On 12/17/2011 4:30 AM, Quentin Anciaux wrote:

The thought experiment is that:

1- Computationalism is true.
2- So it means there exists conscious program.
3- You just stumble accros one.
4- You run it.
5- During the run you've seen that some parts are never accessed.
6- You remove those parts.
7- You run it again... it must still implement the conscious program (3) by 
points 1 and 2.
N- you can build a machine that implements and can only run 3 but that can't handle counterfactual, but as the computation is the same as 3, it must be as conscious as when it was running on a complete physical computer. N+1- you can restore the handling of conterfactual by adding inactive piece. But If N was not conscious, adding inactive pieces shouldn't render it conscious.


Yes, that's my understanding of the argument. But I find it curious that if we substitute "intelligent" for "conscious" it no longer seems paradoxical. We readily conclude that removing the ability to handle counterfactuals makes the machine unintelligent. Yet most people think a machine should be intelligent in order to be conscious. Bruno thinks it must understand finite induction. Yet there are very many people, whom we assume we are conscious, but have never heard of finite induction (their "finite induction" register is missing). Maudlin comments that "intelligence" is dispositional and so is different from consciousness. But why isn't a computation dispositional too? If it were not then it seems that you run into the paradox of the rock that computes everything.

I think the paradox arises from neglecting the fact that intelligence, computation, and maybe consciousness are all relative to a context or environment. Intelligence is the ability to learn from interacting with the environment AND acting to some purpose in the environment.


You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to 
For more options, visit this group at 

Reply via email to