Re: The problem of cognitive closure

From: Eliezer S. Yudkowsky (
Date: Fri Mar 16 2001 - 18:41:44 MST

Mitchell Porter wrote:
> But the bare concept of monads-made-of-qualia
> is not meant to explain the details of phenomena
> (that would be the role of a specific theory
> that fleshes out the concept in a particular way),
> it's meant to explain how there can be any such
> thing as observation of a Universe to begin with.

As long as you have the *expectation* that there are specifics of how the
monad internals affect gross cognitive phenomena, then the existence of
monad internals is not closed with respect to a superintelligence -
ordinary sensory data (nondestructive human brain scans) will discover
them in due course. A computational superintelligence might conceivably
experience cognitive closure with respect to the actual specifics
(likewise humans, one expects), but would certainly not be cognitively
closed with respect to their existence (again, like humans), so this
cognitive closure wouldn't translate into global catastrophe.

-- -- -- -- --
Eliezer S. Yudkowsky
Research Fellow, Singularity Institute for Artificial Intelligence

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:36 MDT