How Do We Know the Age of the Earth? (Patreon)

In astronomy we talk about billions of years like it’s no big deal. But how can we be sure about timescales so far beyond the capacity for human intuition? Our discovery of what we now call deep time is very recent - as recent as our discovery of the true spatial vastness of our universe. And it came as scientists tried to measure the age of the Earth. What they found was as shocking and humbling as anything seen through the telescope.

---

“How long has the world been here?” The question is surely as old as human civilization -  perhaps older. The earliest written record we have of a cosmogony - an origin story for the cosmos - came from the Sumerians, followed soon by the Babylonians. There were floods, dismemberments of gods - all that good stuff that was later adopted by other traditions. But the main thing was this notion that the world had a beginning. The early Jews adopted this idea, and in turn it was picked up by the early Christians, and the idea of a single, unique creation event became a cornerstone of some of the earliest religious texts – like the Book of Genesis in the Hebrew and Christian bibles.

The Genesis story was so influential that many people used it to try to figure out the age of the world. Probably the most famous guess came from the Irish bishop James Ussher, in the early part of the 17th century.  Ussher looked not only at the Bible, but at hundreds of other ancient texts, trying to align the different – and often conflicting -- histories that they presented.  Usher eventually concluded that the world began at 6:00 p.m. on Saturday, October 22, 4004 BC. An oddly precise prediction given the source material - and obviously a bit lower than the value given to us by science.

The French naturalist Georges-Louis Leclerc, Comte de Buffon, was among the first to calculate the age of the Earth using what we would now call scientific methods, publishing his result in 1778. It was ingenious really. He assumed the Earth started as a ball of molten rock, which subsequently cooled down to its current temperature. Buffon spent six years measuring the cooling rates of materials in his lab, and in the end calculated that the Earth was 74,832 years old.  And yes, that’s also a weirdly precise number. Both Ussher and Buffon are docked half a point for excessive significant figures. Buffon’s premise was clever, if flawed. His 75 thousandish years was too small - and we now know why. I’ll come back to that.  

At around the same time as Buffon was staring at warm lumps of iron, the Scottish geologist James Hutton wandered Britain, pondering the ages of its rock formations. He became convinced that they were formed when molten rock pushed through the crust from Earth’s interior. This directly gave us igneous rocks like granite, while sedimentary layers resulted when these materials eroded and were deposited on ocean floors - to be later pushed up again in an endless cycle. Hutton’s biggest breakthrough was that reasoned that these processes were driven by the same forces operating in the world today - so-called uniformitarianism. But those processes were excruciatingly slow, and so Hutton realised the Earth must be unthinkably old.

We can thank Hutton and his 1788 book the Theory of the Earth for opening scientific eyes to the possibility of an ancient planet, and the idea of what we now call deep time. The initial temporal vertigo was shocking. Hutton collaborator, John Playfair, put it well: “the mind seemed to grow giddy by looking so far into the abyss of time.”

Hutton didn’t propose a beginning for the Earth - instead he assumed an infinite series of cycles. This was also the notion of the great German philosopher Immanuel Kant, a few decades earlier. Kant wasn’t afraid to throw out some numbers, based on pure speculation - to quote: “There has mayhap flown past a series of millions of years and centuries, etc. etc. ...  Creation is not the work of a moment.” By the way, Kant was also the first to have speculated about the existance of galaxies beyond the Milky way - Island Universes, as he called them.

Deep time and deep space. We’d known since Copernicus and Galileo that earth was just one planet among several in our solar system. Astronomers now swear by the Copernican principle - Earth is not in a privileged position in the universe, and so the laws of physics work the same everywhere. Hutton’s uniformitarianism is a sort of temporal Copernican principle for geology. The idea was further developed and popularized by another Scottish geologist, Charles Lyell. In Principles of Geology, published in the 1830s, Lyell talks about millions of years of geological processes. Hespeculates on a true beginning for the Earth, but doesn’t try to date it. Instead he imagines that it must be vastly older than geology can yet see, and he draws an astronomical analogy - that the universe must be vastly larger than what was visible at the time.

Lyell’s work was incredibly influential - and not least to a young scientist named Charls Darwin. Darwin read Lyell’s book during his famous voyage on the HMS Beagle. Lyell’s ideas gave Darwin the millions of years he needed  his theory of evolution, and it’s excrutiating slow mechanism of natural selection. But Darwin was also into the geology. In his Origin of Species, he estimated a minimum age of the earth based on the erosion timescale for chalk formations in Southern England. His figure of 300 million years was actually too high for that formation, but we now know much lower than the true age of the planet.

Of course geology and evolutionary biology are now intimately connected. We can trace the progress of evolution by mapping the fossil record to the geological clock. The ordering of  the appearance of ancient species is found when we date the rocks that their fossils are found in. We’ve now found fossils as old as 3.5 billion years - but to understand how we can possibly know that age, we have to turn from geology to physics.

