Tuesday, December 4, 2018

Fermi or DIE! More studies on Fermi paradox.

In this post we review some new (after 2010 at least) solutions to the Fermi paradox, or "the Great Silence".

Aestivation hypothesis

Proposed in That is not dead which can eternal lie: the aestivation hypothesis for resolving Fermi’s paradox (2017), by Anders Sandberg, Stuart Armstrong, Milan Ćirković. 

[Aestivate: to sleep through the summer, or a period too hot for living. The opposite of "hibernate".]

Idea: Aliens are sleeping until the far future, when the universe is cold enough, to maximize computation. The reason is that cold = more efficient use of energy for computation, according to the Landaur limit: $E\ge kT\ln(2) \text{Joule}$ need to be dissipated for an irreversible change of one bit of information.
... this can produce a $10^{30}$ multiplier of achievable computation. We hence suggest the “aestivation hypothesis”: the reason we are not observing manifestations of alien civilizations is that they are currently aestivating...
The phrase "That is not dead which can eternal lie" comes from a poem by H. P. Lovecraft, a philosopher ahead of his time who really understood how human thoughts matter little in a cosmic viewpoint.

A civilization would aestivate if:
  • The civilization goal is fundamentally about manipulation of information.
  • There's some advanced technology that allows computing efficiently (close to the Landaur Limit) at very low temperatures that the universe will reach in the far future.
  • Its goals are so big that the local computational resources it can use right now are not enough, and it is reasonably secure in its safety so that it can go to sleep without worrying getting attacked.
This doesn't seem to be a great solution to the Fermi Paradox, since it's unlikely how all advanced civilizations would choose to aestivate.

However, there are possible tests for a weakened form of aestivation hypothesis: "There exists aestivating civilizations in the observable universe".

If we do discover signs of aestivation, it would not only be a great discovery, it would also be a kind of alien wisdom imparted to us: the Old Ones chose to go to sleep, maybe we should too?

Deadly probes scenario

Idea: Some civilization released self-replicating probes that always destroy civilizations that they come across.
In order to work as a Fermi explanation the replicators must fulfill four conditions:
1. Cause the great silence.
2. Be compatible with our existence.
3. Be silent enough in their activities not to be visible.
4. Be impossible to overthrow by an new civilization.
Condition 3 is so that we don't observe the probes. Condition 4 is so that the probes actually do their job of killing. Condition 3 and 4 are in conflict, because strong probes tend to fight and stir shit up. There are 2 possible ways for the probes to be strong and remain in the game.

  1. Kill all civilizations it can get to before they launch their probes. This requires the probes to be everywhere so that it can monitor all the stars, and if they are so densely populated, they need to be very stealthy.
  2. Be in stable equilibrium with probes from another civilization. Some simulations show that "high and low density [of probes] cases favour each side manufacturing more probes, causing more attacks and resource use." That is, TOTAL WAR for the probes, which is very visible from their energy use.
Condition 2 is because, well, we have existed for quite some time and haven't been killed yet. Condition 1 is what a solution to the Fermi paradox is. Conditions 1 and 2 are in conflict, because strong probes would have killed us too, and weak probes would have not killed enough to cause the Great Silence.

The solar system is not special in the galaxy, so if the probes are good at killing they'd have killed us. If the probes are trying to be stealthy by lying in wait until a civilization is close to launching probes, then, since humans have been emitting strongly patterned radio waves into space since ~100 years ago (no need to get more accurate, since we are doing a Fermi estimate...), and it's entirely within our ability to launch a von Neumann probe within this century, so that gives about 200 years of slack between "rising civilization detected" and "too late to stop it". For the probe to do its job, it should have probably killed us in the first 100 years, just to be safe.

Conclusion: They might have contributed, but deadly probes are not the main cause of the Fermi paradox.

Wasteful colonizers

Burning the Cosmic Commons: Evolutionary Strategies for Interstellar Colonization (1998), by Robin Hanson. It uses economic analysis to propose that space colonization frontiers would be dominated by fast and wasteful colonizers who try to be fast at all costs.

The result would be that they colonize rapidly and leaving behind sparse oasis separated by empty spaces, with no regard for future use. Civilizations arising in these oases would find it hard to spread around.

