minds and brains
johnhkm at logicworld.com.au
Wed Dec 23 22:21:40 EST 1998
>As Turing originally posed his test, it was simple engineering answer to
>philosophers. If you can't tell the machine from a human through a teletype
>interface you might as well agree that it has a mind (soul). But Turing's
>Test can be described with words and the philosophers have had a field day
>with it. Big Blue passed the test in a very restricted universe. You can't
>tell Big Blue from a man at the chessboard.
Wrong. See, Shadows of the Mind, Roger Penrose, pp 47,8 An excerpt:
"This much is obvious to any human player with a reasonable familiarity with
the rules of chess. However, when the position, with white to move, was
presented to Deep Thought - the most powerful chess computer of its day,
with a number of victories over chess grandmasters to its credit - it
immediately blundered ... "
What continually amazes me is the notion that since we cannot physically
identify a soul it must not exist. We cannot physically identify any number
of concepts and theorems (even 'numbers' for that matter), does this then
mean that they do not exist? Furthermore, our sense of self is so
overwhelming that we must operate throughout our lives as if such a self
were very real and substantial. We would not be human if we did not. We
would also be rather stupid and incapable of choice.
For decades the scientific community has been content to remove the self
from discussions about consciousness. They believe a bottom up approach will
do, but after decades of just such a strategy we are no closer. Emergent
properties! This is a cop out strategy, a refusal to deal with the problem.
At the end of this century that cop out is haunting us. If the little man
inside our heads is just an illusion then surely most of concepts are
illusory for the very greater part of thinking requires this counterpoint to
guide us in our formulations. It is the most powerful illusion we have and
everyone has it. So Mr. Dennett may rave all he likes about why the
Cartesian theatre is a myth but in so doing he completely fails to deal with
why this myth is in everyone. It must be more than a mere epiphenomenon, it
must play some essential role in cognition.
>We should leave soul (mind, self, intelligence, whatever) to the
>and proceed with the design of a machine that can think and decide using
>only neuromimes. As a first step we examine the brain to see how the
>do these things. If we can explain the how the brain thinks, how it
>we shall know how to design the machine.
Big if, and again, you are avoiding the question, you are creating excuses
not to deal with something. I agree that it may lie outside the domain of
science, but does that entitle us to just do away with it as being mythical
or illusory? If it lies outside the domain of science then scientists have
no right to question its existence or its validity as a concept.
The brain is not like a computer, it is messy, poorly designed, uses a
multitude of strategies, and is subject to so many external and internal
variables that ideas of determinism and free will become irrelevant. I do
not believe in free will but I do believe in choice. If one does not believe
in choice one's whole life is a lie.
Alan Turing once commented that if you want a computer to do mathematics
then it will have to be a device capable of making mistakes (and, by
implication, be conscious of the same). So much for algorithms etc. To
proclaim that a bottom up approach will eventually solve all these problems
is to exercise a religious faith in science. You can't have your cake and
eat it too.
Putting the cat amongst the pigeons.
More information about the Neur-sci