:07
The other issue with inflation is that
it doesn’t avoid a beginning of time,
00:05:11
nor a point of infinite density—a singularity at
that beginning. You can push that singulary as far
00:05:18
back as you like, but in inflationary models
it has to be there, just like in the regular
00:05:23
Big Bang. And any theory that predicts
a singularity is automatically suspect.
00:05:28
And even if inflation didn’t have its issues, it’s
worth exploring other options. So what about the
00:05:33
option where, instead of asking what happened at
the beginning of time, we ask what happens if time
00:05:39
never had a beginning? Cyclic cosmologies exist
in many ancient traditions, but also in modern
00:05:45
cosmology. Soon after we noticed that the universe
was expanding, scientists came up with models in
00:05:51
which the universe eventually slows and starts
contracting, then bounces into a new Big Bang,
00:05:57
and repeats over and over. But these models
didn’t manage to do away with the beginning
00:06:02
of time because both entropy and the lifespan
of the universe had to increase with each
00:06:06
bounce. That means we couldn’t extrapolate the
bouncing back in time indefinitely. There had
00:06:12
to be a first. And anyway, these cyclic universes
didn’t solve the problems that inflation solves.
00:06:19
But it turns out that cyclic cosmologies can give
us everything we want. To explain the horizon,
00:06:25
flatness and monopole problems without
inflation, and at the same time eliminate
00:06:30
that pesky beginning of the universe. In fact,
it turns out that the same type of field that
00:06:36
causes inflation can also be tweaked to give an
infinitely regenerating universe. This is the
00:06:43
idea of the ekpyrotic universe—named after the
cyclic cosmology of the ancient Greek Stoics,
00:06:50
in which the universe is rebirthed in
fire—ekpyrosis—between unending cycles. By
00:06:58
the way, this is a very different idea to the
conformal cyclic cosmology of Roger Penrose,
00:07:03
and I refer you to our previous episode
for that equally awesome proposal.
00:07:08
The ekpyrotic universe was first proposed in 2001
in a paper by Justin Khoury and collaborators.
00:07:14
Now, I’m going to come back to the ideas
of this paper, but first I want to give the
00:07:19
part of the story told in a followup paper
by two of the original authors. In 2002,
00:07:25
Paul Steinhardt and Neil Turok showed how the
same type of quantum field proposed to cause
00:07:31
inflation—the inflaton field—can be tweaked to
resurrect the universe rather than blowing it up.
00:07:38
The extreme accelerating expansion of inflation
is driven by the same type of quantum field as
00:07:43
we think now drives the relatively
chill acceleration that we attribute
00:07:47
to dark energy. It’s a scalar field—the
simplest type of quantum field in that
00:07:52
it’s just a simple numerical property—a field
strength—in space everywhere. The field also
00:07:59
has a potential energy associated with that
numerical value. Sometimes the relationship
00:08:04
between field value and its potential energy is
simple—stronger the field the more the energy,
00:08:09
sometimes it's complicated. For example, in some
versions of inflation, the field value slowly
00:08:14
drops and energy decreases, but then the energy
reaches a minimum value and any further change
00:08:20
in the field would add more energy. Therefore
the field becomes stable and inflation stops.
00:08:26
But there are lots of different ways you can
relate the field energy to the field value. And,
00:08:31
as shown by Steinhardt and Turok one of
those ways gives you a cyclic universe.
00:08:37
We normally think of dark energy as being due
to a constant energy density everywhere in space
00:08:43
that does not change over time. But maybe dark
energy changes only very slowly. For example,
00:08:51
if there’s this quantum field that is slowly
changing in value—say, decreasing—with barely
00:08:58
noticeable changes in the associated energy. Then
we have accelerating expansion far into the future
00:09:04
even as the field value drops. But eventually the
energy in that field fades and becomes negative,
00:09:11
and as that happens acceleration slow, then
expansion slows and the universe briefly halts.
00:09:20
And then the universe recollapses. The field
potential bottoms out in a minimum and rises
00:09:28
back to zero. Now, we might expect the
field to get stuck in that minimum,
00:09:33
however during this contraction,
gravitational potential energy is
00:09:37
converted into kinetic energy of the field
so that the field value blows past this
00:09:42
minimum. In the final phase of contraction
the field has no potential energy—no “dark
00:09:49
energy” equivalent—and the kinetic energy
of the field gets converted into radiation.
00:09:56
This liberates the universe from
the constraints of this quantum
00:09:59
field and it quickly starts expanding again.
The radiation spawns matter and then dissipates,
00:10:07
the matter dominates for a while,
and finally dark energy takes over
00:10:11
and we find ourselves back where we
started in the roughly the modern era.
00:10:16
So how does this version of a cyclic universe
solve all our problems? Well, the magic happens
00:10:22
in the contraction phase—what the authors call
the ekpyrotic period. When the universe was
00:10:29
at its largest, matter was so far-flung that the
only meaningful energy in the universe was in its
00:10:35
scalar field, and the only meaningful structure in
the universe were the quantum fluctuations in that
00:10:41
field. Now as the universe collapses very slowly,
those fluctuations are amplified. The shape of the
00:10:49
scalar field is tuned so that these fluctuations
have a scale-invariance—equal frequency for all
00:10:56
sizes of lumps, just as is observed in
the CMB and as is predicted by inflation.
00:11:03
That contraction phase also smooths out the
universe, solving the horizon problem. And
00:11:10
the universe does not reach arbitrarily high
temperatures, so no magnetic monopoles ever
00:11:15
need to be created. There’s no singularity, and
there’s no significant difference between one
00:11:21
cycle and the next, so this model is consistent
with cycles extending back in time forever.
