Friday, June 25, 2010

The simulation argument

I've heard the simulation argument from time to time in different forms and decided to look into it a little bit more thoroughly. So I gave Nick Bostroms article from 2003 entitled "Are you living in a computer simulation?" a proper read through the other day.

So, what is the simulation argument anyway? Well, Bostrom summarizes it succinctly in the very last sentence of his paper: "Unless we are now living in a simulation, our descendants will almost certainly never run an ancestor-simulation" (p. 11). Not entirely clear? Let's look it a little more in depth.

First and foremost, the simulation argument is based on the notion that some form of functionalistic theory about consciousness is true. That is, it has to be true that humans, simulated in a computer, have some kind of qualia, that it actually feels a certain way to be one of those humans. This is somewhat of a leap of faith but, as I will come back to later, the beauty of the simulation argument isn't that the conclusion necessarily is true, but rather that the conclusion seems to follow given the premises.

Second, the simulation argument argues that future generations, given that slightly less future generations don't manage to start a global nuclear war or something equally unhealthy, will have a ridiculous amount of computing power available. If it is likely that somebody, or even just someone, would run a sufficient amount of simulations of their own prehistory, where the inhabitants would be unaware of the fact that they lived in a simulation, it would follow that most humans that have ever lived (simulated or real) would be simulated. And since there is no way of telling whether one lives in a simulation or not, wouldn't it be most probable that we actually lived in one of these simulations rather than in the "real" world?

Bostrom introduces a lot of twists and turns to this argument and the paper itself is well worth a read through just because it's pretty damn thought provoking. But, once again, what I find most fascinating with the simulation argument is that the conclusion actually seems to follow from the premises. Whether the premises are true is a different matter and here, I'm a bit skeptical.

The main question I find myself keep coming back to is the question of why anyone would want to run a simulation like this in the first place. In a posthuman society, capable of extremely demanding technological feats, people (or more likely cyborgs/robots) would probably be able to satisfy their needs directly without having to go through the fuss of changing the outer world to please themselves (this counterargument is based on the assumption that such a simulation would have no practical benefit but rather just be something created to entertain its creators, something which indeed could be questioned). Bostrom actually brings this up himself:
But perhaps many of our human desires will be regarded as silly by anyone who becomes a posthuman. Maybe the scientific value of ancestor-simulations to a posthuman civilization is negligible (which is not too implausible given its unfathomable intellectual superiority), and maybe posthumans regard recreational activities as merely a very inefficient way of getting pleasure – which can be obtained much more cheaply by direct stimulation of the brain’s reward centers. (p. 9)
However, it would be enough that just one individual would run one simulation, for half of the subjects ever to live to be simulations rather than the real deal. Bostrom points this out, and it's a good point indeed.

The simulation argument can be varied and applied to different situations. For example, imagine that you would want to live forever (in my experience, most people don't want that, something I find extremely strange, but let's not dwell on that right now). How would you achieve it? Well, one way would be to make sure that in the future, there would be a large number of simulations being run of your life, with the additional feature of an afterlife (assuming that no afterlife will be happening after you life). Given this, it would be more likely that you would live in one of these simulations than that you would be the original you.

In the end, I find the functionalistic assumption to be what makes the truth value of the conclusion of this argument doubtful. Show me that that assumption is true, and I will probably start saving up money for those computers I'll need to be immortal.

1 comment:

Richard B said...

You might want to read Barry Dainton's restatement of Bostrom's Simulation Argument as well. I believe that Dainton makes it clearer that simulated consciousness is not necessary, only a sufficiently complete and consistent perception of consciousness. You or I may now lack qualia, assuming it is necessary for authentic consciousness, but if we are simulants which perceive that we are conscious persons then that simulated perception is sufficient for Bostrom's Argument.

Best wishes,