Well, yesterday was quite a day, and I’m still sifting through the consequences.
First things first. As with all major claims of discovery, considerable caution is advised until the BICEP2 measurement has been verified by some other experiment. Moreover, even if the measurement is correct, one should not assume that the interpretation in terms of gravitational waves and inflation is correct; this requires more study and further confirmation.
The media is assuming BICEP2’s measurement is correct, and that the interpretation in terms of inflation is correct, but leading scientists are not so quick to rush to judgment, and are thinking things through carefully. Scientists are cautious not just because they’re trained to be thoughtful and careful but also because they’ve seen many claims of discovery withdrawn or discredited; discoveries are made when humans go where no one has previously gone, with technology that no one has previously used — and surprises, mistakes, and misinterpretations happen often.
But in this post, I’m going to assume assume assume that BICEP2’s results are correct, or essentially correct, and are being correctly interpreted. Let’s assume that [here's a primer on yesterday's result that defines these terms]
- they really have detected “B-mode polarization” in the “CMB” [Cosmic Microwave Background, the photons (particles of light) that are the ancient, cool glow leftover from the Hot Big Bang]
- that this B-mode polarization really is a sign of gravitational waves generated during a brief but dramatic period of cosmic inflation that immediately preceded the Hot Big Bang,
Then — IF BICEP2’s results were basically right and were being correctly interpreted concerning inflation — what would be the implications?
Well… Wow… They’d really be quite amazing.
Would this story be bigger than the discovery of the Higgs particle? Certainly at least as big. To try to compare the two gets us into silly discussions; you can’t know the long-term implications of a discovery at or near the time it is made. But the immediate list of implications would certainly be longer.
Before I start: we’ll need some definitions, given in this article, of:
- A unit of energy called GeV (roughly the mass-energy of a hydrogen atom)
- The Planck energy, Planck mass and Planck length (processes involving the Planck energy or mass in a region the size of the Planck length will generally involve both quantum mechanics and gravity ["quantum gravity"])
- The energy scale associated with dark energy (which is not quite the amount of dark energy — for one thing, despite the name, dark `energy’ is actually an energy per unit volume.)
If any of these is unfamiliar to you, you may want to read that article first and to have it handy, in another tab or window, to serve as a glossary. You may also want to have my History of the Universe or my BICEP2 Discovery Primer hand.
Implications of BICEP2’s Discovery — IF … IF …
IF IF IF BICEP2’s measurement is correct, at least roughly, and IF IF IF it is being correctly interpreted, at least roughly, then there’s a long list of broad implications that I can think of, or that others of my colleagues have pointed out to me in conversations. Some of them are vague — areas where the implications are likely to be important but are not yet very clear. Some of these things were already somewhat implied by the previous experimental successes of the theory of inflation, while some stem directly from BICEP2.
IF … IF …
A puzzle that bothered scientists for decades, as to how the observable patch of the universe (i.e. the part that we can actually observe today; the universe may be much, much larger than this — see here) could be so uniform, would indeed be firmly solved, by a period of cosmic inflation. The extremely flat geometry of the universe would also now be firmly explained.
We would also have confirmation about how the universe became hot — about how the Hot Big Bang got started. The picture would be this: a large amount of dark energy first makes the universe big, via inflation, and then the dark energy turns into energetic particles, making the universe hot (and still expanding, albeit more and more slowly [until relatively recently]). Some people like to say that inflation puts the “Bang” in “Big Bang”, but remember that it also makes the universe flat and uniform and huge (typically much larger than the observable patch) before it heats it up.
The existence of cosmic inflation would itself be another feather in the cap of Einstein’s theory of gravity — since it is Einstein’s theory that predicts that the presence of a positive cosmological “constant” [not necessarily constant], also known as “dark `energy’ ” [not really energy, but energy density and negative pressure in just the right combination] actually causes the universe to expand, rather than (as we’d naively expect from gravity) contract.
