Making a computer simulation therefore being in a computer simulation?

Question

Suppose we were to make a perfect computer simulation of our own world (granted, the Heisenberg uncertainty principle prevents this so consider it to be mostly, and not exactly, like our world). The inhabitants are so like us that they do the exact same thing: they, too, make a simulation of their own world thus creating a plethora/infinity of simulated worlds within simulated worlds (Thus they create a simulation, which creates a new simulation and so on). We would have no way of knowing which is real and which is not.

The sheer number of simulations within simulations means there is an extremely high chance we are in a simulated world. Therefore, refraining from creating simulated worlds means we don't open up a plethora of simulated worlds within simulated worlds. Therefore, our odds of being in a simulated world are heavily decreased because the probability of us being in the real world is (our world{1} / no simulated worlds+our own {1+0}).

Of course this implies that if we were to create a simulation of ourselves, turning it off would likely mean that our simulation controller would turn *us *off too therefore deleting our universe.

My question: Are there any flaws in the above reasoning? If not, that's a valid answer too by the way- do tell me if this breaks the rules, I can't say I'm sure.

Answer 1

Note: This answer was written in response to an earlier version of the question, and is not intended to address the version that is current as of 1/28/18

The line of reasoning you're following is essentially the same as Professor Nick Bostrom's notorious "Simulation Theory," and while it's implausible to many, it has successfully convinced people from Elon Musk to the Bank of America.

If we do become able to create fully realistic, immersive simulations, then it will indeed become much harder to justify our belief in the primacy of our own reality, to the same extent that the simulation is convincing. As of now, we still live in a world where there might be some as yet undiscovered reason that we cannot create simulations on the order necessary to duplicate our own world. That wouldn't entirely rule out our being some other reality's simulation, but it would weaken the argument that it's not only possible but likely.

Answer 2

People in the fields of history, archaeology and other human studies frequently need to judge between different theories in order to proceed with their work. Some metrics that distinguish a good theory from a bad theory are:

• Is it plausible? (I.e. could it happen?)
• Is it less ad-hoc? (I.e. is it free of extra complications?)
• Does it have the greatest explanatory scope? (I.e. does it explain many things?)
• Does it have the greatest explanatory power? (I.e. does it explain things well?)

The situation you described results in two competing theories: first, that our being a simulation is more likely (much more likely) than our not being a simulation, resulting in strange risks such as that we are in danger of disappearing if we unplug computers. The second theory is the ordinary theory, that simulations are simulations and reality is reality and ne'er the twain shall meet.

I think that you have mistaken probability for a good argument in favor of the simulation-simulation theory. Probability is only relevant for theories that are also plausible, and less ad-hoc... and the simulation-simulation theory isn't really plausible (because of the inversion of cause and effect) and it's more ad-hoc than the ordinary theory.

Answer 3

You cannot grant that Heisenberg's principle makes this impossible, and Heisenberg's principle is an observed fact of our reality. But this is a perfect simulation of our reality, so it has to model that principle. Yet it would have to be immune to that principle in order to exist. Admitting your premises cannot be met, yet proceeding, is a flaw in an argument.

If we created a perfect simulation of our world, given that our world does in fact have quantum dynamics, that randomness would have to apply to it, too. So the two would immediately diverge in many tiny ways, that would quickly accumulate into macroscopic differences.

For one thing, this means that what our simulated world does has no controlling effect on what the world around it does. The simulated world could choose to avoid making nested simulations, even though we created it, and it shared a state with some point in the past.

So only a bizarre amplification of determinism that ignores not only all existing notions of free will, but also a large part of modern physics could make this argument hold water. Whether or not we choose to make simulations cannot determine whether we are in one.

Even the limited degree that quantum theory allows the future to predict the past does not apply here, because the simulation itself would have to be quantum in nature and totally unobserved. But a simulation runs on some piece of equipment which continually makes decisions based on state. If this machine is therefore macroscopic and observable the simulation is endlessly observed. So this reversal of implication could only happen if this piece of equipment were some kind of unobservable quantum matrix. But then this would not be a simulation, because its state would not be determined. Something far more uncertain than reality cannot also be like reality in a meaningful way.

That means that either we are already in a simulation or we are not, whether or not we decide to multiply simulations or turn them on and off.

Answer 4

Some points:

1)

The sheer number of simulations within simulations means there is an extremely high chance we are in a simulated world.

That would be true if we knew the total number of possible worlds for us to live in and if each possibility was equally likely or if the likelihood of each possibility was known. But we don't know that.

2)

Suppose we were to make a perfect computer simulation of our own world .

That would be impossible, it takes more than one atom to simulate an atom. There wouldnt be enough atoms in our universe to build the simulator. The simulation must be a simplified version of reality, therefore unperfect.

Quantum mechanics, by preventing us from studying the infinitely small, make our universe look like it's already a simplification and perhaps this plays a role in the simulation hypothesis popularity.

3) we dont need a simulation to be automatons (robots), by all scientific accounts we are determined by nature and nurture.

4) As already mentionned by @christo183 in the question comments, how would you know if we werent in a simulation ? We wouldnt. This hypothesis can't be falsified and therefore outside of scientific scope. It is the same as "does god exist".

5)Creationism for tech people:

The simulation hypothesis, that we live in a simulated world, is just a rehash of old arguments about god aka the creator. Looking for proofs of a simulation is EXACTLY like looking for a creator, aka a god. Because a simulation is a creation that needs a creator.

Where religions were making some kind of magic family in heaven (god/mary/jesus) at the image of family on earth, the simulation hypothesis, gathering a lot of buzz after Elon Musk mentionned it, needs magic engineers in the sky. It's exactly the same fallacy Voltaire outlined: “If God has made us in his image, we have returned him the favor.” .

All the supposed things that could be evidence of a simulation fall under the same fallacies debunked long ago about creationism and intelligent design.