Since the CMB is the oldest thing we can see, how do we know for sure what happened before the CMB?

Question

I keep wondering about how can we be so sure about what happened before the CMB, given that it is the oldest thing we can actually see. It seems like we are very confident about what really happened from $10^{-12}\ \mathrm s$ after the Big Bang until $380\,000$ years after the Big Bang, but how?

How do we know the universe was expanding before the CMB? We are sure it is expanding since after the CMB until now, but are we really sure it was expanding before the CMB, and if yes, how can we be so sure?

How do we know the temperature of the universe was approximately $10^{12}\ \mathrm K$ at $10^{-12}\ \mathrm s$ after the Big Bang?

I have read that we were able to recreate temperatures as high as $10^{12}\ \mathrm K$ in the Large Hadron Collider and to observe it directly. OK, but how would that be an evidence about what happened $10^{-12}\ \mathrm s$ after the Big Bang?

Answer 1

You are quite correct that we can't see what happened before the CMB (this time is known as recombination) but this is not unusual in Physics. For example we can't see what happens at collisions in the Large Hadron Collider. All we can see is the debris that comes flying out of the collisions. But we understand the physics involved so by measuring the properties of the debris we can calculate what happened in the collision. That's how the Higgs boson was discovered. It wasn't directly observed but its existence was shown by precise measurements of the particles that we can detect.

And the same applies to the universe. The CMB is the debris that came flying out of the Big Bang, so by measuring the properties of the CMB we can calculate what happened at times before recombination.

The obvious question is how we know our calculations are correct. The way we approach this is to try calculating the same thing in different ways. For example Higgs bosons can be detected in several different ways, and if those different measurements gave different masses for the Higgs boson we'd know at least some of our calculations must be wrong. This is harder for the universe since we only have the one universe and the creation of the universe isn't an experiment we can repeat. But we can still cross check various different calculations and at least make sure they are consistent, which is exactly what is done.

Recombination happened about 370000 years after the Big Bang, and in fact the physical properties of the universe at this time are easy to understand. The density and temperature are in the range that we can recreate in the lab so we can directly probe the properties of plasma under these conditions. Indeed even as far back as nucleosynthesis, which happened only a few minutes after the Big Bang we still understand the physics well from experiment.

For example you mention a time $10^{-12}$ seconds after the Big Bang, and this time is normally taken to be the end of the electroweak epoch. From this time on the interactions between particles in the universe occur at energies that can be probed in colliders so we can experimentally determine what would be happening from this time onwards. Incidentally the temperature at this time was more like $10^{15}$ K than $10^{12}$ K.

But it is certainly true that as we go back towards the Big Bang there comes a point where the density and temperature exceed anything we can study experimentally, and we can be less sure what happened then. This is still an active area of research.

Answer 2

It isn't the oldest thing we can see.

Most of the hydrogen, helium and deuterium nuclei that are around in the universe now, were created in the time period between a few seconds and about 15 minutes after the big bang.

The abundances of these nuclei in the universe is a direct probe of the physical conditions, and time evolution of those conditions, at the epoch of primordial nucleosynthesis.

The only important free parameter in the standard big bang model, as far as these abundances are concerned, is the ratio of baryons to photons, which in turn can be found from fluctuations in the cosmic microwave background formed hundreds of thousands of years later.

There is complete concordance in the estimated primordial abundances of He and D. Spectacularly precisely in the case of D, where the primordial abundance is quite sensitive to the nucleosynthetic conditions and the primordial abundance can be estimated precisely.

However, aside from this, there is the fact that we don't need to see something to know what has occurred. The cosmic microwave background and the fluctuations within it are the consequence of events that happened at earlier times. Unless one wants to abandon physical reasoning, then there is no difficulty in accepting that the cosmic microwave background, and it's evolving temperature with time (which has been measured) is very strong evidence that the universe was much denser and hotter in the past, with all the physical consequences that would imply.

Of course you can push that too far. There are details of the physics itself that are poorly understood before about $10^{-12}$s, although things like the baryon to photon ratio, measured in the CMB, encode the mysterious matter/anti-matter asymmetry and allow it to be probed; even though it cannot be "seen".

But post $10^{-12}$s the physics is reasonably understood, so if we have a good idea of what the conditions are in the period of a few seconds after the big bang (from primordial nucleosynthesis) and a few hundred thousand years after the big bang (from the CMB), then we can reasonably extrapolate back to $10^{-12}$s.

In the same way that if you follow the final part of the path of a launched projectile, it is perfectly reasonable to measure that trajectory and follow it back to identify where the launch site was.

Answer 3

Interesting question! Let me see if I can shed some light with an analogy. Btw, I shall be referring to John's answer at points.

Studying or researching in astronomy is very much like criminal investigation. You have the crime, you search for clues which are used to reconstruct the events of the crime. Here, we have a crime, the construction of the universe, with clues spread here and there. One of the very well known clue is the CMB, and the oldest clue we can find (or have found, more on this later). As John tells us, we can recreate this clue and see what follows. A natural deduction. But can we work backwards from this clue, leading to the crime?

Working backwards, we need a method that creates the CMB. We have a theory, that at one point matter was coupled to light, the universe was opaque and exactly 380,000 years later, they decoupled, and light was finally free to travel the universe, and this is what we see as the CMB. Is this correct? Prossibly, and Sherlock shall tell you, that by the balance of probability, it's more probable than possible because the black body spectrum and anisotropies of the CMB predicted by this theory very precisely match that is observed. We have taken a very important step!

Now, how come they decoupled? Because at this time "recombination" occurred, electrons and protons formed hydrogen, which was transparent to light. Voila!

But before this? Nucleosynthesis, we know this since once again, balance of probability, and experiments. Before this? Formation of protons and neutrons and other composite subatomic particles. And finally before this, we expect the CvB (Cosmic neutrino Background), when the neutrinos decoupled from matter. We are trying to observe CvB to see if we are on the right track.

Going further back...

Here we have, as John again mentions, the end of the electroweak epoch. Since he mentions this, let us go and see if we can't trace it further back. Here we run into trouble. There are several ways to reach this stage. Which is correct?

Here's an analogy that tells us what we can do right now. Watson notes about Sherlock says "I nearly fell into the error of supposing that you were typewriting. Of course, it is obvious that it is music. You observe the spatulate finger-end, Watson, which is common to both professions? There is a spirituality about the face however... which the typewriter does not generate. This lady is a musician."

Our clue is that we have to reach the end of the electroweak epoch, this is our "spatulate finger-end". And hence we create theories, our "typewriters" and "musicians", all of which predict this. But we are missing our "spirituality about the face", and this is what we are trying to find, and what John mentions as "an active area of research".

There's one thing that pops to mind. There's a theory that states that the universe began as the collision of two "branes", as opposed to a sudden inflation out of a pinpoint of energy (or uncertainty, which is again debatable). Researchers say they can settle this by observing the gravitational waves formed during that event. If they are mild enough, then the former theory gets credibility, if not, then the latter.

This is how astronomy works. Criminal investigation.