[ExI] Drake Equation Musings

[Anders Sandberg]

On 2016-05-17 20:02, BillK wrote:
> On 17 May 2016 at 11:59, Anders Sandberg wrote:
>> Remember the Carter argument: the probability of life emerging may be very
>> low, and there could be secondary low-probability steps between it and
>> intelligence. So most planets never get it, and among those that get it the
>> secondary steps never happen. The planets with observers on them will be the
>> ones that have the unlikely combination of getting all the steps, and if the
>> natural rate of them happening is slow, we should expect them to be about
>> equidistant across the time interval the planet is habitable. So early life
>> is not evidence of easy life unless there is no reason to think there are
>> any hard steps. However, we do have some evidence for hard steps (e.g. at
>> least one transition in genetic coding, which is something we know tends to
>> be stable over 10^80 cell divisions).
>>
>> In short, observer selection bias makes data from Earth suspect in updating
>> our probabilities.
>>
>> Note that your idea is nicely testable: if we do find life on Mars, Europa
>> or Ceres, then we have reason to think life is indeed common and not a great
>> filter. That is bad news for our future, though: while the probability of
>> intelligence evolving factor decreases the most given this, the expected
>> lifespan of civilization decreases second most.
>>
> You seem to be depending on the 'Rare Earth' claim to validate
> ignoring the evidence of life everywhere on Earth, even extremophiles
> which live in environments that would destroy most earth life. I doubt
> whether it is valid to use observer selection bias to ignore evidence
> unless you first refute the arguments against the 'Rare Earth'
> hypothesis.
Huh? How do you impute that assumption in my text, and how do you get 
that paragraph to work logically?

I am not using a rare earth argument. Maybe I was oversimplifying my 
explanation of the Carter argument ( https://www.jstor.org/stable/37419 
): it allows for life being super-easy too. Extremophiles does not prove 
life is everywhere, since they merely show that once you have life, then 
it can spread into available niches.

Think of it like this: what evidence would present itself to you in a 
universe where life was very easy, or one where it was very rare? In 
both cases you would see a planet with a biosphere compatible with you. 
You would find life adapted to a broad range of conditions. You would 
see life showing up early (either because it was simply likely, or 
because observers require a few rare steps). None of these observations 
give you any information allowing you to choose between the two universes.

Now, if you had evidence that there wasn't lots of aliens around, that 
would favor the later conclusion - as well as the "life is easy, 
intelligence hard" and "intelligence doesn't last" conclusions (plus 
"weird" things like the zoo hypothesis). It weakens the "life is easy" 
conclusion a bit, but since the uncertainty of intelligence and future 
longevity parameters are big enough to soften the change in probabilities.

Panspermias may be possible, but just create correlated patches with 
high life probabilities:
http://aleph.se/andart2/space/the-drake-equation-and-correlations/