The idea is not new, but currently there is a bit of a buzz on the internet due to the article “No, we probably don’t live in a computer simulation” by Dr. Sabine Hossenfelder on the Backreaction blog. In it the author states the following:
The simulation hypothesis, as it’s called, enjoys a certain popularity among people who like to think of themselves as intellectual, believing it speaks for their mental flexibility. Unfortunately it primarily speaks for their lacking knowledge of physics.
Among physicists, the simulation hypothesis is not popular and that’s for a good reason – we know that it is difficult to find consistent explanations for our observations. After all, finding consistent explanations is what we get paid to do.
Proclaiming that “the programmer did it” doesn’t only not explain anything – it teleports us back to the age of mythology. The simulation hypothesis annoys me because it intrudes on the terrain of physicists. It’s a bold claim about the laws of nature that however doesn’t pay any attention to what we know about the laws of nature.
I suspect that part of the reason for the buzz is the author’s choice of language, but also because it is an interesting idea to think about. The whole concept of a simulated reality arguably traces back to the dream argument in philosophy. That has been around which has been around since Zhuang Zhou in the 4th century B.C. with the butterfly dream. Of course, essays like this tend to see responses like the one Gizmodo and “Think We’re Living in a Computer Simulation? Prove It.” To which my response is: can’t be done.
The biggest problem with arguments about a simulated reality is that ultimately they are unfalsifiable. This comes from Karl Popper and the epistemological theory of falsifiability and is discussed at length in “The Logic of Scientific Discovery.” In short, in order to properly test a theory, you need to be able to show that it can be false. The major problem with simulated reality is that you cannot develop an experiment that will prove (or disprove) that you are in a simulated reality! One argument is that an experiment could be developed that forces a glitch in the simulation. This would prove the simulation’s existence. The argument against is that if the behavior is repeatable, you may have only uncovered a new natural law. Unless you know what the rules of the simulation are a priori you have no way of knowing it’s a glitch. Unfortunately without Out of Place Artifacts like Star Ocean: Till the End of Time it is unlikely we will have a clear indicator of something odd either.
Another problem with the augments for or against is that they make assumptions about how the simulation would be conducted. One common argument is that you would need massive amount of computing power to run the simulation. This is incorrect. The problem is not how fast the simulation runs, but how much storage is needed for the simulation state. To a resident of a simulation, they have no way to perceive the passage of time outside the simulation. So even if it takes several thousand years of “real time” to compute one second of the simulation, residents of the simulation will perceive only one second as having gone by. This does present some problems since the the hypercomputer may fail due to entropy long before the simulation is “complete” but again, the residents of the simulation would only be aware of the entropy of their own universe.
If you take the time to read though the comments on the post by Dr. Hossenfelder a lot of the commenters make comments about the unfalsifiability of the problem although some commenters fall into the trap about computing speed. Incomplete information seems to also be a problem. I suspect that most of the commenters are not at Dr. Hossenfelder’s level with regards to physics and comments reflect this. Ultimately, as a scientific argument this is a Kobayashi Maru since it can’t be won. As a philosophical argument, it’s fun but still can’t be won.