Are We All Just Sims?
The simulation argument explored
Oh so many people from writers on Medium to Elon Musk are jumping on board the argument that we exist in a simulated reality. While the idea first occurred to me in middle school, I hadn’t ever put much thought into why it would or wouldn’t make sense, preferring instead to laugh at the absurdity that God could be a teenager like me that enjoyed playing video games.
If we are in a simulated reality, then our parent universe — the one in which the simulation is being run — must be more similar to our universe than different. Structures must be able to form and maintain themselves, as well as grow in complexity. Otherwise those structures must have always existed in a sufficiently complex state to create a simulation. There are many possible universes where nothing at all could really happen besides collisions that result in energy loss; the laws of physics in those universes being so different that nothing could ever form: not stars, not galaxies, not life, not simulations of other universes.
Expanding on what exactly “more similar to our universe than different” implies: processes to store information (memory) must exist, as our universe moves through states (which we perceive as time) overall from higher entropy to lower, as our universe is most likely an isolated system. The memory of the parent universe would store the information about the states of every bit of energy in our universe. There must also be an idea of energy similar to our own that allows the movement of objects and the changes of state in the simulation machine, otherwise the simulator would be static.
Back to energy. Simulating an entire universe would require more than that universe’s total energy to simulate. This implies that our parent universe has much more energy than our universe. While this is not an impossible claim, there is yet no reason to believe it one way or another. If we assume the parent universe can have much more energy than ours, it is necessary that the parent universe does because the simulator running our universe must maintain all the energy here all the time. The simulator likely could not just run sections of the universe at a time because the universe is very connected through transfers of energy; running a part of the universe, then another part, and sticking them together could be a very messy process of ensuring consistency throughout the universe. Maybe a species so advanced that it could run a simulation of an entire universe could figure out this consistency issue, but anything short of perfection would mean could possibly see an instance of inconsistency: the laws of physics not being applied, or being applied incorrectly, in some way.
It seems more likely that the simulating machine must hold all of this energy in memory, the way a computer or gaming console loads an entire map into memory, even if it is only rendering parts at a time. Our simulator would be much more complex than a game, because it would have to load the entire universe into memory so our brains could render whatever sensory information is bombarding us at the time. On top of this, the machine must exert extra energy to change states within itself, and if we assume energy loss like in our universe, that only means more energy must be used by this machine.
If we take the position that the machine does not exert extra energy to change states in our universe, then we must assume that it only uses the same amount of energy as already exists in our universe. This is because if the simulator uses less energy than exists in our universe, then it cannot simulate all the energy in our universe — an application of the pigeon hole principle in which the simulator’s finite energy must be recycled in one part of the universe to be used in another. If the simulator has more energy than the universe and does not use it towards simulating us, then that energy must be going towards other applications, wasted, or not used, because if it was put into our universe, our universe would not be an isolated system of energy. So if the simulator takes no extra energy to change states, the simulator must have exactly the energy of our universe. In this case, either we are at the “base” universe and our simulator is indistinguishable from our universe, or our parent universe has developed machines that are perfectly efficient to give us the isolated system we understand we are in. While the latter is possible with a speculative imagination, it is only speculation and should not be taken as any sort of definitive argument.
However, if we do not assume that the parent universe beings have developed perfectly efficient machines, then the simulator must use more energy than exists in our universe, implying that the parent universe must have much more energy than ours. Not only would it need the energy to run the simulator, it would need energy to create the beings that made the simulator, and the space to hold a still-accelerating universe. If this is the case, then we could never hope to run a simulator that simulates our entire universe. Our only possibility, given the laws of physics in our universe, is to make a smaller universe of the same laws, or a universe of different laws. Given our universe’s laws, we act as a bottleneck in the recursive algorithm of creating universes.
If we make a universe more simple than our own in some way, with definitely less energy, but still able to create life that is itself able to create another universe, eventually that recursion must end in a universe that has such a small amount of energy it cannot create a simulation of another universe. Most likely this final universe would also not be able to generate life, or much of anything, because it would have such little energy. In the second-to-last universe, if the intention was to create a universe with more life, or able to simulate another universe, it would require the same amount of energy that that second-to-last universe has already, because that universe is already at the minimum energy necessary to create a universe that can create a universe. If we assume that this second-to-last universe is an isolated system, in that the machine running it maintains a constant amount of energy for the simulation to use, then the second-to-last universe would have to use all of its energy to simulate another universe capable of simulating another universe. Which is impossible, because it would require energy to make the simulator. So we have arrived at a final step in the recursion. The finite amount of energy in our universe, then, implies that universes cannot be nested within each other in a ring topology. It must be a line with at least one end.
What if we create a universe that has perfect energy transfers like our hypothetical parent universe might? Nothing would change. While our child universe may develop machines that run more energy-efficiently than ours, we did not solve the problem of finite energy. We would at best only delay the end of the nested simulations.
The finite amount of energy in our universe, then, implies that universes cannot be nested within each other in a ring topology. It must be a line with at least one end.
Let’s look for the other end of that line of simulators. Our parent universe could have much, much more energy than ours, as I previously stated, like our universe would have much, much more energy than our child universe(s). But the parent can only have a finite or infinite amount of (useful) energy. A finite amount of energy would imply that our parent is an isolated system, and we are using some subset of energy from that isolated system in our own isolated system. Then our grandparent universe would have yet more energy, albeit finite, and we can take this logical exercise to its natural conclusions: some base universe, wholly containing ours, has an arbitrarily large amount of energy that can sustain at least one simulation of a universe. In this case, the line of our ancestral universes could extend forever, with each universe having a finite but very large amount of energy, until there is some starting point which has mysterious but unique origins with respect to every simulated universe within it. This argument just as easily implies that ours is the base universe with a mysterious yet unique beginning as much as any parent universe we may be in is.
On the other hand, some universe we are nested in may have infinite energy, that is, there will be no heat death for that universe. This would have to be due to some different laws of physics there or that it is not an isolated system, getting more useful energy from somewhere else. In either case, this universe could create infinite simulations of other universes, each with infinite or finite amounts of energy; only those with finite energy would have to end their recursion. This infinite energy universe could be nested in another infinite energy universe; if this is the case then we get the same conclusion as in the finite case. Otherwise this infinite energy universe could be a part of a set of infinite energy universes that form a ring or mesh topology where energy flows through all of these universes in some direction(s), with our universe in a line of universe(s) tangent to this ring/mesh, with energy only flowing into our universe once (since we are an isolated system).
Now to answer the title question: I don’t know. In writing this piece, it occurred to me that answering the question definitively is akin to asking whether or not there is a divine deity that created the universe. For all our human knowledge and speculation, we really don’t know enough to make a complete argument one way or another. At this point, I’ve concluded that thinking our universe is a simulation is no more than a belief.
