This just isn't right. It's a horrible logical fallacy, in fact. The most likely number of sentient species depends on how likely it is for intelligence to form. We have no idea what the chances are of this, but we do know that:
- Earth was formed 4.6 bn years ago
- Water turned up 4.4 bn years ago
- Simple life arrived 0.1bn years after that. OK, 100 million years is a long time but in the scheme of things, that's pretty quick. A tick for life forming relatively easily.
- It then took a billion years before prokaryotes appeared. That's really quite a long time to go from something of ultimate simplicity to something still really very simple.
- Another 1.5 billion years before eukaryotes appeared. Blimey, it's taken 2.5 billion years just to get some other simple one-celled organisms
- Another billion years before we get to multi-cellular organisms. So simple life: 0.1bn years. Multi-cellular life: another 3.5 billion years!
- Everything else is pretty fucking quick, all things considered. That to creatures as we know them in about 0.3 billion years.
Now, I find that interesting because when we think of the development of sentient life, I think we as laypeople tend to focus on (a) the likelihood of finding life, and then (b) the evolution from, like, squirrels and shit to us. But actually this tells us that the difficult bit is going from "life" to "really, really simple but slightly more complex life". It takes literally billions and billions of years. That's how long it took for the couple of random things to happen that made really basic multi-cellular life possible.
So we know that the odds on the necessary random things happening are really stacked against, even if the planet is perfectly set up to create and sustain life in the first place. And then of all the planets we know, how many are really even vaguely set up to create and sustain life?
So maybe 1/1000 planets can get to sustainable life and the typical return period for such life to become multi-cellular is 4bn years. Or maybe we got lucky and the return period is actually 100bn years.
If so, it could well be that across the whole galaxy (say) of
100 billion terrestrial-like planets the odds are such that the mode of the distribution of planets with multi-cellular life is zero but with a mean of, say, 0.5. I'm thinking of something like an
Overdispersed Poisson distribution such that you'd mostly simulate 0 but sometimes get 1, very rarely get 2, super-rarely get 3 and basically never get 4+.