Don't We Already Know This?

Ok, I've had a chance to actually look at the paper, and I have to retract some of my earlier harsh criticisms. The criticism really needs to be leveled at the various synopses, including the main one linked by the article. This paper isn't a claim about the likelihood of extra-terrestrial intelligence. It really doesn't set out to do any such thing.

What Watson is doing in this paper is taking some newer theories about the development of the Earth's biosphere and using them to update a mathematical model of the evolutionary process.

He takes the model, which basically says, "Let's assume that the process of evolution from biogenesis to intelligent life doesn't happen by slow, smooth steps, but rather that there are one or more 'critical' events that have to occur, which are individually fairly unlikely." Some natural questions, then, would be how many such events were there, and how unlikely are they?

So, as a simplifying assumption, he takes "fairly unlikely" to mean that, in the entire habitable period of the planet's development, the expected number of occurrences for the event is less than one. Now, that doesn't mean that the expected number is zero - an expectation, even for discrete events, can be a fraction, which means that on some planets it happens, on some it doesn't. If the expected number of occurrences is 0.1, then on 10% of all planets (with equal length habitable periods), most likely one of them would have the event occur.

In a model like that, when the critical events do occurr, it's most likely that they happen in roughly equally spaced intervals - regardless of the actual improbability of any one of them. So, if we take the time when biogenesis occurred and the time when intelligent life evolved as two of the points, the model then tells us how likely it is that some given number of critical events could have occurred with that pattern. We can compare those and find out the most likely number of critical events, and Watson comes up with four.

Now here's where we stop and take note. First, he didn't say, "let's assume that A, B, C and D are critical events, and let's (arbitrarily) assign them a 0.1 / 5Ga probability of occurrence, therefore intelligent life is (again arbitrarily) only 10-5 likely to occur". He went the other way - coming up with 4 events based on the observed fact that from all evidence, terrestrial intelligence developed quite late in the habitable period of Earth.

The current thinking among scientists is that the Earth became habitable about 4.5 billion years ago, and has between a half and one and a half billion years more during which it will be habitable. This means habitable by any life-forms whatsoever, not "friendly for humans", so it's got nothing to do with global warming, or anything like that. It's more about changes in the Sun's life cycle.

So, this is all good speculative science. We take a model, line up the key points with other models, then ask some questions about what that implies. In this case, we lined up the "critical events" model of evolution with the models of the development of the biosphere, and asked, "what would be the most likely number of critical events?" The model says four. So we sanity check it - are there four events in the evolution of intelligent life on Earth that could fit the model? Sure - the development of prokaryotes, the development of eukaryotes, the development of cell-differentiation into organs, and the development of language are good candidates, fitting the model.

And it is science, not religion. Watson isn't saying, this is the One True Model of Evolutionary Development, and thou shalt believe it or be considered stupid. He's pointed out where every idea in the paper came from - even telling you what papers you can read to see for yourself. You can see what's assumption, and which of his conclusions follow from them.

It's the (evidently non-scientific) journalist who wrote the cover article that seems to have jumped to the conclusion that Watson is saying the development of intelligent life is unlikely. If you accept the assumptions outlined in the paper (they're not crazy assumptions, but neither are they indisputable) then the likelihood of intelligent life on any given Earth-like planet (with a habitable period of about 5Ga) is fairly small - less than one in ten-thousand such planets are expected to actually have life.

But if there are, say, a hundred thousand Earth-like planets in our galaxy, then we'd expect that about ten of them actually have life. And that's just throwing out a number - the galaxy's a big place, and there could be millions of Earth-like planets out there, I've no real idea. And the galaxy's just a tiny corner of a much bigger universe. Watson's numbers don't indicate that intelligent life is unlikely - quite the opposite.

Which takes us to Fermi's Paradox - if there are hundreds, or thousands, of intelligent life forms out there, why don't we see evidence of them?

Our own intelligence only started making the sort of signals that programs like SETI have been looking for about a century ago. So we've been emitting a SETI-recognizable signal for about that long. The only civilizations that could have observed it are ones within 100 light-years of Earth. If you imagine a map of the galaxy, where each civilization "lights up" at the point where they start emitting SETI-recognizable signals, they'd have a bubble expanding around them at one light-year per year. Our first opportunity to detect them would be when that bubble reached us.

We're also pretty far out on one of the limbs. If these civilizations are spread fairly uniformly through the volume of the galaxy, and if there are only a few hundred of them, then it's not at all surprising we haven't heard from them.

So, I take back my criticism of Watson. Interesting work. I very strongly criticize the attempt at sensationalizing it by the journalists. People can be stupid enough on these topics without your making them worse with misinformation. This wasn't Watson trying to give an answer to Fermi. This was a scientist discussing an idea with other scientists.