How We Know the Universe is Ancient (Script) (Patreon)

The universe is precisely 13.8 billion year old - or so our best scientific methods tell us. But how do you learn the age of the universe when there’s no trace left of its beginnings? 

We see the same constellations on the night sky as did our astronomer ancestors. Familiar star-maps are recorded in cave paintings tens of thousands of years old. We could be forgiven for assuming that the universe above is fixed and unchanging. But the Earth was also thought to be timeless - until we learned to see billions of years of change in its geological layers. Well, the universe also changes and also had a beginning. In our recent episode we learned how we calculate the age of the Earth based on radioactive decay in its most ancient rocks. 

But there are no rocks from the beginning of the universe. There aren’t even photons from the time right after the Big Bang. So today we go deeper into deep time to understand how we can possibly have learned the birthday of the universe.

Actually, you know what? Forget for a moment knowing the age of the universe - our knowledge that the universe even had a beginning is relatively recent. Let’s talk about that first. The discovery of the beginning of the universe corresponds to our discovery that there even IS a universe outside the Milky Way galaxy. This is something we explored in our STELLAR series, but it’s worth a review.

At the beginning of the 20th century, astronomers were arguing about the nature of these faint, fuzzy patches of light on the sky known as spiral nebulae. Were they blobs of gas in the Milky Way, or vast, distant groups of stars - other “Milky Ways”, or as Immanuel Kant called them, island universes. We now call them galaxies. It all changed in the 1920s. First Vesto Slipher found that the spiral nebulae are moving away from us at incredible speeds, based on their Doppler shift - the lengthening of the wavelengths of their light due to their motion.

Then Edwin Hubble figured out the distances to these objects by watching their stars pulse. He located a star type called a Cepheid variable, whose pulsation rate is proportional to its brightness - encapsulated in a simple period-luminosity relation. By measuring the pulsation rate he could calculate its true brightness, undimmed by distance. He figured out that in order for these stars to appear so faint in his telescope, they had to be many millions of light years away.

Combining Slipher’s velocities or redshifts and Hubble’s distances, we learned that essentially galaxies are racing away from us. And the further the galaxy the faster it seemed to be retreating. 

This was a shocking discovery.  Until then, science had provided no reason to imagine that the universe was anything but static – that it had always been there, and that it had always looked pretty much the way it looks now.  The new picture was far more dynamic.  We live in an evolving, expanding universe.

The recession of the galaxies makes perfect sense in the context of Einstein’s then-new general theory of relativity. The Russian cosmologist Alexander Friedmann solved Einstein’s equations and found the possibility of a universe that could change in size - a result that Einstein himself dismissed at the time. In fact even before Hubble’s observation the Belgian physicist and Jesuit priest, Georges Lemaître, suggested that Vesto Slipher’s observed redshifts could be a sign of the universe’s expansion. 

Lemaitre put forward the idea that the world began in a state that he referred to as a “primeval atom.” Think of it this way - if the universe is expanding now, then in the past it was smaller. Rewind the clock according to the raw Friedman equations and there’s no alternative -  the universe must have once been in an extremely hot, dense state, and expanded from there.  

Astronomers call this the big bang model of the universe.  The name “big bang” was coined by the astronomer Fred Hoyle during a 1950 BBC radio broadcast.  But not to popularize the idea - rather to mock it. Hoyle was the last great holdout against the idea of a dynamical universe with a finite age.  His own “steady state” model.

This idea that the universe has a beginning brings to mind the creation stories of various mythic or religious traditions. Some have even tried to claim that the Big Bang is a validation of their tradition. Pope Pius XII certainly thought so. But priest or no, Georges Lemaître himself disagreed, stating: “As far as I can see, such a theory remains entirely outside any metaphysical or religious question.”

So if we can predict the big bang by rewinding the Friedman equations, surely we can also predict how long ago it happened. Think of it this way - if a particular galaxy is racing away from us, and we know how fast it’s moving and how far away it is, we can figure out how long it must have taken to cover that distance. In other words, how long ago was it right on top of us? Even doing that for a single distant galaxy gets us a rough estimate of when the Big Bang happened. However to accurately calculate the age of the universe this way you need to account for a few more things. 

First up there’s the fact that galaxies aren’t only moving away from us due to the expansion of the universe - they also have random motion as they’re tugged by the gravitational fields of nearby galaxies and clusters. We can deal with these “peculiar velocities” just by averaging over many galaxies.  

Then there’s the fact that the expansion rate has changed over time. You can’t just assume that the galaxies were always moving away at their current speeds. Now we’ll come back to that later. For now, we’ll see that we can do a lot just having a good idea of the current expansion rate. 

