The simulation hypothesi.., p.12
The Simulation Hypothesis, page 12
Are We Simulated Characters in an Ancestor Simulation or Conscious Players in a Video Game?
The most important conclusion that Bostrom makes is that we are more likely to be a simulated being in an ancestor simulation than we are likely to be non-simulated beings. While we can’t know the exact percentages to plug in to his equations, he clearly shows that there is a nonzero percent chance that we are living in a simulation.
Bostrom’s conclusion is that the simulated beings are really artificially intelligent characters in simulations rather than conscious entities that have presence in the real world. This would mean that as we interact with other individuals in the world around us, if we cannot tell that they are computer-generated beings or AI, then we can assume that whoever created the simulation was able to build powerful computers and get to the simulation point.
If we are simulated beings in a simulation, then other characters around us are already passing the Turing Test. This means that if the Turing Test will ever be passed, then it has probably already been passed! This assumes that it was passed in the reality “above” our simulation, and while we are trying to solve it “inside our simulation” on our computers (which could be, for all practical purposes, simulated computers!).
As a video game designer, this reminds me of our attempts to create realistic NPCs—non-player characters. As games have gotten more sophisticated, their AI characters have gotten more sophisticated as well. I recall early text games like Zork that had players that would talk to you and those attempts to make the characters seem realistic. AI has advanced well beyond that, but we do not currently have NPCs that can pass the Turing Test. Once we do (in 10 years or perhaps 100 years), the possibility that we are already interacting with people who are actually NPCs goes up considerably.
Bostrom, in fact, thinks that “we” are the simulated consciousness. He makes the point that perhaps we don’t need to pass the Turing Test. but we are constructed in such a way that we can’t tell if other “humans” (as we think of them), or other characters in the simulation, are simulated or not.
Video games, on the other hand, as we know them, involve an actual player that is outside the game who is playing the video game. A given video game might have many NPCs, or many instances of the same NPC, but it wouldn’t be a video game unless there were one or more players. In the simulation hypothesis that I am making in this book, there are billions of players who are inhabiting our consciousness, but there is nothing that says some percentage of the “characters” we come in contact with couldn’t be simulated consciousness.
What is Consciousness?
Whether we are dealing with the Turing Test or simulated characters, or self-aware robots or AIs like Data or HAL 9000, thus far we have been taking the idea of “digital consciousness” for granted.
We really don’t have a definition for what digital consciousness might actually be because we really don’t understand what human consciousness is all about. Max Tegmark, an MIT professor and author of Life 3.0: Being Human in the Age of Artificial Intelligence, recalls that his friend, neuroscientist and brain researcher Christof Koch, was discouraged from researching consciousness by one of his mentors, Nobel Laureate Francis Crick (co-discoverer of the structure of DNA). 21 Why? Because it is a thorny issue in Western science.
The materialist Western science point of view seems to put consciousness into one of two possibilities:
Consciousness as an experience is very subjective and, therefore, it falls outside of the purview of the “physical sciences” and thus is ignored.
Consciousness is really the result of physical processes—i.e. chemical reactions. More specifically, neural activity in the brain is responsible for consciousness, and, as we are able to map out that neural activity more fully, we can easily reproduce consciousness artificially.
While No. 1 has been the perspective of many scientists over the past hundred years or so, lately, perspective No. 2 has been gaining favor with the emerging field of neuroscience.
Digital Consciousness vs. Spiritual Consciousness
The Eastern mystical and Western religious views of consciousness are different from those of the sciences. In the religious views consciousness exists independent of, and outside, the physical body. Consciousness enters the physical body around birth and leaves it at death, according to these traditions. Consciousness even leaves the body during dreams in some of these traditions.
We don’t know which, the spiritual or the material, is closer to “reality,” and this is a debate that has been going on for many years. If consciousness exists outside the physical body, where is it? The simulation hypothesis would posit that it is stored outside the rendered world, a world that we cannot see because we are caught inside the simulation! This might be uploaded to some server or stored on local devices (i.e., our bodies) in ways we haven’t yet discovered.
In both of these cases, the materialist view (that consciousness is a result of a pattern of neurons firing) and the spiritual point of view (that consciousness exists outside of and is downloaded into the body), there is one commonality:
Consciousness is really a set of information and a processing of that information.
From the spiritual point of view, this information takes on various forms, including a soul that is indestructible in Hinduism, to a soul that survives death and lives in eternity in Heaven or Hell in the Judeo-Christian-Islamic traditions, to no permanent soul at all in Buddhism, where it is seen as a “bag of karma,” In all these cases, whatever information is associated with the “player” is beamed down and linked through some mysterious process to the biological machine that is our body.
In the materialist point of view, if only we could have enough computational power to model all of the connections in the physical brain, we would have a “copy” of a person at a certain point in time.
