So you've started a website...
The aim of this blog is to share my love of Science. Not just the facts we've amassed over the centuries, but the process we have refined for doing so. I suppose the most sensible thing to do in my first proper blog would therefore be to explain what Science actually is.
Usually when people talk about Science they mean subjects like Physics, Biology, Chemistry, Astronomy, Geology etc. These are the things we learn in school and have to memorise in order to pass an exam. The text books we read are essentially lists of facts the author seems to just know. What’s rarely included is the story of how the facts were discovered and verified.
This approach to teaching Science is unfortunate but sadly unavoidable. If books explained how each fact was arrived at they would be fifty times their size. Sadly, a lot of people come away from Science class seeing it as a collection of information, rather than a reliable method. Science, in reality isn't just a noun referring to facts, it's a verb referring to a process. A process which ought to work like this:
Let’s start at the top. All Science begins with not knowing something. Science is born in ignorance. Maybe we don’t know what happens to apples when we let go of them, maybe we don’t know how often chimpanzees fight in the wild or maybe we don’t know what will happen if we mix two chemicals together.
First thing we do is get out and do an experiment. The alternative is to try and work everything out in our heads, what’s called a priori thinking. Science tends to be very cautious of the a priori approach because it makes the assumption that people think perfectly. Science tends to assume that even brilliantly clever people make mistakes, so nobody can just decide what is true from the vacuum of their brain. Nope. If you want to know about a thing, you have to actually go and look at said thing.
After the experiment is completed, we’ve got some data. We might have discovered that all accelerate toward the ground, that chimpanzees fight once every hour and that the chemicals turn blue when mixed. We’ve learned a simple truth about the world but we’re lacking an explanation. So we come up with a “hypothesis”.
A hypothesis is another name for an educated guess. It’s a potential explanation which must explain everything, not contradict itself and, most importantly, be falsifiable.
Imagine a dragon
Falsifiability is a key principle of Science. It’s the idea that a hypothesis must be capable of being proven wrong. Consider the following thought experiment known as “Sagan’s dragon”. Suppose I claimed to have an invisible dragon living in my house who made the Sun rise. Not only is he invisible but he floats and never leaves footprints, he is non-corporeal so you can’t throw a blanket over him, he’s also silent and gives off no heat signature. In fact there is no way of detecting his presence at all, but he’s definitely still there. Prove me wrong.
Such a hypothesis does provide an explanation for what makes the Sun rise. In fact, it might be accurate. There really might be such an invisible dragon living in my house. But if you can’t prove the idea wrong how do you test to see if it’s true?
The problem with a non-falsifiable hypothesis is that if we accept it, we have to ask: why not an invisible walrus too? Or an invisible chipmunk-robot? Why not an army of undetectable pixies living in my washing machine who control all aspects of the weather? Our world may be filled with such creatures and we would never know.
If you accept one non-falsifiable hypothesis, you have to accept every other one since they have an equal amount of evidence for them: none. Our picture of the world quickly becomes infinite under such a view. It might really be like this but we can’t know with any confidence so we have to ignore such claims, not because we think they’re untrue (they might be right) but because they are un-knowable.
Once we have our hypothesis clearly defined, we put it to another test, one which measures it specifically. This is the kind of thing we learn about in school as “fair testing”. Remove anything which could throw the results off and anything which could be a distraction. Arrange it so that only a few things are changing and only a few things are being measured.
Then do your test repeatedly. This is to get around coincidences because even the simplest “sugar dissolves better in hot water” hypothesis has lots of potential misleading factors. The temperature of the room, the quality of the sugar, the type of water etc. You need to test over and over to make sure the hypothesis wasn’t confirmed or rejected by accident the first time.
At this point your hypothesis is either proven wrong or it looks hopeful (note: we don’t say we’ve proven it right, just we could be onto something). If it's disproven we chuck it out without a thought. At least that's the idea. Unfortunately though, we’re human. We want our idea to be correct because we’re proud of it and if our hypothesis is proven wrong it’s completely normal to get frustrated or disappointed.
In an ideal world, proving a hypothesis wrong would be seen as good because we’ve still learned something, but humans aren’t ideal and it can be difficult to think like that. Even a hard-nosed Scientist feels a slight pang of annoyance if their cherished hypothesis fails.
We do our best to put up our hands and say “ok, I guessed wrong, back to the drawing board” but we’re full of emotional biases and the desire to achieve positive results, even though negative ones are just as useful.
In addition to these instinctive responses, there are all sorts of other things which could make our conclusion faulty. Bad equipment, poorly designed tests, mistakes carried out by accident and even conscious dishonesty. That’s why Science introduces the next crucial stage: we bring in other people to check it.
Prove me wrong, I dare you
After you've done your experiment, you don't just sit on it and decide you're correct, you cast your evidence on the ravenous mercy of other Scientists and hope for the best. This is frightening because you are opening yourself up to heavy criticism, but it’s vital. You need a second pair of eyes to look over what you've said and see if you've made any errors.
If you’re doing Science in your kitchen then you can get your friend to check the results but if you’re working in Scientific research the standard way to release your results is to submit them to a journal - a sort of weekly Scientific newspaper.
The journal’s editorial team will then submit it to "peer review". This is where they send your article to other researchers in the same field to try and find flaws. This is usually done anonymously (to avoid bribery etc.) and it is a brilliant idea. You are essentially showing your work to your rivals, the people most likely to criticise and destroy it. So if they can’t find fault, you really might be onto something. Your hypothesis has passed another hurdle and the results are published.
It works in theory
Now the hypothesis (and the experiment you used to test it) is in the public domain, so other people can see what they think. If they repeat or adapt your experiment and find the hypothesis holds up then you can start getting excited. But if they can’t reproduce your findings you have to swallow pride, abandon ship and start all over.
But let’s say you’re lucky. Let’s say you’re in that tiny fraction of a fraction of people who come up with an accurate hypothesis. This is when it picks up speed. Once the hypothesis has been tested many times, by lots of people over a long period and has been confirmed on every occasion, explaining every result, then the hypothesis is probably correct.
The hypothesis graduates from speculation to confidence and people start referring to it as a “theory”. A theory is not one person’s guess. It’s a statement about the world which has been put through the grinder and stood the tests of both time and savage scrutiny. This is what gets published in your high-school text books and how you win your Nobel prize. But we don’t close the case because the Scientific method goes in a circle.
We might use words like “Scientifically proven” or “Scientific fact” but these are shorthand terms for what we really mean. No theory is beyond question. Ever. A theory is something we’re 99.99% confident of, but any theory is based on falsifiable ideas and a single piece of evidence can bring it crashing down. Rightly so.
Take gravity for instance. Isaac Newton drew up a hypothesis to describe the attraction between two objects in 1687. The more people tested it the more confident they became, and pretty soon we had a “theory of gravity”. A Scientific fact.
But later measurements of Mercury and its orbit around the Sun turned out not to fit Newton’s theory. So, rather than sticking to Newton blindly, we decided his theory of gravity was incomplete.
Then in 1915, Albert Einstein published his own ideas on gravity which explained the anomalies and accounted for why Newton had seemed correct up until then. Einstein's ideas were tested, shared and confirmed over and over. Today we call it Einstein’s "theory of general relativity". But we don’t say job done and declare Einstein’s theory to be iron-clad. In fact, Einstein’s theory seems to break down in certain circumstances so we still don’t have the complete picture.
The key point is: if, at any point, someone had declared their knowledge of gravity complete, they’d find themselves looking pretty stupid a few years later. Scientists are aware of this so they’re very reluctant to make bold claims unless they have incredible evidence on their side. Most of the time Scientists guess wrong and they know it.
Ultimately, Science is a self-correcting, painstaking endeavour which makes progress at a rate of inches per decade. Is it a perfect system? Of course not, it's subject to all sorts of problems and limitations. But when you consider the alternatives i.e. guessing or going with your gut instinct, it becomes very obvious why a thinking person has no choice but to adopt a Scientific approach to truth-gathering.
I love science, let me tell you why.