That rate is encapsulated in something we call the Hubble constant, which just tells how fast a galaxy is moving away from us given its distance - the more distant, the faster it’s moving  away. In the near-century since Edwin Hubble’s great discovery, thousands of astronomers have devoted their lives to measuring the Hubble constant - and a big part of the motivation is that it tells us the age of the universe. In fact if you take the fraction one over the Hubble constant - in the right units - you have the age of the universe, at least assuming no gravity. And if your Hubble constant was determined by looking at lots of galaxies, then it gives you an age averaged over lots of galaxies - and that averages out those pesky peculiar velocities. 

And when astronomers did this calculation in the early 1930s, they figured that the universe is a bit less than two billion years old. Even then that figure didn’t sound right.  By then, geologists had already found rocks here on earth that were at least 3 billion years old! 

So there was a problem - but it wasn’t a problem with the theory, it was a problem with the observations. Edwin Hubble had got the distances to the galaxies wrong. It turns out that Cepheid variable stars come in two types, with two different period-luminosity relations. Hubble had observed a brighter variety of Cepheids in distant galaxies, but he then used a period-luminosity relationship measured from a different, fainter class of Cepheids that were measured in the Milky Way. So, he measured the periods of his Cepheids and calculated what he thought were their true luminosities - but the numbers he got were too low. He thought they were intrinsically fainter than they really were, and so concluded that they had to be closer to us than they really were.  In fact he got distances wrong about a factor of two too small. Overnight the universe doubled in size. 

This whole 2-different-types of Cepheid variable issue was figured out by Walter Baade. In 1943, beneath the dark skies of wartime blackouts, he used the 200-inch telescope at Mt. Palomar to peer deeper than anyone had before into the Andromeda galaxy. With a newly calibrated Cepheid period-luminosity relation, and also incorporating other distance measures which I’ll come back to, Baade calculated a new age for the universe. At a conference in 1952, he announced that it must be 3.6 billion years old.  At last, the universe was older than the earth – although still a way off our modern value.

To understand the next step in improving this number, I need to explain another method astronomers had been using to get distances to galaxies, and so calculate the Hubble constant and the age of the universe. It’s similar to the Cepheid method because it’s based on comparing the expected, true brightness of stars with the apparent, distance-dimmed brightness. It seems fair to assume that on average the brightest stars in a galaxy have pretty much the same average brightness from one galaxy to another. So if the bright stars in a given galaxy appear faint on average in our telescopes, it means that galaxy is further away, if they appear brighter, then it’s closer. 

But astronomers had originally made an awkward mistake - they had been counting bright clouds of hydrogen gas - so-called HII regions - as stars, which threw their numbers off. It was a young astronomer named Alan Sandage who figured this out. Sandage was a student of Walter Baade and also worked closely with Hubble. By eliminating the problem with HII regions he came up with a new age estimate for the universe - 5.5 billion years. 

That was in the 50s. As the decade went on, Sandage continued to revise those estimates.  He eventually concluded that the expansion rate of the universe – the Hubble constant – was about 75 kilometers per second per megaparsec, which gives an age for the universe of around 13 billion years, assuming no gravity and dark energy! But that’s remarkably close to the currently accepted value. Sandage figured that with gravity playing its part to slow down the expansion, the universe must have been expanding faster in the past and so it should have taken less than 13 billion years to reach its current size. He estimated its age as between 7 and 13 billion years old.

Now there were some glitches. For a while we thought we’d found globular clusters - ancient, dense groups of stars - that were 15 billion years old. As bad as finding rocks older than the universe. But as our understanding of stellar evolution improved, those ages came down to under 13 billion years.

To arrive at our current, very precise value of 13.8 billion years, you need to take into account the effect of matter slowing down expansion through its gravity AND the effect of dark energy speeding it up. Those can be put into the Friedman equations and a bit of calculus later gets you an age. The real hard part is figuring out how much of all that stuff there actually is. The mass of the universe - which is mostly in dark matter - can be found by adding up the gravitational effect in galaxies and galaxy clusters, and also by tracking the past expansion history of the universe to observe the slowing effect of all of those galaxies on the whole universe. It was in an attempt to do the latter that astronomers discovered the anti-gravitational effect of dark energy, and had to start adding that into their equations also. As it turns out But that’s a story for another playlist. Long story short, 

These days, the gold standard for measuring the age of the universe is to get the matter content, the dark energy, the expansion rate, and more all from the one source - the cosmic microwave background radiation - the oldest light we can see, released when the universe was a mere 400,000 years old and MUCH smaller and hotter.  And for the details on the cosmic microwave background and how we use it to calculate the contents of the universe? Guess what - here are some episodes we prepared earlier. 

Based on analysis of the cosmic microwave background map produced by the Planck satellite we get the relative amounts of matter and dark energy and a number of other quantities. Quantities important for plugging into the Friedman equations, which define the way the universe would expand in the many billions of years following the release of the cosmic microwave background, and also how it had expanded before that time. There are other approaches to getting the necessary numbers to fuel the Friedman equations - and despite some intriguing conflicts - different methods are mostly in agreement. They converge on a single number. It’s been 13.8 billion years since the fiery beginning of time-as-we-know it, the birthday of spacetime.