Reality could be fake.
I began this response as a comment to an article over at The Frontal Cortex titled Reality is Fake?, but it quickly grew too long in length. My thoughts and response to Jonah after the jump.
The “Simulation Theory” has been making waves recently by suggesting that our reality is in fact the result of a simulation being performed by a more intelligent and sophisticated race of individuals. I’ve been meaning to comment on the paper for some time now as it underlines many of the ideas that the Distributed Neuron project touches on. The paper can be found here. It is a good read even if you don’t believe any of it (and indeed, there are certain points I am skeptical about, particularly the mathematical “proof”). But I feel, on the whole, the theory is interesting and may have some validity.
Back to the subject at hand. Jonah at The Frontal Cortex contends this theory isn’t possible because the human brain, let alone the entire universe, is to magnificently complicated for us to simulate. Jonah says it best himself:
In other words, you can’t simulate a brain until you know exactly how the brain works. Otherwise your simulation will suck. So even if we have a computer of infinite power, until our knowledge is perfect, that hypothetical computer won’t be able to conjure up a pretend reality. A microchip is only as good as its code, and our codes aren’t very good.
And with this I agree. We can’t simulate our own minds until we know how everything works. It is just far too complicated and any attempt to simulate it without full details will, plainly put, suck.
It however assumes that our form of biological intelligence is the only form of intelligence possible. I recently touched on this subject and feel intelligence is an emergent property of matter. If you believe that intelligence emerges from discrete subunits (in our case, the behaviors of neurons), it really doesn’t matter what those neurons are actually created out of, be it biological or digital.
Taking this idea and running with it, the Simulation Theory becomes a lot more credible. Hypothetically, imagine we are the advanced race of individuals in the Simulation Theory, maybe 10,000 years from now. We sit down and write out the laws of a universe we want to simulate. We make “matter”, which is composed of “atoms”, which in turn are composed of subatomic particles. We write out the relationships between all these particles, how they interact, various constants of charge and mass and other properties. The laws of the simulated universe are complete because we design them to be complete. Do these laws mimic our physical universe? Maybe, but they don’t need to. If intelligent life is an emergent property of subunits, it should be irrelevant how that matter actually interacts. Intelligence will eventually arise.
So we start the simulation with a big ol’ bang. Matter is in the universe obeying the laws we have set forth, bumping around doing its thing. Given a sufficiently large amount of time (lets say 13.7 billion simulated years), a cluster of matter shows intelligent life. The organisms on a particular planet live out their daily lives, form relationships, feel anger and joy. Given the physical laws of the universe we have designed, they have evolved intricately complex clumps of matter they call “brains”. They have manipulated their environment to build shelter, transportation, entertainment.
Is their existence trivialized any by the fact they are simulated? Are they any less real to each other because they are digital simulations?
No. Their existence is no less real in relation to each other than our existence is to our friends and family. Their reality is unique to them. They struggle with the laws of their universe just as much as we struggle against ours. They would have no way of knowing they are “digital”. Their sensory input tells them they are composed of matter and nothing in their universe can prove otherwise. Short of us (their creators) dropping into the simulation to say “Hey guys, we made you”, there is no way they can prove their universe is the creation of another species. We designed their universe so it was self-contained. We designed their universe to be incapable of interacting with our universe.
And importantly, we did not intelligently design them. They are the product of evolution in their own set of parameters, their own reality and universe. We designed their sandbox, so to speak, and let the laws we set forth create them. In this way, we did not need to understand every complexity of our universe and our own brains to create intelligent life. We merely needed to simulate an environment and let it run its own course. Based on the laws we wrote intelligence emerged.
It is in this way that the Simulation Theory is valid. It would be foolish to think the world has been simulated like The Matrix, starting at an arbitrary point in human history. For the Simulation Theory to be credible, it would have to assume our entire known universe and timeline was a result of simulation. The evolution of our species is no less beautiful or magnificent, it is merely simulated by another species. Furthermore, who can define “simulated”. The universe is real to us, which is all that matters.
As a note, I’m not claiming to be some new-age Simulation Religion freak. I frankly have no idea about the origins of the universe and if there is/isn’t a God, or some race that is acting as a surrogate God. But I do believe there could be validity in the Simulation Theory.
Not that we will ever know 
August 21st, 2007 at 10:57 pm
What if intelligent life, as an emergent property, is reliant on particular interactions? Intelligence would then emerge only if the interactions occured in the right way.
There is the argument that if the laws were altered then microscopic changes could have devastating effects at the macro scale. I.e., if the laws are coded incorrectly at best intelligence doesn’t emerge, at worst the simulated universe blows up. So if we want to simulate intelligence that we’d recognise, I suspect we’d have to get the laws right. And even if we wanted to simulate intelligence of some other kind, if we got a basic law wrong I suspect there’d be a high chance the universe wouldn’t develop far enough for intelligence of any kind to emerge.
But, taking into account the possibility of getting those basic laws wrong, I agree with you. That we currently have difficulty accurately simulating complex objects or interactions doesn’t mean we won’t be able to do it in the future, not does it mean some greater intelligence might not be able to. I’m sure that at some point ancient humans with stone tools thought they were the pinacle of intelligent life.
November 1st, 2007 at 4:20 pm
And what if the simulation did ’suck?’ Would it matter at all? That is, would we be able to notice it? Keep in mind that we have a purely and strictly subjective relationship with our existence. We obviously cannot detach our consciousness from our mind and look back upon our mind to determine whether it’s a good ’simulation’ of intelligence or not.
Perhaps in the ‘true’ universe where the simulation is executed from, the fabric of reality is much more detailed and indeed far different from our perception. ‘Sentience’ and ‘emotion’ here in the simulation might be a laughable farce compared to the entity or system that created it.
For example, imagine characters from an SNES cartridge becoming aware of their creators. It simply could not happen. Two-dimensional existence and looping dialog would seem as advanced as intelligence could possibly get. As a metaphor within a metaphor, try imagining what hue ultraviolet would have if your eyes could detect it. It is outside our range of perception, therefore we have no reference point to imagine it with; ‘violet’ is just a convenient way of naming it, and ultraviolet telescopes merely make it tangible by converting it into a visible wavelength.
Perhaps, even, physical law is so drastically different in this overworld that it makes simulations such as this universe simple to accomplish, even spontaneous. To a human mind, the universe seems vastly complex and impossible to emulate. But is it? It may be deceptively simple, only seeming complex to the simple creatures that live within it. But how can that be so? Rules such as the speed of light, the Planck length and Heisenberg’s uncertainty principle make it impossible to create a computer powerful enough to simulate an entire universe, even within an infinite amount of space.
But one must keep in mind, hypothetically these physical laws could also be simulated.