This explains the title "Burning the cosmic commons", because it's a form of "tragedy of the commons". In a rush to use as much commons (here in the form of cosmic resources), the colonizers selfishly favor speed over efficiency.

In the following article (Eternity in 6 hours), there's a critique of this theory:
Robin Hanson has argued that a spreading interstellar species will tend to (culturally) evolve towards a tendency for rapid expansion with no concern for leaving useful resources for future use. However, his model (a staple of science fiction, based strongly on human modes of exploration on Earth) presupposes short jumps between colonies, leading to many generations of colonies and hence making them more susceptible to drift towards this attractor state...
Basically, if colonization is not step-by-step, settle-and-explore, but one-shot, then it could be centrally controlled, avoiding the tragedy of the commons.

Eternity in 6 hours


[Turns out the title is a reference to Jesus Christ, the deity of Catholicism who spent 6 hours dying before coming back to life. These 6 hours were considered so severe by the Catholicists that they call it an "eternity".]

Idea: not a solution, but more questions. Turns out intergalactic colonization is not hard, and even humanity could do it in the foreseeable future. Using fixed launch systems, such as blasting out probes that have just enough fuel to decelerate, using lasers stations based in the solar system, aka photonic propulsion, it would cost only six hours of the sun’s energy to send out all the probes needed to colonize the entire visible universe.


It's utterly trivial for a future humanity to do it.
Picture a future ‘President of the Solar System’ proclaiming: “Everyone turn off their virtual reality sets for six hours, we’re colonising the universe!”... If humanity were to survive for, say, a million years, and we were to spread over a few star systems, and spend nearly all available energy for our own amusement, we could still colonize the universe, many times over.
This of course makes the Fermi paradox worse: why haven't we been colonized yet, if a billion galaxies could have sent out their probes to reach us already?

Not only is it easy to colonize, it's even easy to do so invisibly.
... the alien probes could then discreetly rush every star in the galaxy... Indeed, discreetly grabbing the cosmos would only be slightly harder than blatantly doing so.

Transcension Hypothesis

The Transcension Hypothesis (2012), by John Smart.

Idea: There is a universal law for complex systems, called the evo-devo: evolution-development. Evolution is creative, makes more varieties, grows exponentially, like a tree. Development is orderly, makes less varieties, trims the tree of evolution, creates convergent designs, like a funnel.

Children spend lots of time dealing with physical world, but adults spend lots of time in their head. Maybe the same is for civilizations. As civilizations get older, they increasingly turn to thinking in inner space, and eventually gets completely absorbed into a ball of maximal inner complexity that on the outside looks just like a black hole.

Maybe civilization is an evo-devo process. Civilizations start out with a big burst of random creativity, with lots of different kinds of history, but they would all end up with the same conclusion: disappear into the inner world!

Advanced civilizations would not want to contact lesser civilizations because it would cause the lesser civilization to evolve like the advanced civilization, thus making the universe a less diverse place. 

This assumes that all advanced civilizations want to increase universe's diversity, which I find very suspicious. In general, any claim to universal morality, that is not simultaneously a universal survival trick, is suspicious. Also, there are always some adults that are intimately in contact with physical world (think of hunter-gathers). The mere existence of just a few such mature civilizations with contact with physical world would ruin the solution.

We really are alone

Okay this isn't a new one at all, but there's a recent paper Dissolving the Fermi Paradox (2018), by Anders Sandberg, Eric Drexler and Toby Ord, that argues that it's quite likely that we are alone in the universe.

It basically runs a Bayesian argument on the Drake equation. It first shows that the prior probability on us being alone is pretty nonnegligible.
... proper treatment of scientific uncertainties dissolves the Fermi paradox by showing that it is not at all unlikely ex ante for us to be alone... 
Then, after updating, the probability became substantial.
... taking account of observational bounds on the prevalence of other civilizations, our updated probabilities suggest that there is a substantial probability that we are alone. 
Conclusion: We are quite likely to be alone, and probably won't die from the Great Filter.
... pessimism for the survival of humanity based on the Fermi paradox is unfounded.

No comments:

Post a Comment

Let's Read: Neuropath (Bakker, 2009)

Neuropath  (Bakker 2009) is a dramatic demonstration of the eliminative materialism worldview of the author R. Scott Bakker. It's very b...