The other week, the blog Borealism published a brief interview with David Kyle Johnson (Kings College) on how our lives would change if we discovered we are living in a computer simulation. Dr. Johnson kindly mentioned some of my work on The Peer-to-Peer (P2P) Simulation Hypothesis (see here, here, and here), which led in turn to the blog owner to send me a few follow-up questions. Borealism published my answers in part here, but here they are in full (I hope some of you find them interesting!):
1. Your theory maintains anchored in the idea that time flows subjectively and that the physical universe is a timelessly existing array of information which our consciousness is able to perceive as it chooses to perceive. In other words, that each person’s consciousness can read the physical information, akin to the laser of a CD player reading the information on the compact disc and playing it back to the observer. In your observations and experiences, how does the quantum world of collapsing wave functions and the observer effect work to support this hypothesis?
Great question, and many thanks for taking an interest in my work and sharing it with your readers!
Let me begin by suggesting this video to any of your viewers who may be (understandably) unfamiliar with quantum mechanics. Quantum mechanics is an incredibly well-confirmed theory of fundamental physics. Indeed, every bizarre prediction it makes has so far been observed to be correct. Yet quantum mechanics paints a very strange picture of reality. Among other things, it entails that every particle in the universe is simultaneously in many places at once (quantum superposition), but that whenever we observe a particle, we will always find it in some particular place in space-time (wave-function collapse). As I note in my 2013 article, ‘A New Theory of Free Will’, Einstein thought this to be so absurd that he once scoffed, “Do you really think the moon isn’t there if you aren’t looking at it?” Perhaps the weirdest thing here is that according to the dominant interpretation of the equations of quantum mechanics—the Copenhagen interpretation—the superposition every particle is in never actually goes away or ‘collapses’. Rather, every particle is always in a superposition (many places at once), but observation makes the particle appear to ‘collapse’ to a single particular state.
My theory provides an elegant explanation of these phenomena. In a peer-to-peer networked computer simulation, each computer on the network is running its own unique simulation. So, for example, if we are playing a peer-to-peer networked internet game on 2,000 game consoles, there are in a sense ‘2,000 simulations’ running, each with its own ‘reality’. But at the same time, insofar as all of the computers on the network are interacting with each other—updating what they are simulating based on the data the other computers on the network give them—there is also in a sense just one simulation: the entire network.
This is how the Peer-to-Peer (P2P) Hypothesis explains quantum superposition and wave-function collapse. Insofar as each simulation on the network has its own unique representation of where objects in the simulated environment are, unless there is absolutely perfect error-correction in real time (a computational impossibility), different computers on the network will have slightly different representations of where things in the environment are. Thus, at the level of the network as a whole, it is right to say that everything in the environment is always in a superposition: each particle in a 2,000-computer network will be represented in the network as being at something like 2,000 different places simultaneously. Yet, whenever any particular individual playing their game looks at the simulated world (on their computer), they will always observe things in one particular place or another: since their computer is one of the 2,000 computers running the simulation. So, in a P2P network, objects really are in a sense always in multiple places at once, but will always be observed by any observer to be at one particular place or other—exactly as the equations of quantum mechanics entail.
2. When we zoom out from the individual experience and look at the shared perception of reality, your theory maintains that the joint choices of all conscious observers work to collapse possible paths that can be taken (through the multiverse) into a single actualized reality, which all conscious observers experience in tandem. How do MMORPG’s substantiate this idea and how do they exist as an example we can read into?
The simplest way to understand this is through the example that popped into my imagination when this P2P Hypothesis first dawned on me. In graduate school while finishing my PhD, I spent a lot of time (way too much time, in fact!) playing Halo 2 online. Although Halo 2 doesn’t use a P2P network architecture (each game is played on a dedicated server), some strange things would happen while playing the game sometimes when your game console failed to network with the server properly: specifically, you would appear to be doing something on your console while something very different appeared to happening on someone else’s. Here is one case I remember distinctly: me and friend stood shooting at each other. On my console, I appeared to be shooting directly at his head. However, on his console, my character was oriented in a slightly different direction, shooting at a wall in front of but to the right of his head. And now comes the fun part: because my console was coding me shooting him and his console was coding me shooting a wall, the server told his console that my gunshots were tunneling through the wall and hitting him in the head, making it appear on his console as though my shots were literally going through the wall. Which is what quantum tunneling is, more or less!
My theory, the P2P Hypothesis, simply holds that this phenomenon occurs not through a single computer server that each of us ‘accesses’, but again, in terms multiple simulations running simultaneously and interacting through a network. To translate this into our reality, my conscious awareness is, as it were, processing the digital data of one ‘reality-simulation’, giving me experiences of tables, chairs, people, and so on—like a character in an ordinary videogame. Your consciousness is then running another version of the simulation, and so on, for every conscious observer in the universe. Insofar as all of our simulations are networked together, we observe what appears to be a single, intersubjective world we all experience together: ‘the Universe.’
