Catastrophe Engines: A possible resolution to the Fermi Paradox

by snarles 3 min read25th Jul 201522 comments


The Fermi Paradox leads us to conclude that either A) intelligent life is extremely improbable, B) intelligent life very rarely grows to a higher-level civilization, or C) that higher-level civilizations are common, but are not easy to spot.  But each of these explanations are hard to believe.  It is hard to believe that intelligent life is rare, given that hominids evolved intelligence so quickly.  It is hard to believe that intelligence is inherently self-destructive, since as soon as an intelligent species gains the ability to colonize distant planets, it becomes increasingly unlikely that the entire species could be wiped out; meanwhile, it appears that our own species is on the verge of attaining this potential.  It is hard to believe C, since natural selection favors expansionism, so if even a tiny fraction of higher-level civilizations value expansion, then that civilization becomes extremely visible to observers due to its exponential rate of expansion.  Not to mention that our own system should have already been colonized by now.

Here I present a new explanation on why higher-level civilizations might be common, and yet still undetected.  The key assumption is the existence of a type of Matrioshka brain which I call a "Catastrophe Engine."  I cannot even speculate on the exotic physics which might give rise to such a design.  However, the defining characteristics of a Catastrophe Engine are as follows:

  1. The Catastrophe Engine is orders or magnitude more computationally powerful than any Matrioshka Brain possible by conventional physics.
  2. The Catastrophe Engine has a fixed probability 1-et of "meltdown" in any interval of t seconds.  In other words, the lifetime of a Catastrophe Engine is an exponentially distributed random variable with a mean lifetime of 1/λ seconds.
  3. When the Catastrophe Engine suffers a meltdown, it has a destructive effect of radius r, which, among other things, results in the destruction of all other Catastrophe Engines within the radius, and furthermore renders it permanently impossible to rebuild Engines within the radius.
A civilization using Catastrophe Engines would be incentivized to construct the Engines far apart from each other, hence explaining why such we have never detected such a civilization.  Some simple math shows why this would be the case.

Consider a large spherical volume of space.  A civilization places a number of Catastrophe Engines in the volume: suppose the Engines are placed in a density so that each Engine is within a radius r of n other such Engines.  The civilization seeks to maximize the total computational lifetime of the collection of Engines.

The probability that any given Engine will be destroyed by itself or its neighbors in any given interval of t seconds is 1-e-nλt.
The expected lifetime of an Engine is therefore T = 1/(n λ).
The total computational lifetime of the system is proportional to nT = n/(n λ) = 1/λ.

Hence, there is no incentive for the civilization to build Catastrophe Engines to a density n greater than 1.  If the civilization gains extra utility from long computational lifetimes, as we could easily imagine, then the civilization is in fact incentivized to keep the Catastrophe Engines from getting too close.

Now suppose the radius r is extremely huge, i.e. on the order of intergalatic distances.  Then the closest Catastrophe Engine is likely on the order of r distance from ourselves, and may be quite difficult to spot even if it is highly visible.

On the other hand, the larger the radius of destruction r, the more likely it is that we would be able to observe the effects of a meltdown given that it occurs within our visible universe.  But since a larger radius also implies a smaller number of Catastrophe Engines, a sufficiently large radius (and long expected lifetime) makes it more likely that a meltdown has simply not yet occurred in our visible universe.

The existence of Catastrophe Engines alone does not explain the Fermi Paradox.  We also have to rule out the possibility that a civilization with Catastrophe Engines will still litter the universe with visible artifacts, or that highly visible expansionist civilizations which have not yet developed Catastrophe Engines would coexist with invisible civilizations using Catastrophe Engines.  But there are many ways to fill in these gaps.  Catastrophe Engines might be so potent that a civilization ceases to bother with any other kinds of possibly visible projects other than construction of additional Catastrophe Engines.  Furthermore, it could be possible that civilizations using Catastrophe Engines actively neutralize other spacefaring civilizations, fearing disruption to the Catastrophe Engines.  Or that Catastrophe Engines are rapidly discovered: their principles become known to most civilizations before those civilizations have become highly visible.


The Catastrophe Engine is by no means a conservative explanation of the Fermi Paradox, since only the very most speculative principles of physics could possibly explain how an object of such destructive power could be constructed.  Nevertheless, it is one explanation of how higher civilizations might be hard to detect as a consequence of purely economical motivations.

Supposing this is a correct explanation of the Fermi paradox, does it result in a desirable outcome for the long-term future of the human race?  Perhaps not, since it necessarily implies the existence of a destructive technology that could damage a distant civilization.  Any civilization lying close enough to be affected by our civilization would be incentivized to neutralize us before we gain this technology.  On the other hand, if we could gain the technology before being detected, then mutually assured destruction could give us a bargaining chip, say, to be granted virtual tenancy in one of their Matrioshka Brains.