Recently, a group of researchers in Switzerland believe that they have created a scaled computer model that reproduces and acts like the brain by modeling and reproducing an individual’s neurons: They believe by modeling a rat’s brain, which might consist of “only” 10,000 neurons (with an associated 30 million neural connections), they had a scale model that could model an entire brain within a few years. The human brain, on the other hand, is composed of networks connecting 1012 neurons through 1015 synapses.22 As of the writing of this book, no one has been able to successfully model that many neural connections, but it doesn’t mean that it’s as far off as we might think.
Ray Kurzweil also makes the point, as many others have, that it’s not just consciousness that is information. The cells in your body have been replaced many times—you are literally not the physical person that you were many years ago. There must be some “information” (which Kurzweil refers to as “patterns”) that defines who you are and tells the cells how to grow. This applies not just to healthy but also to diseased cells; theoretically if all the cells are replaced, any diseased cells should just disappear on their own. This doesn’t happen because of patterns of information. Even biological entities, at their core, follow a computational structure.
Regardless of which approach you subscribe to, the materialist or the spiritualist, information seems to be a key part of consciousness. And if consciousness is information, then it can be represented digitally and could potentially be downloaded and uploaded to “servers,”, even if it is well beyond our ability to do so at the moment.
As we start to view consciousness and biological reality as a type of information, the simulation hypothesis seems more and more likely, whether you ascribe to a materialist or a spiritual worldview!
Does Simulation Explain Our World?
In this part of the book, we have gone through how we might build something akin to The Matrix, showing that it is technically feasible within the next few decades or century at the most. As Bostrom’s Simulation Argument shows, if any civilization ever reaches this point, then it’s very possible (even likely) that we are already in a simulation!
The next sections in this book will explore some of the reasons the simulation hypothesis actually explains many unexplained questions about our world. This includes some of the paradoxes and unusual aspects of quantum physics as well the religious views expressed by Eastern mystics and Western religions.
Surprisingly, it is the religious views that are closest to the video game analogy: that our physical world is a kind of “illusion,” populated by conscious beings that exist outside the simulation. Bostrom didn’t seem to contemplate this possibility, but it may be one of the most intriguing aspects of the simulation hypothesis, bridging the gap between two domains of knowledge that rarely overlap, religion and science.
In the last part of the book, we’ll then wrap up by exploring additional evidence of computation, giving voice to some of the skeptics, and considering the bigger philosophical questions and implications of this hypothesis in the last part of this book. In fact, we’ll see how the simulation hypothesis may be the one model that ties everything about our world—our science, spiritual and religious teachings—together in a coherent way
Part II
How Simulation Explains Our World: The Physics
The great difference between old and new physics is both much simpler and much more profound: both the old physics and the new physics were dealing with shadow symbols, but the new physics was forced to be aware of the fact—forced to be aware that it was dealing with shadows and illusions, not reality.23
—Ken Wilber, Quantum Questions
The layman always means, when he says ‘reality’ that he is speaking of something self-evidently known; whereas to me it seems the most important and exceedingly difficult task of our time is to work on the construction of a new idea of reality.
—Wolfgang Pauli, Theoretical physicist
Chapter 5
Conditional Rendering and the Collapse of the Probability Wave
For those who are not shocked when they first come across quantum theory cannot possibly have understood it.24
—Niels Bohr
As I began looking more deeply into quantum theory, I found many parallels between the world of video games and this puzzling area of physics that might offer some evidence that we are indeed living in a simulation.
It is my contention in this next part of the book that the simulation hypothesis explains things about our physical world that have been difficult to explain. While it has been productive in finding out the rules of the physical world, modern physics has been unable to answer the big question of why the universe works in this way.
This is true not just of quantum physics, but also of questions that arise from the findings of classical and relativistic physics. We’ll explore each of these questions in the next few chapters (though not necessarily in this order):
Is space quantized—i.e., does it consist of pixels like a virtual world?
Is time quantized—does the universe have a clock speed and proceed like a computer simulation?
What is the purpose and nature of quantum indeterminacy? Is it an optimization technique similar to what rendering engines do in video games?
Why does “observation” cause the collapse of the quantum probability wave?
Do parallel worlds actually exist? Or are they just probabilities on a “virtual grid” of possibilities, like a video game?
Does matter actually exist, or, like pixels in a video game, are subatomic particles simply turned on or off as needed?
Why is the speed of light both a fundamental constant and an absolute limit?
Does the physical world have a physics engine like a video game? If so, does it allow for “jumping” around like video games and instantaneous communication?
Is there an objective universe or is it, like a video game, information that is rendered on each of our “computers” (i.e., our minds)?
These are big and complicated questions. If we adopt the simulation hypothesis as a model, it turns out that it gives us some answers that explain both the purpose and underlying mechanism of many mysterious aspects of physics.
