The world’s been around for a while. And although there’s no shortage of cool stuff in it for us to look at, some people aren’t content with being limited to everything that’s around now. They reckon that some of the stuff that happened “before now”, in a period of time experts refer to as “the past”, might also be quite interesting. To learn things about the past, and what exactly went on there, we’ll have to have some way of extracting this information, from things we can look at in the present.
Sometimes this is simple. If you want to know what was on the mind of any given emo poet in the last few years, there’s LiveJournal. This and many other equally vital pieces of information are stored, regularly and reliably, and provide an easily accessed insight into yesterday, yestermonth, and beyond. There are also news archives, parish records, and diaries kept by rather more notable historical figures than the emo poets of the 21st century, to give us a more rounded view. Sometimes these can be unreliable sources, riddled with political and personal bias (that Adrian Mole clearly can’t be trusted, for one), but they at least provide some insight into what the past was generally about, even if we can never totally avoid the danger of inadvertently buying into someone’s own line of bullshit.
But sometimes we want to know about things that nobody wrote anything about, or passed down any credible verbal stories about, or filmed on their camera-phones. One important part of this is finding out how old something is, for those times when you don’t also manage to dig up the diary of the guy who put it there. Dating historical artefacts is a massive and complicated area of science, embracing many fields of study, in none of which I am qualified to claim even rudimentary competence, let alone expertise. This, then, will be a brief description of radiometric dating, a broad and shallow summary of how it works, without too much detail just yet as to how we know it works as well as we do. There’ll be a few other articles branching off from this one soon, looking at some of the common complaints and objections raised to it, to the extent that my poor monkey-brain can cope with the intellectual rigour of it all.
Up and atom
Before we can get into the meat of this topic, we’ll need to cover some basic particle physics. But you guys like reading lengthy essays on the internet about particle physics, right? I mean, porn can only hold your attention for so long before you start lusting for some hard science action, am I right? Well, if not, you can skip down to the next heading in bold, but I’ve written this bit now so I guess I’ll leave it here.
Good. So, physics. Broadly speaking, all the stuff that everything’s made of is itself made of three types of basic particles: protons, neutrons, and electrons. (Don’t worry about the dozens of others. It’s fine to simplify for now. There won’t be a quiz.) In an atom of any particular type of stuff, some protons and neutrons will be clustered together in a big dense ball (the nucleus of the atom), and some number of electrons will be floating way out around them (relatively very far, but still well under a billionth of a meter away).
Let’s pick one particular type of stuff as an example. Look at a pencil. The stuff in the middle that you write with is graphite, which is a form of carbon. Each atom of carbon is made up of those particles as described above: specifically, six protons and six neutrons bundled together in a ball in the centre, and six electrons circling around them1. That arrangement is what makes it carbon.
If, instead, you have just two protons and two neutrons in the middle, with two electrons orbiting, you get helium, the gas responsible for one of the most univerally hilarious phenomena known to science: making people’s voices go squeaky.
The numbers of particles don’t always have to match up exactly, like they do in those two examples. Sometimes there are more neutrons than protons, or very occasionally vice versa – but you can’t just throw any old mess of protons and neutrons into a nucleus and expect to make an atom. There are forces pushing them apart and pulling them together, kinda like gravity or magnetism pull at things sometimes, and so only some arrangements are stable.
For instance, the carbon I described up there has six of each type of particle. But that’s not actually the only way to make carbon. The number of protons is what defines what element a molecule is, so anything with six protons is carbon. You couldn’t have a nucleus with only those six protons, though, and no neutrons, because then there’d be a strong magnetic force pushing them away from each other (the same kind of magnetic force which makes your fridge magnets not want to get too close to each other), so it’d all fall apart. But if you put six neutrons in there as well, they provide a sort of binding force (I hope it’s clear that I’m still greatly over-simplifying all this) and hold the whole lot together. Because this particular type of carbon (one of carbon’s isotopes) has twelve particles in the nucleus (six protons, six neutrons), this is called Carbon-12. Carbon-12 is stable – the nucleus bit in the middle won’t ever just fall apart on its own – and around 99% of the carbon you’ll ever see will be like this.
Carbon has one more stable isotope, Carbon-13, which has seven neutrons instead of six in the nucleus. There’s still six protons (otherwise it wouldn’t be carbon), and six electrons (otherwise it would be ionised, which isn’t worth worrying about just now). This makes up almost all the rest of the carbon, the stuff which isn’t Carbon-12.