00:11:27
Steinhardt and Turok claim that the
scalar field needed to achieve all of
00:11:31
this magic is no more finely tuned than
the field needed to achieve inflation,
00:11:36
and so this cyclic model is just as plausible as
the inflationary model because they both result in
00:11:42
the same observables—at least as far as current
observational sensitivity allows. There are ..
00:11:48
But is there any more motivation to believe that a
field of the needed variety actually exists? Well,
00:11:55
these guys say yes, and the mechanism
was proposed the year prior by a team
00:12:00
including these authors and led by Justin Khoury.
00:12:03
This does require a little more than a
modification of the inflaton field. It
00:12:07
requires an entire new dimension of space. 6 new
dimensions really. One motivation for the type of
00:12:14
scalar field needed, with its particular potential
energy curve, lies within M-theory. This is an
00:12:22
encompassing framework for string theory, in which
our universe, with its 3 dimensions of space and 1
00:12:27
of time, exists as a single slice in a greater
object with 4 large spatial dimensions. And
00:12:33
with 6 coiled, compact dimensions, but we don’t
need to worry about those for this description.
00:12:39
Our universe would be something called
a brane—short for membrane—living within
00:12:45
the higher dimensional space—itself
called the bulk. In this picture,
00:12:49
our universe is one of the boundary layers of the
bulk. We call it the visible brane. Normally this
00:12:56
brane is just chillin—it’s pretty empty and is
not changing in size. Things get interesting
00:13:02
when another free-floating brane within the
bulk—called hidden brane—smashes into us. Which,
00:13:09
apparently, is something that can happen.
Try not to let it keep you up at night. This
00:13:14
collision dumps a bunch of energy into
the visible brane, sparking a big bang.
00:13:20
This connects to the description
of the scalar field because we can
00:13:24
interpret the value of that field as
the distance between the visible brane
00:13:29
and this incoming hidden brane. So on
the graph we saw earlier, movement to
00:13:36
the left—decreasing field value—corresponds
to decreasing distance between the branes.
00:13:41
Khoury paper assumes a simpler form of the
potential than Steinhardt and Turok, with energy
00:13:48
decreasing exponentially as the branes approach,
which is like a purely attractive force between
00:13:53
the branes. In the more complex version where
potential energy decreases then increases again,
00:13:59
we have attraction then repulsion of the
branes. Either way, when the branes collide,
00:14:05
the energy is dumped into the visible brane
causing space there to start expanding. The
00:14:12
hidden brane recoils, propelled back
the way it came with the energy of the
00:14:17
bounce until eventually it’s pulled inwards again.
00:14:21
In this interpretation, the quantum fluctuations
of matter manifest as wiggles in the incoming
00:14:27
hidden brane. These result in different parts
of that brane arriving at different times,
00:14:33
and so there are variations in the start time of
expansion in the visible brane. These ultimately
00:14:41
result in density and temperature fluctuations
in the resulting cosmic microwave background,
00:14:47
which, again, have scale invariance. And we
can interpret the solutions to the horizon,
00:14:54
flatness, and magnetic monopole problems in the
context of colliding branes. Both the visible
00:14:59
and hidden branes exist long before the
collision, and so they can reach thermal
00:15:03
equilibrium over a large enough region
to explain the smoothness of the CMB. The
00:15:09
branes can be very flat over the range that
eventually become the observable universe,
00:15:15
so there’s flatness achieved. And this type of
Big Bang doesn’t start as a singularity—there’s
00:15:22
a limit to how hot it gets—and so we don’t
need to create magnetic monopoles here either.
00:15:27
So, did we just manage to save
the universe from ever ending,
00:15:31
or save it from ever starting for
that matter? And at the same time,
00:15:36
did we save ourselves from having to be part
of an eternally inflating multiverse? Let’s
00:15:42
not get ahead of ourselves. For one thing,
if this ekypyrotic behavior is due to our
00:15:47
universe colliding with others within a higher
dimensional space, that’s hardly less extravagant
00:15:52
than the eternal inflation. This M-theory stuff
may not be the cause of the peculiar potential,
00:16:01
but even so, we’re going to want ways to test
this against the also-untested inflationary model.
00:16:09
Although the ekpyrotic model predicts almost
exactly the same observables as inflation,
00:16:14
there are potential differences. There may be
slight differences in the spectrum of density
00:16:18
fluctuations, but more concretely we would
expect differences in the gravitational waves
00:16:23
produced in the inflationary versus
ekpyrotic Big Bangs. In either case,
00:16:29
the extremely energetic early universe would
have generated gigantic gravitational waves,
00:16:34
weighted towards lower frequency in the ekpyrotic
case compared to inflation. No currently planned
00:16:40
detector will be able to see these waves, but it’s
conceivable that one day we’ll build a detector
00:16:46
that can sense they now extremely faint buzz of
ancient spacetime ripples, and read from the m
00:16:53
the nature of the beginning of this universe.
Those waves may also have left a signature on
00:16:59
the matter that formed soon after, and we may one
day be readable in the polarisation of the CMB.
00:17:07
Honestly, it’s crazy to even imagine that we
may one day be able to test ideas like this,
00:17:14
and actually have a good idea, one way or another,
00:17:18
whether there’s an infinite multiverse
extending through inflating space, or if
00:17:23
we’re just one universe in an endless temporal
chain of expanding and contracting spacetime.
