This conference has been fairly insane, with very little downtime and spotty internet access. I’m still chewing on some of the talks I saw and the conversations I had; in the meantime, here’s the second half of the post from last week…
Figuring out what happened during the first billionth of a billionth of a second after the Big Bang is rather hard, mostly because there’s no light we can see from that time. In fact, there’s no light that we can see from the first 380,000 years after the Big Bang. For most of that time, light was trapped in a plasma, a dense soup of electrically charged particles a lot like the interior of the Sun. Light can’t travel in a straight line in a plasma — it keeps running into electrically charged particles, which deflect the light and exchange energy with it. But as the universe expanded, it cooled down, and once it got cool enough, those charged particles — mostly electrons and atomic nuclei — got together and formed atoms, which don’t deflect light nearly as much because they have no electric charge. At that point, 380,000 years after the Big Bang, light was able to propagate freely for the first time in the history of the universe; the light from that time — the oldest light in the universe — is now the cosmic microwave background radiation (CMB). The CMB is an immensely rich source of information about the very early universe, letting us look back at the universe when it was about 0.003% of the age it is now. But that’s still much, much later than the era we’re interested in. The CMB lets us see the tree when it was a sapling — but we want to know what the pinecone looked like.
Can we do better? Well, there’s no light older than the CMB, but there are plenty of things other than light. Most of the atomic nuclei in the universe –hydrogen and helium, with a little lithium thrown in — were forged in the first 20 minutes after the Big Bang, when the entire universe was as hot as the center of the Sun.1 By looking at the relative abundances of those elements in the universe — how much helium there is vs. how much hydrogen there is, for example — we can learn a lot about what the universe was like during that first half-hour. But that’s still not going to get us what we want: in the first three minutes after the Big Bang, the universe was too hot for atomic nuclei to remain stable, so all those nuclei we see today didn’t start forming until then. (Think about that: too hot for nuclei to remain stable. That’s much, much hotter than anything anywhere in nature today2 — billions of degrees — and hotter than anything humans have created other than the brief, tiny flashes of intense heat generated in particle accelerators.)
So, we’re trying to learn about a period in the universe’s history so far back that there’s no light left over, and there’s no atoms or atomic nuclei left over either. So how can we learn about that first fraction of a second? The key is that the tiny differences in density I talked about last time come from that first fraction of a second. In the violent and hot early history of the universe, those differences in density were stretched and warped a bit, but they should have persisted. And again, as I mentioned last time, those early differences in density — we call them “primordial density perturbations” — seeded all the structure that we see in the universe today. Along the way, they also affected the cosmic microwave background radiation. So by looking at the structure of the universe today, as well as the cosmic microwave background radiation, we can learn about the primordial density perturbations that came into being a fraction of a second after the Big Bang; and through those perturbations, we can learn something about what the very, very, very early universe was like.
What, specifically, are we trying to learn about those primordial density perturbations? As I’ve mentioned before, a lot of cosmology is statistical, and this is no exception: the statistical properties of these density perturbations carry interesting information about the very early universe. Specifically, what I go after (as do many other people) is the distribution of those perturbations. For example, say we plotted out the magnitude of these density perturbations — particularly over-dense regions go on the right, under-dense regions go on the left, and regions of average density go in the middle. It’s just like plotting the amount of money that you spend in your house on food each week over ten years — most weeks, you might spend something pretty close to the average amount, and then some weeks you spend a lot and others you don’t spend as much, and you end up with something that might look like this:
while for the primordial density perturbations, we might find something that looks like this:
These plots aren’t plotting the same thing, but they’ve got the same overall shape, one which shows up an awful lot in nature, and which goes by several names: bell curve, normal distribution, or Gaussian distribution (after Carl Frederich Gauss). This kind of curve has all kinds of special properties, and shows up all over the place in nature — you’ve probably seen it before, or heard of it, and I’m not going to go into all the special things about it here. But it’s called a normal distribution for a reason: all other things being equal, it’s the distribution that you’d expect to see for nearly any given thing that you plot. (This is because of something called the Central Limit Theorem, and I’m really really not going to talk about that here, at least not today.) It’s sort of the default answer in that way, and a plot like the second one I’ve got there wouldn’t be a surprise to anyone if it turned out to be an accurate description of the primordial density perturbations.
But, of course, science is about surprises. And if that second plot turned out to be wrong — if the primordial density perturbations turned out to be distributed in a different way, some non-Gaussian distribution — that could tell us an awful lot about the very early universe. This is what I research. I don’t look up in the sky to find clues about whether the primordial density perturbations are non-Gaussian — instead, I try to figure out new kinds of clues, and determine how well we can find these new clues using future sets of data that we’ll get from telescopes and satellites over the next few years. This helps theorists who are building new models of what the very early universe might have looked like — ideally, they would like to find evidence that their models are correct, so knowing how well we can find clues about non-Gaussianity can tell them whether their theories have any hope of being tested over the next few years. It can also help observers who want to know what kinds of observations they have to make if they want to be able to test whether the perturbations are Gaussian to a certain degree of accuracy.
So, in a nutshell: I try to figure out how much we can learn about the way stuff was spread out in the very early universe by looking at the way stuff is spread out now, so we can learn more about what happened at the beginning of time.
- This process is called Big Bang nucleosynthesis, and it only created hydrogen, helium, and lithium. Other elements didn’t get created until much later, by stars. All the elements on the periodic table between beryllium and iron were created by fusion in the cores of living stars; all the elements heavier than iron — from the gold in your ring to the tungsten in your lightbulbs to the copper and zinc in your pocket change – were created by supernovae as stars died. [↩]
- Except perhaps at the center of very young neutron stars. [↩]