And then there’s one more type, Carbon-14, which has – anybody? Bueller? – eight neutrons. There’s very little of this stuff around on Earth – about one in a trillion carbon atoms are of this type – and unlike the other two, Carbon-14 is radioactive. It won’t just sit still being carbon, like its isotopes tend to do (and for which your DNA should be grateful), but over time it decays into different types of matter, because all the stuff in the nucleus can’t quite hold itself together indefinitely. In this case, Carbon-14 will undergo what’s called beta decay.
people atoms, so many people atoms2
Beta decay means that, every so often, one of the neutrons in an atom of Carbon-14 will suddenly transmogrify into a proton, and emit an electron (which would make your Geiger counter click, if you happened to be waving one around)3. Because of this extra proton, it’s now an atom of Nitrogen-14, and not carbon any more. This could happen at any moment, depending partly on the stability of the atom in question. Some radioactive stuff will tend to linger for years; other elements will decay into something else much more quickly. How long the radioactive stuff stays around is determined by its half-life.
The half-life of a radioactive element is the amount of time in which you would expect half of a sample of it to have decayed – or, the time after which the odds are 50/50 of any particular atom decaying. The reason I use words like “expect” is that it’s still random, so you can never be certain exactly when your Geiger counter will click, but it’s guided by a measurable process.
Time for a dubiously appropriate real-world mathematical analogy. Imagine you’ve got a thousand coins laid out on a table, all currently turned to heads. You flip them all once, and get rid of the ones that come up tails. You’d expect about half of them to go – maybe not exactly 500 out of 1000, but pretty close, because that’s the average you’d expect when the odds are 50/50 either way. Radioactive decay is like that: after a certain time (one half-life, or one round of coin-flips), you expect half your atoms to have decayed, on average. Flip the remaining 500-ish again, and about half will go again, one further half-life later, and so on.
Now, if I know that this is what you’re doing, I can figure out how many times you’ve been through this process of flipping all the coins and removing the tails. If I know you started with 1000, and now there’s around 125, I can be almost certain that you’ve done it three times (because when you take 1000 and halve it three times, 125 is what you’d expect to end up with). In a similar way, if scientists know how fast some radioactive stuff is decaying, then by looking at how much of it there was and how much there is now, they can tell how long it’s been decaying for, and thus how old it is.
The half-lives of radioactive stuff range from tiny fractions of a second, to billions of years. So, onto some specifics of how we can use this to actually work stuff out.
All we hear is, radiocarbon
To a lot of people, any way of finding out how old some really old stuff is, is basically “carbon-dating”. And by “a lot of people”, I mean me until a few paragraphs ago. It’s actually only one of many ways we can use radioactive decay to measure how old something is, but it’s a useful one, so let’s focus on that first.
Most carbon is Carbon-12, as you may remember from that bit you probably skipped over. A very small proportion is Carbon-14, which has a half-life of around 5,730 years.
We’ve found this out to a reasonable level of accuracy (the error bars usually given are about 40 years either way), basically by just taking a sample of C-14 atoms, and watching how fast they decay. Fortunately we don’t need to wait an entire half-life to get a useful answer, because unlike the individual coin-tosses in the example earlier, radioactive decay is a continuous process – so, if half of a sample will have decayed after 5,730 years, a much smaller (but mathematically calculable) sample will have decayed after a much shorter, more easily measurable time.
It might sound like only a really, really tiny amount would decay in a short enough time to measure, if it takes millennia for half of the stuff to go, and that it could therefore never be measured with any useful accuracy. But atoms are really, really, really small, and a tiny fraction of a gram can contain billions of them, easily enough to get a meaningful measurement of how fast the stuff is decaying. So, we know Carbon-14’s half-life, to within a low (and also well known) level of certainty.
So now that we know the rate of C-14 decay, we just need to know how much of it something started with, and measure how much is left now, to tell how old it is.
Plants take in carbon dioxide from the air in photosynthesis, some of which will be C-14. Other living things that don’t photosynthesise (such as “animals”, in technical biology jargon) tend to eat these plants (or other animals, with plants somewhere further down the chain). So every living thing is continually taking in and excreting some C-14, mixed in with all the rest of the carbon in their diet. The overall level of C-14 in their bodies stays fairly constant, depending on how much of it is in the air they’re breathing, and the stuff they’re eating.
