What the Universe Wants: Anthropics from the POV of Self-Replication

post by moridinamael · 2018-01-12T19:03:34.044Z · LW · GW · 2 comments

Meta: Once again I'm trying to submit this to my "personal page" and not the front page, but I really have no idea who will actually see it.

Here's a pretty simple idea I hadn't heard yet anywhere in science or science fiction. It's likely I'm not the first to think of this. It's also likely that it's nonsense, but I think it's fun nonsense. Please enjoy this in the spirit of the wild-ass speculations straddling the border between sci-fi and futurism with which Less Wrong used to be filled.

The rules of the universe are in some sense objectively unlikely, but by the anthropic principle, we shouldn't be surprised to observe them. A universe that doesn't support observers doesn't get observed.

We happen to be animals capable of creating new intelligent life by default, through a fundamental drive to replicate our genes, and a physics that supports that process. We breed.

We should condition "anthropically" on this evidence and update in favor of believing that intelligent observers are usually/commonly also "replicators" driven by natural selection propelled by lower levels of self-replication.

If you'll forgive my abuse of language, genes "want" to self-replicate, "causing" humans to "want" to self-replicate.

There is an argument that the universe may be more complicated than it strictly needs to be to support observers. If you could run a sentient observer in a Conway's Game of Life universe, and you include some kind of complexity-weighted distribution over universes, then shouldn't "most" universes and "most" observers exist in minimally complex universes?

Here is the step of the argument with the most uncertainty: There is a concept that each black hole creates a new universe.

Let's assume that black holes do indeed create new universes. If that is the case, "most" universes should be the ones just complex enough to support creation of new, similar universes. Natural selection in action, again.

However, that should put evolutionary pressure on generating universes just complex enough to create such singularities. Observers are not a requisite.

Under the above assumptions, we're in a universe that's complex enough to support both intelligent, self-replicating observers, and a potential mechanism for self-replication of the universe itself.

It strikes me as elegant and suggestive that natural selection could be operating at both scales because -- again, sorry to abuse language -- because the universe "wants" to create inhabitants that are both intelligent and driven to replicate themselves, so that those inhabitants will eventually become strong enough to hijack the process of creating a gravitational singularity and intentionally building new universes like this one.

(NB: I am not saying the universe is intelligent. I am saying natural selection has put pressure on the universe to look a certain way. I mean "want" in the same sense that we say natural selection "wants" to create a survival machine, or a single-celled protist "wants" to eat, grow and divide.)

So, by analogy: the universe is the organism; the black holes are the unfertilized eggs; the pockets of self-replicating sapience are the sperm destined to fertilize them and continue the cycle.

If you extrapolate from these assumptions, there are some potential consequences:

1. The universe wants enough pockets of intelligent sapience to gaurantee its continuity, but will avoid spawning thousands of civilizations within a range that they can wipe each other out. Thus, no surprise that we seem to be alone in our Hubble volume. Fermi Paradox solved.
2. The universe wants its inhabitants to want to fulfill its cosmic drive to spawn new similar universes. We infer can, by the Omohundro drives, that all replicators, and all other agents seeking to maximize anything, will want to create new universes, if such a thing is possible. Especially if it turns out that agents within the universe can influence the dynamics of the new universes in ways that appeal to us. Thus, we probably can influence the dynamics of new universes.
3. If we assume ourselves to be in a universe created by the process described above, that does imply that our creators did not just "move in" to the universe they created, because if they had, I feel like things would look different.
4. Unless they're waiting until a certain celestial epoch before they move in. The harvest isn't ripe yet. Perhaps our godparents want a cold, flat universe that supports optimal computation. And this make sense evolutionarily, too -- the universe hasn't "spawned" yet, we haven't reached our ascension and started creating our own universes. The life cycle would be interrupted if Cthulu invaded now.
5. I can't really determine if this increases or decreases the likelihood of malevolent AGI. Does the universe "want" paperclippers? Paperclippers at least want to build new universes that support paperclips. But complex sentient life also wants to create space for itself and its children. I lean slightly in favor of the universe being optimized for paperclippers, because paperclippers will tend to want to "clone" the universe to preserve the exact specifications of their paperclips, while intelligent life may optimize its child universes in all kinds of ways that don't preserve what our universe "wants" evolutionarily.
6. Therefor, if we don't want to be selected against, we should appease our dark master Natural Selection and spawn a bunch of clones of this universe in addition to whatever else we want to accomplish.
7. Using these assumptions to solve Theodicy and the Hard Problem of Consciousness are left as an exercise to the reader.

2 comments

Comments sorted by top scores.