And (IF… IF…) the confirmation of cosmic inflation would mean that those who pointed it out and its advantages, and developed the idea — people like Starobinsky, Guth, Steinhardt, Linde — ought to be able to celebrate (though not all will do so, because they abandoned the idea…)
Also celebrating would be those who pointed out the possibility of a detectable signal from gravitational waves in the polarization of the CMB, which is what BICEP2 has apparently observed. I believe these would be Kamionkowsi, Kosowsky, Stebbins, Seljak and Zaldarriaga (but the cosmologists should correct me if I’m unfair here.)
Were there any doubt left concerning Einstein’s prediction that gravitational waves exist — perhaps some leftover worry about the 1993-Nobel-Prize-winning indirect detection of energy carried being off by these waves, via precision measurements of a pulsar, one of a pair of closely-orbiting neutron stars discovered by Hulse and Taylor — it would be over.
And if anyone still wasn’t sure that gravity is due to a spin-two field, not a spin-zero field, the BICEP2 result would settle the matter; you can’t get B-mode polarization of the CMB without combining the spin-two polarization structure of the gravitational waves with Thomson scattering.
If there were any doubt that gravity is controlled, just like everything else, by quantum physics, it would erased; BICEP2’s observation would imply that just like other fields, which are subject to quantum jitter (i.e., random “fluctuations“), space and time (somewhat more precisely, the metric that determines distances) undergoes the same type of quantum fluctuations as other fields, fluctuations that any quantum version of Einstein’s theory of gravity would predict. No details about quantum gravity are needed for this conclusion.
The amount of dark energy required during inflation, in order that BICEP2 could observe this gravitational wave signal, would (IF… IF…) be enormous. The energy scale of dark energy [defined here] would be about 1016 GeV, just about 100 times less than the absolute maximum possible, which is the (reduced) Planck scale. This need not have been the case! This energy scale could have been trillions of times smaller, and yet still given acceptable amounts of inflation of the universe and an acceptable Hot Big BAng. Amazingly, the dark energy during inflation, if BICEP2 is right, is much larger than it needed to be, and almost as large as it could possibly be!
Moreover, if the energy scale of dark energy [defined here] had been just a few times weaker than BICEP2 observes — 1000 times less than the absolute maximum — then the gravitational wave signal would have been so faint that it would have been completely drowned out by another process (lensing of E-mode polarization, the solid red curve on the first figure here.) So most of us thought, with this wide range of possibilities, that it was a real long shot that BICEP2, or any future similar experiment, would ever see any signal due to gravitational waves from inflation. That the gravitational waves would turn out to be large and powerful enough to be observed would be an amazing piece of luck for scientists wanting to understand the universe. (IF… IF…)
[The original version of this paragraph overstated what we'd know; thanks to my colleagues for pointing out a blunder concerning the heating after inflation.] Also, this would mean we would now have an estimate for how hot the observable patch of the universe could have become after inflation ended and the region containing the observable patch became very hot — the start of the Hot Big Bang. The energy scale of the dark energy during inflation determines the maximum possible temperature at the start of the Hot Big Bang, more or less. To say it another way: after inflation with dark energy of scale 1016 GeV ended, the universe would have become hot, potentially so hot that the average particle rushing around in the hot dense soup of particles had a motion-energy of 1016 GeV, though perhaps quite a bit less than this. That’s a maximum temperature of as much as 1029 degrees [yes, that's a 1 with 29 zeroes after it!!!]
The Large Hadron Collider’s data provide direct insights into physics at the energy scale of around 1000 GeV or so. [The mass-energy mc² of the Higgs particle is 125 GeV.] The BICEP2 measurement would arguably be our first direct evidence (IF… IF…) concerning physics at higher energy scales than the LHC (though one could argue we have a little information from the existence of neutrino masses.) And not just a little higher! Since the scale of the dark energy at inflation is at 1016 GeV, we’re talking 10,000,000,000,000 times higher!!! We’ve been trying for years, using various methods, to find evidence concerning how physics works at or near the Planck energy and Planck length. All previous efforts have come up with nothing; proton decay might have given us insight, but it is too rare if it happens at all; neutrino oscillations haven’t given a clear pictures; cosmology might have revealed big surprises concerning the Planck energy, but none have previously shown up. But if BICEP2 is right, then, for the first time, we are seeing phenomena which occurred near this energy scale!
Similarly, data from cosmology has made us very confident that we understand physics in the early universe back to the time when the first atomic nuclei formed during the Hot Big Bang — a few minutes after the Hot Big Bang started, when the temperature was such that the typical particles had energy of about 0.001 GeV. And we have had reasonable confidence that we have a decent understanding, using what we’ve learned about particle physics recently, back to times of a billionth of a second, and temperatures corresponding to an energy of a hundred GeV or so. But now our understanding may be taking (IF… IF…) an enormous leap, back to the very start of the Hot Big Bang, at a temperature corrresponding to an energy of as much as 1016 GeV, and even earlier, into the frigid inflating universe.
The success of the details of the equations that form the heart of inflationary theory suggests that not only did Einstein’s theory of gravity describe physics at very early times and very high energy scales, the basic principles of quantum field theory worked back then too. It could easily have been imagined — and many have imagined, with concrete ideas backed up with equations — that there might be important principles that we are unaware of, or modifications of the ones we know, that would give very different predictions from standard inflationary theory. So far, there’s no sign of that at all. BICEP2’s result (IF… IF…) would be yet another sign that despite having done all our particle physics and gravity measurements at much lower energy scales and longer distances, those accessible to the LHC and to our previous experiments, we’ve actually understood the principles that govern the behavior of phenomena at much, much higher energy scales and much shorter distances!
The energy scale inferred from BICEP2’s measurement, 100 times smaller than the Planck energy — 1016 GeV or so — has appeared in particle physics before! If you take the three non-gravitational forces of nature — the strong nuclear force, the weak nuclear force and the electromagnetic force — and you first consider how they are rearranged when the Higgs field is turned off, into strong nuclear, weak isospin, and hypercharge forces — and then you look at how the strengths of the forces drift as you study them at shorter and shorter distances and higher and higher energies, you find the forces all become about the same strength at an energy scale of about 1016 GeV or so. This is called “unification of coupling constants” (i.e. force strengths) — or, originally and more ambitiously, “grand unification”, a grander notion that the three non-gravitational forces are actually manifestations of just one type of force. [Unification of that unified force with gravity is yet another question; that's what string theory might do, though string theory does not require grand unification be a separate process from the unification with gravity.] Originally, back around 1980, inflation was imagined to be associated with grand unification, and if so, would have an energy scale of about … yes, 1016 GeV or so. But that idea died out long ago as people learned more about both unification and inflation. Yet now we must wonder: could that part of the original idea, in some vague way, have actually been right?
Over the years, scientists have invented a plethora of variations on how inflation might have, in detail, taken place. BICEP2’s observation of a gravitational wave signature would sweep away most of them, leaving just a few. (More will be invented in coming months, though!) It also would sweep away some alternatives to inflation and many speculative ideas for how the universe might behave, or might at least have behaved at earlier times. Despite “tension” (meaning mild disagreement) between BICPS2s current data and the Planck satellite’s data, a rather simple observable patch of universe, with rather simple laws of nature and a rather simple form of inflation, are consistent with what we know.
One of the variants of inflation that would be excluded by BICEP2’s data is the notion (admittedly long-shot anyway) that actually the Higgs field could play the role of the inflaton after all. It turns out that no observably large gravitational wave signal would be expected if that were true.
BICEP2’s result would represent the first time that, without any theoretical speculation about quantum gravity, an experiment has forced us to consider processes involving physics at the Planck scale, where quantum gravity is important. Specifically, for a variant of inflation to give such a large gravitational wave signal relative to the size of the other non-uniformities in the cosmic microwave background, the Planck energy scale becomes important. The inflaton field, which (by definition) is the field whose stored “potential energy” is the dark energy, must change by an amount close to or a bit larger than the Planck energy scale. [Sorry for the necessary technical-speak here. This whole business is inherently confusing for non-experts. The dark energy may change or may not change at all while the inflaton field is changing; that's a separate question. What it means for a field like an inflaton to change by the Planck energy scale is that if a particle interacted with the inflaton field as strongly as it possibly could (the way the electron interacts with the Higgs field, but stronger), then as the inflaton field changed by an amount comparable to the Planck energy scale, the particle's mass would change by an amount comparable to the Planck mass.]
