RE: consciousness based on information or computation?

From: <hal.domain.name.hidden>
Date: Fri, 29 Jan 1999 08:47:40 -0800

Jacques Bailhache, <Jacques.Bailhache.domain.name.hidden>, writes:
> Is it really a fundamental difference between static and dynamic structures
> ?
> Any machine can be represented by a string. A program can be considered
> either as a static character string, or as the description of the rules of
> evolution of a dynamic process.

I am uncomfortable with saying that a process is equivalent to the rules
which generate it. It seems to suggest that writing down the rules is
equivalent to actually working out all the ramifications of those rules.
It is true that the rules uniquely specify the process, but in practice
it still takes work to find out exactly what the details of that process
are.

In particular, if the process in question is a consciousness-containing
universe, what is actually necessary in order for that consciousness to
exist? Running the program seems good enough to me. Is just writing the
program enough? Hans Moravec takes the view that you don't even have to
write the program; the mere fact that such a program potentially exists
automatically means that the consciousness exists (and is, in fact, how
and why our own consciousness-containing universe exists).

I prefer a simpler approach to the string-process question. I would
say that a "dynamic" three-dimensional process can be represented by a
"static" four-dimensional space-time structure. This is the standard
viewpoint for relativity theory. Time is just another dimension (albeit
one with different properties than the spatial dimensions).

Another way to think of this is to imagine a two-dimensional world.
It's been demonstrated that you could, in theory, have a two-dimensional
universal computer. Then any process operating in that computer could be
represented as a three dimensional structure, where we use the third
dimension to represent time. This solid, static three-dimensional
structure then can encode any computational process.

Some versions of the many-worlds model consider the universe to be
branching at each point into multiple universes. This is harder to
capture in a static picture. You can do it but it doesn't seem as
natural. However another way to view this model is to have, instead
of a single universe which branches into many, many initially identical
universes which then differentiate themselves. These are equivalent ways
of looking at the same phenomenon. With the differentiating-universes
model we again have a simple flow of time within each universe and a very
simple representation of processes in that universe as static structures.

Hal
Received on Fri Jan 29 1999 - 09:44:38 PST