Back before he went into politics and spawned Mecha Hitler, Elon Musk advanced the idea that we exist within a simulation. It was even claimed that was allegedly funding efforts to escape this simulation. This is, of course, just another chapter in the ancient philosophical problem of the external world. Put briefly, this problem is the challenge of proving that what seems to be a real external world is, in fact, a real external world. This is a problem in epistemology (the study of knowledge).
The problem is often presented in the context of metaphysical dualism. This is the view that reality is composed of two fundamental categories of stuff: mental stuff and physical stuff. The mental stuff is supposed to be what the soul or mind is composed of, while things like tables and kiwis (the fruit and the bird) are supposed to be composed of physical stuff. Using the example of a fire that I seem to be experiencing, the problem would be trying to prove that the idea of the fire in my mind is being caused by a physical fire in the external world.
Renee Descartes has probably the best-known version of this problem. In his Meditations he proposes that he is being deceived by an evil demon that creates, in his mind, an entire fictional world. His solution to this problem was to doubt until he reached something he could not doubt: his own existence. From this, he inferred the existence of God and then, over the rest of the Meditations on First Philosophy, he established that God was not a deceiver. Going back to the fire example, if I seem to see a fire, then there probably is an external, physical fire causing that idea. Descartes did not, obviously, decisively solve the problem. Otherwise, Musk and his fellows would be easily refuted by using Descartes’ argument.
One often overlooked contribution Descartes made to the problem of the external world is consideration of why the deception is taking place. Descartes attributes the deception of the demon to malice because it is an evil demon (or evil genius). In contrast, God’s goodness entails He is not a deceiver. In the case of Musk’s simulation, there is an obvious question of the motivation behind it. Is it malicious (like Descartes’ demon) or more benign? On the face of it, such deceit does seem morally problematic but perhaps the simulators have excellent moral reasons for this deceit. Descartes’s evil demon does provide the best classic version of Musk’s simulation idea since it involves an imposed deception. More on this later.
John Locke took a more pragmatic approach to the problem. He rejected the possibility of certainty and instead argued that what matters is understanding the world enough to avoid pain and achieve pleasure. Going back to the fire, Locke would say that he could not be sure that the fire was really an external, physical entity. But he has found that being in what appears to be fire has consistently resulted in pain and hence he understands enough to avoid standing in fire (whether it is real or not). This invites an obvious comparison to video games: when playing a game like World of Warcraft, the fire is not real. But, because having your character fake die in fake fire results in real annoyance, it does not really matter that the fire is not real. The game is, in terms of enjoyment, best played as if it is.
Locke does provide the basis of a response to worries about being in a simulation, namely that it would not matter if we were or were not. From the standpoint of happiness and misery, it would make no difference if the causes of pain and pleasure were real or simulated. Locke, however, does not consider that we might be within a simulation run by others. If it were determined that we are victims of a deceit, then this would presumably matter, especially if the deceit were malicious.
George Berkeley, unlike Locke and Descartes, explicitly and passionately rejected the existence of matter. He considered it a gateway drug to atheism. Instead, he embraces what is called “idealism”, “immaterialism” and “phenomenalism.” His view was that reality is composed of metaphysical immaterial minds and these minds have ideas. As such, for him there is no external physical reality because there is nothing physical. He does, however, need to distinguish between real things and hallucinations or dreams. His approach was to claim that real things are more vivid than hallucinations and dreams. Going back to the example of fire, a real fire for him would not be a physical fire composed of matter and energy. Rather, it would be a vivid idea of fire. For Berkeley, the classic problem of the external world is sidestepped by his rejection of the external world. However, it is interesting to speculate how a simulation would be handled by Berkeley’s view.
Since Berkeley does not accept the existence of matter, the real world outside the simulation would not be a material world as it would a world composed of minds. A possible basis for the difference is that the simulated world is less vivid than the real world (to use his distinction between hallucinations and reality). On this view, we would be minds trapped in a forced dream or hallucination. We would be denied the more vivid experiences of minds “outside” the simulation, but we would not be denied an external world in the metaphysical sense. To use an analogy, we would be watching VHS tapes, while the minds “outside” the simulation would be watching 8K video.
While Musk did not present a developed philosophical theory, his discussion indicates he thought we could be in a virtual reality style simulation. On this view, the external world would presumably be a physical world of some sort. This distinction is not a metaphysical one. Presumably the simulation is being run on physical hardware and we are some sort of virtual entities in the program. Our error, then, would be to think that our experiences correspond to material entities when they, in fact, merely correspond to virtual entities. Or perhaps we are in a Matrix style situation and we do have material bodies but receive virtual sensory input that does not correspond to the physical world.
Musk’s discussion seemed to indicate that he thought there was a purpose behind the simulation and that it has been constructed by others. He did not envision a Cartesian demon, but presumably envisioned beings like what we think we are. If they are supposed to be like us (or we like them, since we are supposed to be their creation), then speculation about their motives would be based on why we might do such a thing.
There are, of course, many reasons why we would create such a simulation. One reason could be scientific research: we already create simulations to help us understand and predict what we think is the real world. Perhaps we are in a simulation used for this purpose. Another reason would be for entertainment. We created games and simulated worlds to play in and watch; perhaps we are non-player characters in a game world or unwitting actors in a long running virtual reality show.
One idea, which was explored in Frederik Pohl’s short story “The Tunnel under the World”, is that our virtual world exists to test advertising and marketing techniques for the real world. In Pohl’s story, the inhabitants of Tylerton are killed in the explosion of the town’s chemical plant and are duplicated as tiny robots inhabiting a miniature reconstruction of the town. Each day for the inhabitants is June 15th and they wake up with their memories erased, ready to be subject to the advertising techniques to be tested that day. The results of the methods are analyzed, the inhabitants’ memories are wiped, and it all starts up again the next day.
While this tale is science fiction, Google and Facebook are working very hard to collect as much data as they can about us with an end to monetize it. While the technology (probably) does not yet exist to duplicate us within a computer simulation, that would be a logical goal of this data collection. Imagine the monetary value of being able to simulate and predict people’s behavior at the individual level. To be effective, a simulation owned by one company would need to model the influences of its competitors, so we might be simulations in a Google World or a Facebook World so that these companies can study us to exploit the real versions of us in the external world.
Given that a simulated world is likely to exist to exploit the inhabitants, it certainly makes sense to not only want to know if we are in such a world, but also to try to undertake an escape. This will be the subject of the next essay.