After popularization of the simulation hypothesis by popular figures like Elon Musk and Neil deGrasse Tyson, it's no longer possible to ignore it. I decided to dig in in order to form an objective opinion. Is it simply a thought experiment, or something to be taken more seriously? The answer is probably a little of both.
The simulation hypothesis has been around for a while, but it was formalized by philosopher Nick Bostrom in a 2003 paper titled "Are You Living in a Computer Simulation?" In this paper, Bostrom argues that "at least one of the following propositions is true:
- The human species is very likely to go extinct before reaching a “posthuman” stage
- Any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof)
- We are almost certainly living in a computer simulation.
By "posthuman" Bostrom, means a civilization whose technology is capable of creating a giant simulation (which Bostrom calls an "ancestor simulation") that contains the universe as we perceive it. In The Simulation Hypothesis, Rizwan Virk calls this "The Great Simulation." There is no consensus on the validity of the hypothesis. As Bostrom himself explains, "Some think it obvious that (1) is true, others that (2) is true, yet others that (3) is true."
Much of Bostrom's argument focuses on the theoretical math behind the hypothesis. Is it computationally possible? But he himself admits it's not all that important, because even if the math were off by an order of magnitude, the hypothesis still stands. More important than the math are the ethical concerns. Would a civilization even want to make such a simulation if they could? Bostrom says yes—an answer that is heavily based on the fact that many people today would want to do it if we could.
The hypothesis makes assumptions about a posthuman race based on our current working knowledge. To me, that's one of its major shortcomings. Once we reach the point of being able to do it, our thinking will have changed. Trying to deduce a posthuman race's intentions is no different from trying to deduce God's intentions. And like many people, I see the hypothesis as a modern replacement of God theory that, whether true or not, solves very little. It simply raises more questions. Sooner or later, even if we are living inside of a simulation which is inside of a simulation, etc. there must be a primary reality in which we would have to finally answer these questions instead of deferring to the computation-centered zeitgeist of today.
There are still elements of the hypothesis that are very attractive though. One of the most convincing arguments for why we may be living in a simulation, is our inability to understand quantum physics. Quantum indeterminacy, one of the hallmarks of quantum physics, shows through experiments like the double-slit experiment that a physical system at quantum scales remains undetermined until observed. Schrödinger's famous thought experiment demonstrates this nicely by asking us to imagine a cat in a box which is both dead and alive until observed. In other words, consciousness may be central to how the universe operates (Something reminiscent of many mystical writings. I recently wrote about The Kybalion, an occult classic that asserts the universe is mental). Physics doesn't have this problem at larger scales, and people have been searching long and hard for a theory which could unify the two systems. The simulation hypothesis offers a convenient solution to this problem by supposing that quantum indeterminacy may be some kind of optimization technique used by a rendering engine.
Rizwan Virk illustrates this well in The Simulation Hypothesis. He explains that video games have for a long time utilized some form of conditional rendering. Early games like Zelda only show the screen that you are actively in. Games like Civilization show dark areas where your units cannot see. This is often referred to as "fog of war". Games like World of Warcraft (called MMORPGS or massive multiplayer online role playing games), which may be the best comparison of all, also use techniques similar to this. What looks like an intentional design is really a technique to prevent the game from rendering more than it needs to. The information in dark areas is saved on a computer somewhere, but not rendered to save computation power. Quantum indeterminacy may work similarly, acting as an optimization technique in our universe. If computing power governs our universe, then why bother using computation when you don't need to. Like a video game, you can simply save the state of the universe and render it once it is observed.
Part of my hesitation in accepting this is how conveniently it fits into our existing metaphors of video games. I wonder if we aren't seeing more noise than signal because our minds are familiar with these ideas. If we're to believe that indeterminacy is a form of optimization, who is to say that a universe outside of simulations wouldn't also need optimization? Ontology, the branch of metaphysics that deals with existence, is still in its infancy, and it relies heavily on science for new information. We just don't have that much information.
AI and Consciousness
If we are in a simulation, what is consciousness? There are two main arguments surrounding consciousness in simulations are:
- We are software based AI whose consciousness lives on a computer outside of the simulation
- We are hardware based life forms that exist outside of the simulation and are plugged into the simulation.
While option 2 has been popularized by the media in movies like The Matrix, this is not what Bostrom means by ancestor simulations. If physical bodies were required in order to populate the simulation, the number of simulations would be limited by the resources required and the odds that we would be living in a simulation would decrease dramatically. If instead our consciousness is purely software based, as in option 1, it is easy to conceive of infinitely nested simulations that require nothing tangible outside of a computer. Because this allows for easy nesting of simulations within simulations, the likelihood of our being in a simulation would increase greatly. If our consciousness exists in this way, being saved on a server outside of our universe, perhaps we could also be easily transported into other states of existence. And maybe dreaming functions in this way.
But what does this mean for our free will? Does this increase the likelihood of our having free will? To me, it changes nothing. Just as God may have gifted (or penalized?) us with free will, so too may our creators have programmed room for our decision making. So we haven't gained any new ethical information. The simulation hypothesis is a re-packaging of traditional God theory with the aid of scientific knowledge.
One concern with accepting the simulation hypothesis is that if we are in an ancestor simulation, as Bostrom posits is the most likely simulation scenario, this implies that we are special—yet again. It implies that a civilization created this universe to simulate their own, and that we are created in their image. Sound familiar? You can take this further with the introduction of "non-player characters" or NPCs, artificially intelligent players that lack our own consciousness. If we do run into alien species in the future, they could simply be NPCs used to make our universe appear more real to us. This kind of ethnocentrism is dangerous to say the least. We have thought we are the center of the universe before, and it has proven incorrect in every way, every time.
And why do we have such an obsession with ethics? Bostrom attempts to form a logical conclusion, based on fear, that all civilizations in the simulation hierarchy would behave ethically because they can not be certain of whether they live in a simulated reality. He says that "all the demigods except those at the fundamental level of reality are subject to sanctions by the more powerful gods living at lower levels... Because of this fundamental uncertainty, even the basement civilization may have a reason to behave ethically." This again screams of God theory to me. If all physics are pre-programmed, would not the master civilization have complete control over the type of morality that developed? Given the power to run a simulation, morality must be programmed in some way. It's hard to imagine genuine free existing in a simulated reality.
There's a certain naivety in creating a metaphorical overlord based on our current working knowledge of quantum theory and computing. The universe may be some kind of simulated reality, but what kind? Without understanding the true nature of reality, how can we draw meaningful conclusions about the differences between a simulated reality and a real reality. In both cases, you must answer to the same questions of finite and infinite. In both cases, you must answer questions about God and the fundamental nature of existence.
To use our nascent understanding of computation as a basis for a grand unified theory of the universe seems likely to result in an outdated theory a century from now. But there is also something extremely attractive about an hypothesis that is at once familiar and could explain phenomenon that have remained illusive up until this point. If computation really is responsible for our reality, it must be a vastly superior form of computation that we can scarcely understand. Perhaps we are starting to understand, or perhaps the future has greater surprises in store for us.