As I mentioned in the introduction, while studying at MIT, my professors told me that if a new model offers better explanations for an observed phenomenon than an existing model, then we should consider adopting that model as a better possibility for how the world works.
Video Games and Quantum Indeterminacy
The first topic we’ll tackle relates to how video games are built and what this can teach us about one of the oddest findings at the heart of quantum physics: quantum indeterminacy (or “QI”). This is the main finding of quantum physics which proved so shocking to physicists and laypersons alike, that Niels Bohr was referring to in the quote at the beginning of this chapter.
Quantum Indeterminacy is the idea that the world may not be “rendered” if we are not looking at it. As we saw in the previous section about video games, techniques in computer science to optimize rendering that led to virtual reality glasses only render that part of the world that can be viewed by the player at one time.
Moreover, while there may be a master “state” on the server in multiplayer video games, rendering is done on each client machine. This corresponds to the finding in quantum physics that a probability wave collapses to a specific reality only when there is an observer. This will lead us into another aspect of quantum physics that seems like science fiction: parallel worlds.
Using an analogy of early AIs in video games, going back to my very simple example of Tic Tac Toe, we’ll ask: Is there some conscious (artificial or otherwise) mechanism that is scanning these probable parallel worlds and choosing one among many? Also, since there are multiple conscious observers in a given “world” who are all affecting it, we will explore the nature of a multiplayer program that has to keep all of these in sync and what this means for an objective reality.
The Old Physics
Before we jump into the crux of this chapter’s point about quantum indeterminacy and how it relates to the simulation hypothesis, some background information will be needed. Let’s take a quick look at what’s often called the old (“classical”) physics models, which were built on the works of Sir Isaac Newton, and the new (“relativistic and quantum”) physics, which began with Albert Einstein but was really fleshed out by a number of eminent physicists in the early 20th century, including Niels Bohr, Werner Heisenberg, Wolfgang Pauli, Erwin Schrödinger, and others.
In the classical view of physics, the universe operates independently of people like us (or observers) and does so in a purely mechanistic way. Newton’s laws of motion could be used to describe the movements of heavenly bodies simply based on their mass and position using basic physics equations. In fact, utilizing Newton’s equations, Pierre-Simon Laplace put together two tomes that described the motions of every heavenly body that was known at the time, Exposition du système du monde and the Mécanique celeste.
In this model, each of the planetary bodies is an independent physical entity that acts on the others per the laws of classical mechanics. It is a purely deterministic model—in order to know where things end up, you simply need to know where they started and which forces are acting on them. In this view of the world, the observer is just that—an observer that has no effect on the motions of the bodies being studied.
This idea, which started in the macroscopic world, was extended to the microscopic world when Lord Ernest Rutherford discovered the nucleus of the atom. The idea was that there were basic building blocks, which were discreet and independent of one another, just like the planets in the solar system. The nucleus of an atom consisted of protons and neutrons, while electrons traveled around the nucleus in orbits, analogous to how planets orbited the sun. This was called the planetary model (or the Rutherford-Bohr planetary model).
The only real challenge to this model was the discovery of electromagnetic fields, which seemed to be a completely different part of the physical world altogether. Michael Faraday and James Maxwell studied these phenomena, and Maxwell’s equations described something new that was not accounted for in the existing models: an electromagnetic field. Still, since this was in the newly emerging field of electricity and not dealing with atoms, Newton’s model of classical mechanics reigned supreme.
The New Physics and the Wave/Particle Duality
As is often the case with science, it was the edge cases that started to redefine the model of the physical universe. The classical model, while still relevant for everyday intuition and motion of big objects, started to fall apart in the early 20th century as physicists delved into nature’s mysteries and found it lacking, particularly in two cases: (1) when objects were traveling at very high speeds, and (2) at the subatomic level with very small objects.
The first major revolution was Einstein’s theory of relativity, which redefined our ideas of space and time. Einstein found that the speed of light was a fundamental constant in our world, and the speed of light didn’t change. Instead, as objects traveled faster to get closer to the speed of light, both space and time seemed to adapt. This finding, which has been verified experimentally, was quite puzzling. Why would the speed of electromagnetic waves (the speed of light) be a fundamental constant in our world? We’ll talk much more about this in Chapter 7.
Einstein also participated in the birth of quantum physics and the redefinition of how atoms and subatomic particles behaved. This started with the Pauli Exclusion Principle—the discovery by Wolfgang Pauli that no two electrons in an atom can occupy the same position or quantum state at the same time. It continued with Einstein’s investigations into the photoelectric effect (for which he received his sole Nobel Prize)—in which he found that little bits of light acted as “quanta,” and that these particles didn’t move continuously from one state to another but “jumped states” as they absorbed energy (this was called the “quantum leap”). Later, these quanta of light came to be called photons.
