We often read that science has “proven” something…. However, for a scientist this is an immediate red flag, as we should never use the term "proven" in a scientific context. Here I will explain why everyone should see red flags when they read “science-proven” or “scientifically proven”.
Why we should never use the word “proven” within a scientific context
Proven means that it is absolute. There is absolutely no chance that there is an alternative explanation. There are plenty of examples where people were convinced they had proven something, only to discover later that they were wrong. Our observations often play tricks on us and even though we do everything possible to verify and make sure we are right; it is never 100% certain that your observation and the interpretation are correct.
If we take a simple example and look at Figure 1a, we are easily convinced that the two grey squares (A and B) are different shades of grey. If we designed a study and asked a group of 100 individuals, the vast majority, maybe even all of them would agree with this statement.
However, the GIF below (Figure 1b) will show that this statement is wrong. The context of everything else around it, and the shade that is introduced makes our brain think that the colours are different. Our conclusion was influenced by everything else we were observing. Carefully removing all those factors will reveal the real truth. This is just an illustration that our eyes can play tricks on us.
Below are two more examples. In Figure 2a, focus on the dot in the centre and move your head back and forwards. This gives the illusion that the circles are moving, even though they are not. The next figure (Figure 2b) consists of perfect lines and squares even though our initial interpretation would lead us to believe otherwise. The bottom-line is that our observations are not perfect.
For a long time people thought the Earth was flat. They were convinced it was flat. They thought it was “proven”. Until it was demonstrated that it was not. Another example is the miasma theory, which proposed that all diseases were transmitted by "bad air", often referred to as dark air or black air, until it was demonstrated that diseases are caused by microorganisms. Another instance is related to heat transfer. It was thought to be a substance or invisible gas which flowed from hot materials to cold materials. Scientists in the 1700s were convinced that caloric would flow from one material to another. When you place a hot piece of metal on a cold one, the cold one will become hot. It was thought this was because caloric was flowing from the hot metal to the cold. You could even feel it with your hands, so scientists considered it “proven”.
"The bottom-line is that our observations are not perfect"
Why science does not “prove” anything
There are 4 main reasons why science cannot "prove" anything:
1). Difficulties with observations
The above examples of optical illusions are a clear demonstration that eyes can play tricks on us. The same is true for our measuring devices and their operators. In science, much is done to prevent wrong observations. Wrong observations come in many different shapes and forms. It may be wrong because it is not actually measuring what we think we are measuring or the measurement may not be reproducible (we refer to this as validity and reliability). Of course researchers do everything they can to make sure measurements are valid and reliable. There are calibrations, verification of data, replication, confirmations by independent measurements, and so on. But we can never completely eliminate error.
2). Difficulties with interpretation
But imagine that the data collected is free from all error (a hypothetical situation), our mind, that has to interpret this data, is not. We see the world in a way that is influenced by many factors: our genes, culture, education etc etc. There are many forms of bias that will creep in as a result of this. Again, researchers do everything to minimise bias. They are trained to recognise potential bias and avoid it. But bias is everywhere, and it can never completely be avoided or removed. An example of bias is publication bias. Studies with a positive finding (for example supplement X has an effect) are more likely to be published than studies without a positive finding (supplement X had no effect). Of course, both outcomes are important but if only the positive outcomes are published this will result in a bias. So even when our measurements and observations would be free of any error, the interpretation of the data is not.
3). Context
To build on this interpretation issue a little more: a truth requires a context. So, if a statement is true, it may only be true in a very specific situation. For example, carbohydrate feeding improves performance. It very likely does during prolonged exercise, but it may not do this during sprint performance, such as a 60m dash. Any general conclusion that seems to generalise more, is less likely to be true. There are always situations where it may not be. So we can’t say something is proven because there may always be situations where this may not work.
4). Incomplete information
Sometimes we simply don’t have the right data or sufficient data to come to solid conclusions. We hear scientists often say: “we need to do more studies” or “ we need more data”. If we have two studies and they don’t have exactly the same results, it would be good to do a third study to find out why. However, resources and time are not unlimited. Science is moving slowly (more slowly in certain areas than others) as a result of both of these factors. The meticulous work of a single research project sometimes takes years. And one research project may only unravel a very small piece of the puzzle. Even in areas where there is apparent agreement and we must be aware that it is possible that, at any time, new and contradictory information could arise. The more studies we have with similar findings, the less likely it will be that we will ever have to revise our thinking, but we can never be 100% certain.
"We can draw conclusions and we can be confident that these conclusions are very likely correct... just never 100% certain"
So can we never draw conclusions? Can we not be confident about anything? Of course we can. We can draw conclusions and we can be confident that these conclusions are very likely correct. We can be confident – incredibly, absolutely, positively confident – just never 100% certain. There are a number of great quotes by the great scientist and thinker Richard Feynman:
“We absolutely must leave room for doubt or there is no progress and no learning. People search for certainty. But there is no certainty.”
“I have approximate answers and possible beliefs in different degrees of certainty about different things, but I’m not absolutely sure of anything.”
“Scientific knowledge is a body of statements of varying degrees of certainty — some most unsure, some nearly sure, but none absolutely certain.”
Can we never use the word "proven"?
We can… there are certain contexts where the word "proven" is justified. It is possible to use the word proven in situations where there is 100% certainty. In Maths and in logic this is possible. 1+1=2 in all situations. This is different in science where we will never achieve 100% certainty.
Of course it is baffling that there are so many people who KNOW that something is true, even though they don’t base it on any scientific evidence or any other evidence… There is no effort to even check it is true, they just KNOW it is… Scientists spend so much time and effort checking if what we believe to be true is really true and then end up with an answer that is likely true, but they are still not 100% certain. Read also the article on the Dunning Kruger effect…. A little knowledge gives people huge confidence and people quickly draw conclusions and shout about it, whilst learning more and knowing more will actually raise more questions and increase uncertainty... and it requires a lot of knowledge before you can climb out of this valley of despair…
So next time you read that science has "proven" something, you can be certain that you are not dealing with science.
Comentários