Where does the multiverse come in? Well, consider again a simple videogame DVD, say for Playstation 4. What is coded on the game DVD? Here’s the answer: every possible move you can make in the game, plus computational rules (or ‘laws of nature’) for what will happen if you do X, Y, Z, and so on. So, for example, we can play Madden 2020 over and over again, playing out the game differently each time. What the Madden 2020 game DVD contains, then, is a vast ‘multiverse’ of possible outcomes—every possible outcome players can experience in the game. My P2P Simulation Hypothesis holds that exactly the same thing is true of our ‘Universe.’ The ‘game DVD’ that our conscious awarenesses are accessing contains every possible set of configurations given the ‘rules of the game’ (the ‘laws of nature’): a multiverse of possibilities. Yet, the actual ‘Universe’ we experience depends on our choices, just like an individual game of Madden. MMORPGs just do this on a much larger level, with large numbers of gamers playing each other simultaneously (as opposed to just two gamers playing each other on Madden).
3. How likely do you, personally, think it is that we exist in a simulation?
I think it is extremely likely. This is because (1) we should believe the best explanations of what we observe in the world around us, and (2) I don’t know of a better explanation of quantum phenomena than that they are produced by peer-to-peer networking. Let me briefly explain both parts of this answer.
This might come as a surprise to a lot of people, but as famed physicist Stephen Hawking points out on pages 41-2 of his 2010 book with Leonard Mlodinow, The Grand Design, there is actually no observational proof of Copernicus’ theory that the Earth orbits the Sun over Ptolemy’s ancient theory that the Sun orbits the Earth. Hawking and Mlodinow write:
Although it is not uncommon for people to say that Copernicus proved Ptolemy wrong, that is not true…one can use either model of the universe, for our observations of the heavens are explained by assuming either the earth or the sun to be at rest. Despite its role in philosophical debates over the nature of our universe, the real advantage of the Copernican system is simply that the equations of motion are much simpler in the frame of reference in which the sun is at rest.
The reason we believe Copernicus’ theory is not that our observations prove it. We believe because it is a better explanation of why planets and other astronomical bodies move the way they do: it explains them in terms of a unified model of mass, gravity, and so on.
So, when people say there is ‘no proof’ that we live in a simulation, I think this mistakes how good scientific explanations work and how we should base our beliefs on the best explanations we have. We have plenty of proof that quantum mechanics exists, and (or so I’m inclined to think) we also some evidence that the peer-to-peer simulation hypothesis may be the most promising explanation for why quantum mechanics exists and has the features it does. If both of these things are true, then we should think it’s likely we’re in a simulation. Since some readers may be skeptical of this, let me conclude with an argument by analogy.
Suppose as a simple thought-experiment that we all lived on the surface of an ordinary wrist-watch, and the hands of the watch rotated in the sky above our heads. If we were standing on the surface of the watch, we would see that a central part of the ‘physics’ of our world would be the movement of the hands. We could observe their movements, and even carry out ‘physics experiments’ on them similar to what actual scientists do with particle colliders. For example, we could smash things against the watch-hands as they move and see the effects that it has on their movement, in much the same way that scientists at the Large Hadron Collider smash particles into each other to see how they behave, decay, and so on. Notice, however, that no matter how many experiments we carried out on the watch hands, our theories of Watchworld physics could never explain why they move the way they do. Those theories could at most describe the ‘laws of nature’ that govern the watch hands (namely, the law that the second hand moves 60 times per minute, and so on).
This is how actual theories of physics such as quantum mechanics and relativity work: they describe how our world’s laws of physics work, but not why they are the laws that govern our world to begin with. Bearing this in mind, now return to the analogy. If we lived on the surface of a watch, we would want to know the deeper reality behind why the hands move the way they do. The only way to determine what that deeper reality is would be to provide a causal-mechanical explanation of what is going on beneath the surface of the watch to cause the hands to move the way that they do. Now, what would a good causal explanation look like? It’s obvious, right? The best explanation would be that there are gears beneath the surface of the watch that are set in a particular way to produce the movement of the hands. Hence, if we were living on the surface of a watch, that’s what we should believe: that even though we cannot directly observe the gears beneath the surface, there probably are gears beneath the surface causing the hands to move around. By a similar token, if the Peer-to-Peer Simulation Hypothesis is the best explanation of why quantum mechanics and relativity exist and work the way they do, then we should believe we probably live in such a simulation even if we cannot observe the computer code, processors, and so on.
Could there someday be further observational support for the P2P Simulation Hypothesis? Yes, and I detail some of them in my published work. The point is simply that I think we already have some very good reasons to think that the hypothesis is likely to be true!
Comments