P2P No. 38 — Don't fall in love with a story

You should not want your story to be true, but find the true one.

We tell ourselves stories to understand the world. Narratives drive the economy. But they can be dangerous. And I am not talking about the tooth fairy (sorry if I lifted the veil off that one).

A narrative is a means to express a perceived pattern, thus, compressing the information required for understanding. You could memorize all leap years or learn the rule instead. Cognitive scientists call this chunking, i.e., summarizing pieces of knowledge compactly and without information loss.

However, what if the story is wrong? When astronomers started making more accurate observations, the geocentric model of the universe could have been shattered. But the belief in geocentric celestial mechanics was so strong that people did not spare effort to tinker with the equations to fit the data. Switching to the heliocentric model has led to a more straightforward narrative.

You might dismiss this story as irrelevant since the scientific method was not even born. I wish this would have been an outlier, but it is not. As Peter Medawar reports in his essay on The Strange Case of the Spotted Mice, scientists can hold such strong beliefs that they sometimes temp the evidence. But let us not attribute things to malice that can be explained by something else. Incompetence? Lack of critical thinking? The answer does not matter as much as recognizing the toughness of the problem in general: aside from temping the evidence, which is unacceptable, sometimes the quandary is disguised. We should not cherry-pick our best results, but not all results are relevant.

We should not fall prey to confirmation bias and neglect all contradictory evidence. It is hard (no, it is exceptionally taxing!) to let go of our brainchildren, but that is the time when we need to remind ourselves what we want:

Do we want to prove ourselves right or do we want to find out the truth?

So let go of all the stories we tell, shouldn't we? By no means: the narratives we form contribute to science; they are the hypotheses we seek to test—even if they fail to explain everything and are only approximate (we still use Newtonian mechanics despite being proven wrong by Einstein's theory of relativity). But we should not fall in love with any of them; they are ungrateful lovers.