I've been thinking for a while about Nick Bostrom's simulation argument.
Bostrom asserts that one of the following three options must be true:
(1) the human species is very likely to go extinct before reaching a “posthuman” stage;
(2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof);
(3) we are almost certainly living in a computer simulation.
How does this work? Well, if simulations of conscious being are possible, then Bostrom argues that most minds in existence across time are simulated, and therefore we are much more likely to be simulated than real.
Ouch!
Bostrom's argument is about the probability of one of these things being true - specifically number 3 - that we are in a simulation run by our distant post-human descendants. In order to refute it totally, we'd need a reason to say that what he is proposing isn't logically possible - and I'm not sure how to do that. I do however think there is a problem with what he is trying to show.
Any one of those options he puts forward are possibly true. If I remember correctly, he doesn't state it clearly in the original paper, but he does say in this video that us reaching post-human level, and being interested in running ancestor simulations increases the probability of (3). I can't quite put my finger on it, but I can't help but think that something is seriously amiss with this line of reasoning. Why would the fact that we did something make it more likely that someone else (or maybe posthuman version of ourselves) has done it already?
If we were in a simulation, then why would we assume that the chances of reaching post-humanity are the same within the simulation as in the real world? It would be like a character in Simcity, Civilizations, or Age of Empires speculating about history in our real world - they would almost certainly get it massively and spectacularly wrong.
In my mind, this indicates the following: The higher the probability of (3), that we are in a post-human simulation, the lower the probability is that we can accurately assign probabilities to (1) and (2). I think this indicates a problem with the argument. Unless there's some reason that simulated humans can accurately say what history in the 'real' world is actually like, Bostrom is back to odds of 1/3 for each.
This pretty much kills the simulation argument. But I'm not satisfied with making it seem weak - I'd like to crush it entirely (and will proceed to do so).
How do we know that those possibilities that Bostrom suggests are the only three options? Assuming one thing is simpler than assuming three - so he can't rely on simplicity to stop me from proposing further options.
Here's Bostrom's three:
(1) the human species is very likely to go extinct before reaching a “posthuman” stage;
(2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof);
(3) we are almost certainly living in a computer simulation.
Now, let's add some more:
(4) we are part of an infinitely branched multiverse, where every collapsing waveform, personal choice or random event gives birth to a new world-line;
(5) we inhabit a multiverse where each universe is a fictional narrative in another universe;
(6) we are part of an organic computer run by mice, aimed providing the ultimate question, soon to be demolished to make way for a hyperspace bypass;
(7) we are dreams in the mind of a sleeping god
(8) etc.;
.....
(X)...
If we have X possibilities, and we assign the same probability to each option, the bigger X gets, the smaller each individual chance must get. As X gets closer to being infinitely large, the probability of each option becomes infinitely small.
If we can't rule out some possibilities, and we can't assign probabilities in any way other than evenly divided between all options, then all options are equally highly improbable.
So there it is: Stop worrying about being in a computer simulation. Start worrying about being a character in someone else's novel.
Thanks for reading. Comments, upvotes and resteems are awesome, and I appreciate them deeply.
Photo by Markus Spiske on Unsplash
