Nick Bostrom is the one who thinks there is a 20% chance we are living in a computer simulation. His paper, found here, is not concerned with establishing this 20% figure, which he affirms is rather his personal gut feeling. It is more of a trivial exercise in logic, but he makes a number of remarks that are very problematic. Take for example. "Suppose that these simulated people are conscious (as they would be if the simulations were sufficiently fine-grained and if a certain quite widely accepted position in the philosophy of mind is correct)."
This brings up all sorts of difficulties.
Suppose I'm a Nuclear Physicist and you are a computer programmer. I approach you requesting a simulator that will include a newly theorized particle, the awesometron. Now the awesometron is a strange particle, because it participates in none of the four forces. It has no mass, no charge, it does not undergo decay. In fact, this is why we want you to simulate it, because it is entirely unobservable.
So, how do you, the savvy programmer build it into the simulation?
You will nod and smile, fake something, and collect your paycheck.
There is exactly the same problem when people talk about simulating consciousness. Because, you know, consciousness doesn't actually do anything. It just is. And this is why they tend to speak in terms of simulating the brain. Consciousness is somehow caused by the brain, so if done with sufficient attention to detail, then surely consciousness will inevitably (how could it not?) come about.
The trouble is this: when we simulate the heart with sufficient attention to detail, what inevitably comes about is simulated blood flow. Not true blood flow. When we simulate an entire human with sufficient attention to detail, what will inevitably come about is... simulated consciousness, which is nothing. (like the awesometron). What might come about is true consciousness, but to claim that true consciousness will inevitably come about, with out knowing anything at all about how it will happen, is totally without justification.
Later I'll try to put together an argument towards why we should positively expect consciousness to not come about when we simulate a brain.
