Simulation theory has been getting a lot of attention lately. However, it seems to have been misinterpreted by some as proof we are most likely in a simulation. This seems to me an incorrect conclusion to come to from the argument.
So, I thought it would be interesting to take a look at what simulation theory is and what some of the possible consequences of the theory are.
First let’s look at what the theory is.
Simulation theory is the hypothesis that we may be living in a simulation. The idea is that if technology continues to improve at any pace at all, we will eventually have the ability to create virtual realities indistinguishable from reality.
If you think this is far fetched consider the history of computer games: Just over Forty years ago we were playing pong. Today our games are photo-realistic, and we have more immersive virtual reality technology in the pipeline.
Just imagine what games will be like in ten or twenty years time, let along another forty years. It seems certain we would achieve simulations if there was any rate of progress at all.
Furthermore, that’s based on the technologies available to us now - we don’t know what break-through technology is just around the corner.
So, it seems not only possible but inevitable technology will reach the point of being able to create simulations indistinguishable from reality.
The consequence of this line of reasoning is that our ancestors may have reached this point already and we are living in one of their simulations.
So that’s the essence of the simulation hypothesis. Let’s now look at the hypothesis as originally proposed by Nick Bostrum.
The Simulation Hypothesis
Nick Bostrum proposed that at least one of the below statements must be 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. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor‐simulations is false, unless we are currently living in a simulation.
When reading this for the first time it seems there is a missing option. Surely, there is the possibility that we aren’t living in a simulation - that we are living in base reality even if (1) and (2) are false?
This is covered by the final point -
‘It follows that the belief there is a significant chance that we will one day become posthumans who run ancestor‐simulations is false, unless we are currently living in a simulation’
So, the argument is that if we will one day reach the point of running simulations we must already be in one.
At first this seems counter-intuitive - we’ll leave this for now and come to back to it later!
First let’s look at the simpler scenario’s (1) and (2).
(1) We are going to be extinct
This is straightforward - One possible scenario is that we will become extinct before reaching a level of technology where we are able to run simulations.
We may will kill ourselves with bio-chemical weapons or find some other interesting way to die. Perhaps some natural disaster will befall us.
So, we will either become extinct or we will reach the posthuman stage and have the technology to run simulations.
Given our previous discussion on technological progress this seems reasonable. We will either reach the posthuman stage (with all the technological magic this implies) or become extinct before we reach this point.
(2) Posthumans won’t run experiments
Another possibility is that posthumans might choose not to run simulations even though they can.
There may be any number of reasons for this;
Firstly, on reaching the posthuman stage we may not want to run simulations simply because our motivations and interests have changed.
Though we see running simulations as desirable today, this desire may cease at the hugely more advanced posthuman stage. In short, we cannot assume our current and future motivations and interests will be the same.
Another possibility is that we may impose strict sanctions on running simulations. Perhaps, as we become aware of moral or ethical issues. To make this viable posthuman civilisations must converge on a path where no one individual has the wealth or freedom to run simulations despite these sanctions.
These are not the only possibilities. There are many other possibilities why we would choose to not run experiments.
(3) We are living in a simulation
Now, let’s examine the final and most interesting scenario for our discussion.
Let’s look at the original scenario again;
‘We are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor‐simulations is false, unless we are currently living in a simulation’
This is a bit wordy so let’s split this up into two;
- 3.0 We are living in a simulation.
- 3.1 The above statement (3.0) is false, unless we are currently living in a simulation.
Why is 3.1 true? Why does it follow that the belief that we are in a simulation is false unless we are already living in a simulation?
To answer this let’s take a detour back to the 1970’s game show let’s make a deal.
The Monty Hall problem
In this game show there are three doors. You pick a door and get the prize behind the door.
Behind one door is an amazing prize but behind the other doors there is a booby prize or nothing.
Let’s say you pick a door. The host then opens another door to show you a booby prize and asks you if you want to change your choice to the third door - the one that you didn’t choose, and the host didn’t open.
What should you do?
Well initially when there were three doors you had a 33% chance of picking the door hiding the amazing prize.
In other words, your chance of being wrong was 66%.
So it follows you most likely didn’t pick the door with the amazing prize.
Therefore, you should swap for the remaining door, because by swapping you have a 66% chance of picking the door with the amazing prize but only 33% chance by staying with your original choice.
Simulations and probability
So the idea is similiar in that it is based on likelihood; There are infinitely more possible simulations than the original base reality (just the one).
So, if you assumed you were in the original base reality you would be wrong most of the time. Whereas, if you assumed you were in a simulation you would be right most of the time.
We can describe this as the below premises and conclusion;
- There is a significant chance that we may one day become posthumans who run ancestor simulations.
- It is far more probable we are one of the many simulations than the original base reality.
- Therefore, if (1) is true then we are most likely already in a simulation.
So now we understand the simulation theory.
Before we finish I’d like to look at one similiar theory from the past.
The evil genius
The idea that reality may not be what we think it is has a long history.
Descartes proposed that he could not trust anything he thought was fact because the world may be an illusion created by an omnipotent evil genius.
This genius may have set the world up to fool Descartes and there was nothing he could do about it. Everything that he took as fact was false.
This is an extension of the classic question ‘how do I know what I perceive as real is actually real?’.
This is known as a sceptical hypothesis - it is an argument that if true, would yield all our beliefs about reality invalid.
This seems similiar to the simulation theory. However, there is an important difference;
It’s been debated that the simulation theory is not a sceptical argument - the argument is that we may be in a simulation of our past.
Therefore, our beliefs about reality hold true even if we are in a simulation of that reality.
However, it doesn’t seem impossible to recast the simulation theory in a way that makes it a sceptical argument. Why would posthumans only simulate their past - why not make up worlds that are not based on reality?
Wouldn’t this make it a skeptical argument? Perhaps, but in the original forms Descartes argument is a sceptical argument whilst the simulation hypothesis is not.
If we are in a simulation there are some interesting consequences for ethics, morality and physics among others.
For example, it occurred to me that time in simulations would probably be different. It doesn’t seem useful to have a simulation run at the same pace as base reality. At least, it would be more useful to have simulations run faster than base reality.
The impact on ethics may come from the possibility that the simulation could be turned off. Should this affect how we live our lives?
So, which of these three scenario’s is true or more likely?
Well, it wasn’t Nick Bostrum’s purpose to show which one is true - only to show that one of these must be true.
One of them is true, we just don’t know which one. Therefore, it seems to me the most intelligent course is to ascribe equal weight to them until we have more information.
Perhaps one day science will be able to discount one or more of the possibilities but, for now at least, it seems we can’t know for sure.
All we can deduce is that if (1) and (2) aren’t true, then it’s more likely we are in one of our ancestors simulations than it is we will be the creators of the first simulation. We’d have to be extremely lucky to be in the original reality as opposed to a simulation.
Assuming we are not amazingly lucky, I certainly hope (1) isn’t true as no-one likes the idea of extinction. Neither does the idea that our simulation could be turned off seem any more palatable.
Perhaps the best we can hope for is (2). Either that or we need to ensure the simulation is ‘interesting’ for our ancestors to keep watching!