X-Message-Number: 11920
Date: Tue, 08 Jun 1999 01:37:08 -0700
From: Mike Perry <>
Subject: :Relative Consciousness, 2nd Try; Neural Nets

Thomas Donaldson, #11906, writes:
>This is to Mike Perry: Can you give me a more operational definition of
>what relative consciousness may mean? ...
> there are several different kinds of memory, and one kind
>can be destroyed (or enhanced) without affecting the others. And one
>thing that happens in multiple personality disorders is that the different
>personalities often simply don't remember one another's experiences ---
>all in the same head! Do these matters bear on relative consciousness?
>Please give me an operational definition: how do I identify relative
>consciousness? Is there another kind, absolute consciousness, or are
>they all relative? And of course, relative to what?

So I'll try again. "Relative to what" is mainly: relative to what universe
you are in; I take seriously here the idea of a multiverse with more than
one universe. (Yes, in this sense all could be said to be relative--but see
comment below on "absolute" consciousness.) Different universes do not
communicate, so there is no way one conscious being  in one universe can
converse or otherwise verify the conscious states of a being in another
universe. No, it doesn't relate much to multiple personality disorders (this
is a very different notion of "relative consciousness"!). And there is a
kind of absolute consciousness too: a reasonable consciousness in some
universe, is consciousness, period, though again we may not be able to
verify it. One issue to consider is that time may not flow in the same
direction in some alternate universe as it does in ours, yet consciousness
could persist. I wouldn't rule out a "counter-clock" world where time flows
backward relative to us, yet a being in that universe finds, naturally, that
time is "forward" for it. Apparently there are very many possibilities here;

there is some interesting discussion in Deutsch's book, *The Fabric of Reality*.

Thomas in #11907 also says:
>Second, to Mike Perry: It is easy for a digital computer to SIMULATE
>a small neural net. No one would claim that it actually is one. And of
>course when the neural net gets larger and larger, the problem of 
>simulating it becomes sufficiently hard that the best thing to do is
>simply to build one rather than try to simulate it. It is that simulation
>which makes the action of the computer symbolic rather than real.

On the web I've seen a discussion of "artificial" neural nets vs. the
biological variety, but the biological NN does not seem to be what you are
referring to here, since you say we can build it. So I'm wondering, Thomas,
what you mean here. It seems to me that, if your neural net is some parallel
device in hardware, a sequential computer could perfectly simulate it,
except possibly for a slowdown in execution speed. It could, for instance,
trigger real electrical impulses that would cause action in a real robot,
not just do these things symbolically. (And a "digital computer" could be a
parallel device anyway.) 

Mike Perry

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=11920