At some point, you might hear a news story about a shipwreck. One of the sailors went overboard and he’s at risk of drowning. Suddenly, a dolphin popped up at his side and gently but firmly started to nudge him safely toward the shore. It is tempting to draw the conclusion that dolphins must like human beings if they take actions like this to save us from drowning. But how can we know this, that dolphins are actually intending to be helpful?

To answer this question, we would need to know how many shipwrecked sailors were gently nudged farther out to sea by dolphins.  The problem is we don’t know this number because those sailors would have drowned and would not be around to report their experiences. If we knew that dolphins randomly nudged sailors toward the shore and out to sea, then we might conclude that dolphins are neither benevolent or evil–they are just playful.

In order for any theory to be scientific, it has to be formulated in a way that it can be shown to be false as well as true. If every outcome confirms your hypothesis, your belief is a matter of faith and not science. 

Interesting concepts in the book I’m currently reading Mistakes Were Made (but not by me) by Carol Tavris and Elliot Aronson. This is one of Dr. Peter Attia’s most gifted books and I know one of @keni ‘s favorites.