- Elon Musk recently said we probably live in a computer simulation.
- He was quoting Nick Bostrom, who said we’re either in a simulation or there will never be one.
- It doesn’t matter to us whether our universe is constrained by physical constants or lines of code.
- If we do live in a simulation, are our programmers gods?
Having settled the problem of my own existence, I then have to worry about whether the world around me is what I perceive it to be. Now I’m on shakier ground because my consciousness receives its information in a heavily processed form. A lot happens between electromagnetic radiation at a frequency of 470nm hitting my retina and me thinking of the colour blue. I’m on even shakier ground when it comes to other people. I have to conclude that because people respond to most things in much the same way I do, they experience the world much as I do.
To put it another way, I have to leap from the observation that as you look, sound, feel and similar to me, you are a human being like me and not just a construct of my perceptions. There are a lot of assumptions tied up in that conclusion, but I can’t navigate the world without those assumptions so I may as well go with them.But what if we’re both part of someone else’s reality?
Reality from the page to the screen
The idea that our reality may be a construct of someone else’s was a running theme in the work of Philip K Dick. Dick’s stories often posed the question of which of several pasts or futures was the real one, or whether someone’s perceptions and memories can be so manipulated that their understanding of reality has no objective basis. Since the 1980s, the subgenre of ‘cyberpunk’ has been showing us virtual worlds with virtual inhabitants simulated on massively powerful computers.
The blend of questions about the nature of reality with cyberpunk was picked up by Hollywood in 1999, when The Thirteenth Floor and iconically, The Matrix, brought the simulation question into mainstream culture.
We’re still talking about it. Inventor and entrepreneur Elon Musk recently said that he and his brother discuss it so much that they had to agree never to talk about it in the hot tub. He was answering a question from the floor of the 2016 Code Conference when he said, ‘there is a one in billions chance that this is base reality’. Musk thinks we are almost certainly constructs in a simulation:
Musk was quoting an argument laid out by futurist Nick Bostrom. Let’s call it Bostrom’s paradox:
Unless we are now living in a simulation, our descendants will almost certainly never run an ancestor‐simulation.
That is the closing sentence to his 2003 paper, though it’s worth summarising the steps it took him to get there.
Bostrom starts with the observation that computing capacity has been rising since the day someone put some beads on a thread and called it an abacus. Now we have simulated worlds populated by thousands of constructed individuals with their own artificial intelligence, not as the cutting edge of research but as a playground. Game worlds may be much simpler than the real world and non-playing characters may have very limitedintelligences but if the trend continues, it’s only a matter of time before we’re able to run a simulation with constructs as intelligent as ourselves. If it’s possible to do that, sooner or later someone will do it.
There’s no shortage of reasons for doing it. Terry Kidd discussed the idea of ancestor simulation to understand events in our past. Who wouldn’t like to model what may or may not have happened the day JFK was shot?
If a simulated world was possible today, it wouldn’t be hard to find proposals for using it. Perhaps we’d want to test variations on consciousness or to put versions of ourselves in different physical worlds, or perhaps we’d just want a new reality TV show free of ethical constraints.
If our descendants do it once for one reason, they will probably do it many times. If they build simulations with enough complexity, the artificial intelligences in those simulations may build simulations of their own. If any sufficiently advanced technological civilisation will produce several simulations, it will have many more consciousnesses running in thosesimulations than what Musk called ‘base reality’. Therefore, we are far more likely to be among the many in the simulations than among the few in base reality.
Applying the Copernican principle to Bostrom’s Paradox
The Copernican Principle states that we are more likely to be alive now than at any time in the past for the simple reason that there are more of us alive now than there have ever been before. The same principle states that if there is a time in the future where there are vastly more simulated consciousnesses than real ones, we’re probably conscious in that time. If we don’t, it’s probably because that time will never come.
If this is base reality, the idea is already conceived so the only reason it wouldn’t happen is if it can’t. There may be a limit to computing capacity that we’ll hit before we’re able to simulate a consciousness equivalent to ourselves. A more likely reason is that our civilisation will hit some catastrophe before we develop the computing capacity we need to do it.
So when Musk said we are probably living in a simulation, he was being an optimist:
We should hope that that’s true because if civilisation stops advancing, that may be due to some calamitous event that erases civilisation.Self help for a constructed reality
If we accept that the most likely and the best case scenario is that the world around us is composed of code, what does it mean to us? Or to put it another way, does it matter whether we are the product of collisions between sub-atomic particles or the product of
someone’s line of code? We are still the product of our consciousness and our perceptions, however we come by them.
Your joys, heartaches, pleasures and pains feel the way they feel, whether they’re experienced in silicon or carbon.
I feel bound to point out that technology may well leave silicon behind before it gets to the point of being able to model a world as complex as ours. Evidently, I was coded as a pedant.
The question that usually follows Bostrom’s paradox is whether a construct can have free will? It’s unanswerable because if we’re constructs in a simulation, we have as much or as little free will as we’ve been programmed to have.In practical terms, it’s sensible to assume we have free will whether or not it’s objectively true. If we don’t have free will, we don’t get to choose whether or not we believe we have it, let alone do anything about it. There would be no point in worrying about it, although we don’t get to choose how worried about it we are.
If we do have free will, we are free to believe we have it or to believe we don’t. If we believe we don’t have free will when in fact we do, we’re not only wrong but we’re not equipping ourselves for good decision making. The only sensible course is to believe we have free will and act accordingly. If we’re right, so much the better. If we’re wrong, it doesn’t matter.
Free will has been a favourite subject of argument among philosophers since Aristotle, and
presumably before. If it was argued about before the possibility of simulated worlds was conceived, it’s safe to conclude that whether or not we’re in a simulation has no bearing on the argument.
Gods and geeks
Many religions include a belief that our world and our species were created by someone or something more powerful than us, which is why the free will argument is nothing new. Is there a meaningful difference between a god who says ‘let there be light’ and a line manager who says ‘let there be code’?
Not in practical terms, though it does raise a theological point: to accept Bostrom’s Paradox and to agree with Musk’s interpretation, that we’re almost certainly in a simulation, is to reject atheism. Not because Bostrom or Musk suggest any religion ever conceived is based on the true nature of god. It’s because anyone or anything who created our universe and put us in it is a god by any linguistic or religious definition. We can only guess at the nature of that god, but he, she, it or they achieved godhead at the momentthey set the simulation running – even if they themselves are running within a simulation of their own.
Whoever or whatever that god is, I have one thing to say to you: dude, you have a warped sense of humour.
Elon Musk asked what was wrong with his summary of Bostrom’s Paradox, and no one was able to find an answer. Neither can I. If you can, please let me know in the comments.