Once they’re dead, though, this cycle comes to an abrupt halt, and the C-14 inside them is left to just radioactively decay. Because it’s only after the moment of death – when they stop breathing and eating other things with carbon in – that the amount of C-14 begins to drop off rather than being continually topped up, we can estimate how long ago the poor sod whose carbon content we’re poking around in died.
That’s basically carbon dating. It’s useful, but obviously limited. It’s not generally used on samples older than 60,000 years, since after more than ten half-lives or so, there’s such a small proportion of the original C-14 left undecayed that it’s impossible to tell with much accuracy how long it’s been decaying. Also, not everything even has any carbon in it to measure. But potassium-argon dating can be useful with rocks upwards of 100,000 years old, and uranium-lead dating is what you want if you’ve got some zircon that’s been lying around for billions of years, and there are many other similar techniques. The basic priciples of radioactive decay are the same in each case.
I wish to register a complaint
Unsurprisingly, not everyone’s happy with of this. Many people disagree with the conclusions of radiometric dating. Just as unsurprisingly, a lot of the complaints stem from a misunderstanding of how science works. In particular, the detractors of radiometric dating methods tend to approach it with an all-or-nothing mindset, and assume that a single result that seems unreliable, or any sign that it doesn’t work perfectly in all circumstances, must nullify every aspect of the discipline.
This drastically oversimplifies things; obviously we should be on the lookout for faulty results and anything which calls these methods into question, but it’s not implausible to suggest that we might be able to understand its particular shortcomings, and come up with a more limited (but still useful) system in which the results we get are still valid, and can be shown to be consistent, replicable, and entirely scientific. The pen I’m currently holding in my hand isn’t plummeting straight down into the Earth’s gravity well, but that just means we need to understand Newton’s laws in context, and be sure of how they can or can’t be applied, if we want to avoid being reckless and leaping to ridiculous conclusions. The scientists who developed these methods have had plenty of time to figure out their shortcomings.
The people who have a fundamental problem with the entire field of radiometric dating, though, often seem less than clear on what errors led to the supposedly faulty conclusions. For many of them, I suspect their doubts didn’t originate with a flaw they uncovered in the scientific methodology being used – but strict Biblical doctrine implies that the world is no more than 10,000 years old, so if that godless, immutable behemoth of Science somehow says that some things have been around for millions of years, then it’s obviously gone wrong somewhere. The details of where these mistakes have been made is rarely given quite so much thought – it’s only obvious that they must have been made somewhere, because the results disagree with the inviolate axiom of Biblical infallibility and Jesusy goodness, on which their world turns.
When fundamentalists are trying to protect their dogma from an onslaught of pesky facts, the most useful pseudoscience tends to concern theories at a slight remove from what’s commonly understood and accepted by laypeople. Nobody wastes any time denying that elements consist of a nucleus of protons and neutrons, and surrounding electrons, just as there’s no real dispute these days that genetic information is passed from living organisms to their offspring by DNA – these are things we all kind of understand, and they don’t threaten anyone’s ideology, so it’s universally settled that we’ve got these things more or less nailed. (This is always subject to change if new data arises, of course, but we’ve observed enough by now that we can be pretty confident we’re on the right track.)
But the controversy comes when the science gets a bit trickier, the concepts less easily grasped by anyone who hasn’t spent time studying them, and the implications less comfortably reconciled with the dogmatic claims of belief systems. Because of this, fields of study which are equally well established and supported by evidence start getting labelled as “controversial”, and it can easily seem like some scientific theories are much weaker than in fact they are. You won’t find many evolutionary biologists who think that the basic idea of evolution (that life on Earth evolved through a process of Darwinian natural selection) is any less likely to be true than the fact that our genes carry our genetic information, but it’s a much simpler task to spin one as a “controversy” than the other. The evidence supporting it is overwhelming, but a lot of it is harder to understand, so it may be less intuitive to accept “Because it’s obviously true” as a reason why the scientific community seems to be so unanimously behind it.
Of course, simply labelling anyone who disagrees with the majority view as an ideologue who doesn’t understand the subject matter, and ignoring their arguments, would be a colossal logical fallacy. If these “scientific” objections are as out of touch with real science as all that, then even I should be able to explain the flaws behind some of them. But it’s taken far too long already to put this together, so I’ll cut this off early, and talk about some of those problems in separate articles soon. In the meantime, you might find enlightenment at the Radiometric Dating Resource List, or these articles by Dave Matson, or the ever-invaluable TalkOrigins, or any of the other articles linked to further above.