An extremely simple possibility for inflation that would still be consistent with all the data (IF… IF…) is just to have a spin-zero field (a bit like the Higgs field, but with important differences — no weak nuclear force effects, for instance) which has a mass and no substantial interactions with any other fields. This is a model introduced by Andrei Linde; it’s amazingly simple. Could it really be correct?
Among the other very simple ideas that are likely to feature prominently in discussions of the near future are ones involving a “pseudo-Nambu-Goldstone boson”, of a type often called an “axion”. [I wrote a little about this in the paragraph just below the figure within this post about the Planck satellite and what it means for variants of inflation.] One of the variants of inflation that is most consistent with both the Planck satellite and BICEP2 (IF… IF…) is an idea called “natural inflation”, from 1990, due to Katherine Freese, Joshua Frieman and Angela Olinto. [Here's a pdf of the paper.] Such a field has the feature that it is periodic (i.e., if you change it by enough, it comes back to itself, the way an angle that can only range from 0 to 2π.) It also has the feature that its interactions with all other fields are rather weak. When this field varies a lot, the dark energy associated with it varies by quite a bit less… so the field could vary by something approaching the Planck scale yet the energy scale of the dark energy could stay rather constant, at one percent of the Planck energy.
In order for this idea to really be consistent with quantum gravity, it might require some modification, of the sort was introduced in 2008 by Liam McAllister, Eva Silverstein, and Alexander Westphal, in which the axion is periodic with a larger range than you’d naively expect. [They did this work in the context of string theory, but I don't believe that string theory specifically is necessary for the idea to work; it might work in other consistent quantum gravity theories, if there are any.] Expect to hear much more about axions. [By the way, it's long been suggested that dark matter itself could be made from a type of axion... but presumably not the same one...]
One last one which is the weakest of the set… In a recent article, I pointed out that if the Higgs particle discovered at the LHC turns out to be of the simplest possible type — a “Standard Model Higgs” — then because this is “unnatural” (in the sense of “highly non-generic”) it might call into question fundamental conceptual issues, perhaps even the whole framework of quantum field theory. But BICEP2’s measurement would seem to support the framework of quantum field theory in a world of three smooth spatial dimensions up to very high scales and very short distances, so that line of thinking would be disfavored. (IF… IF…) Of course, measurements of the Higgs particle are still in the early days — at this point we can only say the Higgs particle is roughly “Standard-Model-like” in type — and maybe the Higgs will turn out to be more complex, or other new particles will show up at the LHC, possibly rendering the discussion moot. But (for reasons I outlined in my Santa Barbara talk two weeks ago) having only the Standard Model, with the Higgs field turned on just a little, plus quantum gravity and an inflaton field (and plus something to explain dark matter and neutrino masses) would pose many grave conceptual problems, ones that anthropic reasoning would not address. [Experts: Anthropic reasoning can be used to argue why the cosmological constant might be small; it can be used to argue why there is a hierarchy between the Planck mass and the proton and electron masses; but it cannot easily be used, without huge and problematic assumptions, to argue why one would find just the Standard Model at the LHC, with one light scalar Higgs field and nothing else new.] So I would tend to see the BICEP2 result (IF… IF…) as a (rather weak!!) argument slightly in favor of supersymmetry and grand unification, with supersymmetry just a little out of reach of the LHC for some reason, or modified in some way to make it harder to observe than expected. [Experts: "mini-split" supersymmetry, which preserves quantum field theory, with unification of coupling constants at the grand unified energy scale, starts to look better than something more radical.]
Colleagues: what have I left out? Eventually this post will be stored on this site as a reference article, so I’d like to make it complete.
One more time I must remind you that we’re still some time away from trusting the BICEP2 result, and quite a long time from trusting the interpretation of the result in terms of inflation. All of the implications I’ve mentioned are therefore provisional. But the list is impressive, and there are probably more I have forgotten to mention, or are known to others but not to me, along with still others that haven’t yet been noticed by anyone. So I don’t yet know if BICEP2’s measurement is right, or if inflation occurred and if it created the signal they observe, but I do know this: there’s a lot at stake.