A popular, famous and often-quoted misunderstanding lies on the very basis of the Evolution theory: “a species evolves into another species by random mutation“. This idea, repeated in many magazines and blogs and friends-talks, is simply and fundamentally wrong!

The bad side is that it’s sometimes used to distort the real mechanism of Evolution and to push Theistic Evolution or Intelligent Design ideas. The good side is that the concept is easy to understand and get it right.

Species are formed by *selection*, most often natural selection. Natural selection, albeit myopic, is not random: since individuals are slightly different, their chances of reproduction are also slightly different, so on average and after a big number of generations the most favorable features will get preserved.

Those differences were caused by mutations, indeed. However, mutations happen in a number larger than many people realize: many of them create differences that are not perceptible; other mutations don’t even create any differences (silent mutations); other mutations create bad cells that are killed right away; and some other mutations are immediately corrected by the DNA replication process itself.

In fact, it is possible to find different pairs of DNA threads between identical twins: over the course of life, cells multiply and mutations (mostly harmless) will occur. Even inside *you* it’s possible to find pairs of cells with different DNA threads. The difference is irrelevant but shows how mutations happen all the time and without purpose.

We *can* say DNA mutations are random because they are unpredictable and non-deterministic. What makes evolution work is the subsequent process of selection, eliminating harmful mutations and increasing the chances of favorable mutations (interestingly, sometimes letting some irrelevant or so-so mutations pass by, giving rise to biological variety, but that is another story).

The analogies with pattern recognition are flawed because they ignore all the other mutations that were not success full. It restricts itself to a final picture, like interviewing the guy that played a pair of dice and got double-sixes 10 times in a row.

If we could only see the final result, certainly would be hard to believe that the process of throwing dice was really random. It would make sense to think that perhaps there was a pattern, so complex we just couldn’t figure out. That something somehow was making sure the double-sixes kept coming, round after round.

But if you could see the hundreds of failed attempts to get double-sixes, all those times where other players got a 2-5 or 3-3 and got kicked out of the competition, then the whole picture gets clear: it was simply random.