The discovery of X-rays in 1895, and the discovery of radioactivity a year later, would open up a new world within the atom – and yield new tools for probing vast stretches of time.  

The New Zealand-born chemist Ernest Rutherford, working in Montreal, discovered that certain elements were unstable, and released energy at an ever-decreasing rate, over time.  He calculated that minerals that contained those radioactive elements could release energy for thousands of years or even millions of years.  We now know that radioactive decay is responsible for keeping the interior of the Earth hot - and explains why the Earth is much older than Buffon’s 75000 year estimate based on the passive cooling of a ball of molten rock.

But the discovery of radioactivity also gave us our more accurate way to figure out the age of chunks of the Earth - radiometric dating. Unstable atomic nuclei decay into lighter nuclei by splitting or by ejecting particles. The rate of decay is expressed in terms of “half-life” - which is the amount of time for a given radioactive nucleus to have a 50% chance of decaying; or equivalently, the amount of time it takes for half of a large number of radioactive nuclei to decay. In principle, if you know how much of the stuff there was to start with you can figure out how long the radioactive material has been decaying.

Figuring out the initial content is typically impossible - at least directly. Instead, there are some clever workarounds. Perhaps the most well known example is carbon dating. In this case we’re interested in the decay of carbon-14 - that’s the version or “isotope” of carbon with 8 neutrons. It’s radioactive, and decays with a half-life of 5,700 years. Carbon-12 and 13 isotopes are stable, and so much more abundant. Carbon-14 exists on the surface of the earth because it’s produced when cosmic rays hit nitrogen in the atmosphere, resulting in a constant proportion of C-14 within atmospheric CO2. That carbon is incorporated from the atmosphere into living organism by photosynthesis, but when the plant, or whatever ate the plant dies, the C-14 content gradually drops. By measuring the current C-14 content relative to the stable C12 and C13 content, the age of once-living material - including fossils can be determined. But this is only accurate to around 10 half-lives - or 50,000 years.

Move useful for measuring the age of the earth is uranium-lead dating. Uranium decays on much longer timescales - 710 million years for the U-235 isotope and 4.5 billion years for U-238, with both decaying to different isotopes of lead. You can figure out how much uranium a given sample had to start by looking at the proportion of uranium to lead. This only works if you can be sure that no lead was in the sample to start with - and in some cases you CAN be sure. Some crystals like zircon tend to incorporate uranium atoms into their crystal structure when forming, while at the same time repelling lead. So any lead you find in those crystals came from uranium decay.

By looking at the ratios of each type of uranium to the corresponding isotope of lead you get two independent ages - which match up if the crystal has not lost any of its lead. But the comparison of the two allows you to correct for any lead loss. The other cool thing about this tequnique is that the different half-lives of each uranium isotope means this radiometric technique is useful from between hundreds of millions to many billions of years.

By the 1920s, British geologist Arthur Holmes would declare that the Earth was between 1.6 to 3.0 billion years old, based on his radiometric dating. This was around the same time that astronomers proved that Immanuel Kant’s Island Universes were indeed other galaxies, many millions of light years away. The world simultaneously got a lot older and the universe a lot bigger.

Older rocks were discovered, pushing back Earth’s age further and further. But beyond a few billion years it starts to get tricky. There aren’t many patches of land left from before then - most of the original crust has been recycled into the mantle. Once such spot is in Western Australia. There,  zircon crystals have been found as old as 4.4 billion years, based on uranium-lead radiometric dating. But to go beyond that date we have to look beyond the Earth. We beleive that the moon formed at the same time as the Earth - both coallescing after a giant planetary impact in the early solar system. The moon is techtonically inactive, so rocks on its surface now where there when it formed. The Apollo missions bround back lunar specimens that have been radiometrically dated to 4.5. billion years.

That jibes with our measure for the age of solar system. Nearly 4.6 billion years - we get the same number from radiometric dating of solar system meteorites and also from our calculations of the age of the Sun.

I should add that science took its time getting to these large numbers for the age of the world. The Mayan long-count calendar includes cycles of 63,000 years. Hindu tradition also has a cyclic cosmology, with nested epochs lasting millions of years, In fact a single day of Brahma is 4.32 billion years. Oddly close to the age of the Earth - although it looks like we survived one day of Brahma AND the last long-count transition, so I’d say we’re good for now.

So the picture has come together. There’s consistency in that 4.5 billion-ish year age across several independent measures. That gives us some confidence in the result. The confidence in the number may be justified, but perhaps we’ve become a bit too comfortable with it. These days we stare down at the unfathomable gulf of the past and shrug, and we forget that entire span of human civilization is the width of a single hair on our heads compared to the stratified depths of the Grand Canyon. Not that our ancient Earth cares - it’s rotated 4.5 billion times through deep space and out of deep time, and it’ll do the same again. Into what I guess you could call deep spacetime.