By Taylor Marvin
Jason T. Wright has a fascinating series of posts discussing the energy and waste heat constraints facing extraterrestrial civilizations, and in his most recent post argues that expansionary alien civilizations are likely long lived. This has interesting implications:
“I am arguing that once a civilization gets going, it’s going to take over the whole galaxy quickly, and that L (the lifetime of a typical civilization) is actually longer than the current age of the Universe. If this means that subsequent civilizations are unlikely to arise, then N= 0 or 1 for most galaxies (0, in fact, since most galaxies don’t look like they’re full of Dyson spheres).”
This is an interesting challenge to the classically understood implications of the Fermi Paradox and explanations for the Great Silence. Even if we accept the Rare Earth argument that the conditions required for the evolution of Earth-like complex life are rare, the sheer number of rocky planets in our galaxy suggests that even intelligent civilizations based on Earth-like biology should be relatively common. This presumed frequency is typically squared with our failure to observe evidence of extraterrestrial civilizations by noting that even if these civilizations are common, they are unlikely to coincide with us in space and time. However, if the first civilization to begin expansion is likely to dominate the galaxy fairly quickly, then this explanation doesn’t hold. As Wright notes, if an intelligent expansionistic civilization had arisen in the past it would be long lived, and we would observe it today. The fact that we don’t is clear evidence that an alien civilization has never begun expanding in our galaxy — or, for that matter, in any other close enough for us to observe a lack of Dyson spheres at our observed time. If we haven’t observed an alien civilization in the local group, it’s likely because there have never been any there.
There are a few ways around this observation. Speculatively, perhaps interstellar civilizations do arise frequently, expand, but quickly evolve into a state that renders them undetectable. Science fiction plays around with this conjecture: Kardashev type III civilizations could exist, but just in some way we’re incapable of recognizing, or advanced civilizations could choose to expand in virtual reality rather than the outside universe.
Another possibility is the classic answer to the Drake Equation: perhaps intelligent civilizations are common, but they universally fail to expand beyond their home system and are short lived. This state is possible even if civilizations do not destroy themselves before expanding. As I have previously discussed, there are reasons to believe that low-birthrate, energy constrained, individually rational species (in short, a species like us) would fail to expand even if it is in their long-term interest. However, this explanation is obviously problematic. Even if a species’ innate characteristics discourage expansion, AIs derived from this particular civilization would not share the same traits: it is easy to imagine a universe where biological species fail to expand, but AI entities descended from their computers do not. Similarly, even if many species destroy themselves before they can expand beyond their home systems, or elect not to, it’s unreasonable to suppose that this tendency is universal. If Wright’s logical conclusion that expansionary civilizations’ L is long and leads to an N=1 galactic outcome, we would only have to be predated by a single expansionary civilization for us to observe alien intelligence. Given that our galaxy has likely been potentially habitable for over ten billion years, this implies that civilizations are extremely rare.
Even more speculatively, another possibility is that civilizations are common, but universally elect not to expand. Because alien civilizations’ behavior would be governed by vastly different biological and economic constraints, this decision would have to be motivated by outside influence. Specifically, alien civilizations, like us, fail to observe evidence of aliens. Civilizations then face a choice: they can either expand, and enjoy the massive first mover advantage that leads to the N=1 outcome, or not expand. Why not? One possibility is that alien civilizations do exist, but are actively hiding. Alien civilizations may reason that the absence of observable aliens is evidence that the universe is extremely dangerous, and other civilizations are either hiding or extinct. This allows for the original Drake equation’s assumption, in Wright’s words, of a “steady-state of short-lived civilizations” that never move from a Kardashev type II to III. Instead they elect not to expand, or only as much as remaining hidden allows. Importantly, it’s possible that this no first mover equilibrium exists whether or not there is actually a malevolent entity that makes the universe dangerous. Since no civilization can be sure that they’re the first intelligence in the galaxy, the possibility of frightened and hiding extraterrestrials is impossible to rule out.
If the universe is dangerous, then expansion — or being detectable, in general — is risky. Of course, not expanding is the ultimate “risky” choice — civilizations that don’t spread beyond their home star are doomed to die along with it. If humans fail to expand beyond Earth in significant numbers, our civilization will die at most a billion years from now. Given this time frame, the risk detection leads to immediate, complete destruction must be very high for civilizations to choose not to expand. For example, say that humans will be completely destroyed 42,000 years after being detected by whatever malevolent entity makes the galaxy dangerous (the Earth is 26,000 light years from the center of the galaxy; this assumes that a destructive force is dispatched at light speed as soon as evidence of humans reaches the center of the galaxy. I’m spitballing here). Given this assumption, the expected number of years human civilization survives is maximized by not expanding if the risk that the universe is dangerous is >.99996. Obviously, all alien species must be extremely risk averse to make a conscious choice not to expand out of fear, especially since the only evidence of a dangerous universe type is the absence of observable civilizations.
So not particularly convincing! But pretend this is the case, and the galaxy is relatively full of risk averse alien civilizations afraid to expand. This suggests an interesting strategic interaction. The first civilization to expand beyond their home star takes a bet that the universe is not dangerous. If they are wrong, they are exterminated. However, if they’re right their civilization enjoys the first mover advantage that leads to a N=1 galaxy, and survives essentially forever. Given the low likelihood that all civilizations in the galaxy are extremely risk averse, lack of evidence that the universe is a dangerous type, and massive first mover advantage, it is unlikely that universal hiding is a stable equilibrium.
Another problem is that colonization is not necessarily risky. As Robin Hanson has noted, “if this colonization effort could hide its origins from those who might retaliate, what would they have to lose?” For expansion to be risky, the hidden malevolent entity must be so all-powerful that it can arbitrarily destroy colonies separated by many light years and remain dangerous across deep time, but still fail to detect hidden civilizations — a narrow criteria! These requirements, and the likelihood that a galaxy full of hiding, non-expansionistic civilizations is not in stable equilibrium, suggests that the ‘aliens are hiding’ answer to the Fermi Paradox is not convincing.
Previous posts on aliens and human expansion: