X-Message-Number: 11226
From: Thomas Donaldson <>
Subject: Mike Perry and Tipler
Date: Sun, 7 Feb 1999 22:45:19 +1100 (EST)

Hi everyone!

Some people were gracious enough to email me the entire copy of the
Cryonet messages that I did not receive. And John Clark sent me his own
message, which seems to have caused my problem -- in ascii form so
that I could read it. I want to thank all of these people.

To Kevin Smith I will point out that IF messages are sent in ascii form
as I understand they should be, there is no reason why ANYONE should run
into trouble reading them. I'm happy to work with him to find out what is
going wrong, but I was not just repeating my former statement when this
happened the last time. If Kevin wants to change the rules as to
acceptable message formats, then he has the right to do so. It's his
system, which he has very graciously provided for all of us. But I'd
still like to work out just what happened.

Finally, as to Mike Perry's answer to my comment about his use of Tipler:
Tipler is certainly an intelligent man, and I would hardly claim that
he's had nothing of value to say. I will point out, though, that merely
believing that human beings are finite state machines is hardly a sign
of originality. Lots of other people have the same opinion. (I will give
my own suggestions on this issue later in this message). So my question
remains: what has Tipler got to say, now, on the issue of our survival
in any form at all? Simply being intelligent is too broad a test to 
answer my question: why are his ideas still of interest?

And here, as promised, are my comments on the issue of whether or not
human beings are finite state machines. First, we know more about how
nervous systems, including human nervous systems, work now than we knew
before. One major observation is that existence of growth: neurons will
develop new connections which they did not formerly have. Nor is it true
that neurons will have at most one connection to any other neuron: for
some reason, the connections can be multiple. Furthermore, despite
attempts to disprove this point, there is growing evidence that we DO
grow new neurons. Not everywhere, but in some places.

And finally, for those who are interested by fractals and their relation
to nonlinear systems, our neurons respond in a nonlinear fashion to 
stimuli.

These are statements of the best available information on how our brain
works. Yes, it has been an advance to work out that we use some form of
neural net, but the exact form is not any of those yet incorporated into
computers.

NOW: whether or not human beings can only adopt a finite set of states,
given the ideas which I have just summarized, is likely to depend strongly
on whether the Universe itself has only a finite number of states
(naturally larger than the number of states of human beings in it). This
is close to but not the same as a claim that it is infinite: even if 
infinite it might still have only a finite number of states. Still, it
would be surprizing if that infinity had only a finite number of possible
states.

If the Universe takes on an infinite number of states, then human beings,
perceiving this Universe, can also take on (collectively, not
individually) an infinite number of states. The strong suggestion that 
we do grow new nervous connections and may well even grow new neurons
means that the possible future states any individual may take are
infinite.... even though at any fixed time they will be finite. And
at any fixed time, there may be only a finite number of possible states,
but if that finite number of states acts like a window on an infinite
universe, then the states in that window will vary infinitely.

Here is an example which may explain what I say in less abstract language:
suppose that we can at any time only see a finite number of colors (3,
for instance: red, green, blue), at time zero. And suppose that the colors
available for us to see are infinite in number. Then the choice of colors
we can see might vary infinitely with time: from (red,green,blue) to 
(umber,blue,mauve) to many other colors for which we have no names (as
yet). If we are finite state machines, than those finite states, to have
any meaning, would have to remain constant, at least for the period of 
a computation. If we constantly add and remove connections, and even
(more slowly) add and remove neurons, then even the wiring of our brains
may have an infinite set of possible states, even though at any fixed
time it remains finite, perhaps even the same size.

There is a second problem which I alluded to when I pointed out that our
brains worked in a nonlinear fashion. It's possible for nonlinear DEs to
produce an infinite variety of behavior, without ever returning to their
original starting point. Easy, actually. (This point is much more an
extension of what we know about neurons and brains than the previous one,
but I think still deserves consideration). Even objects in the game of
Life can do the same, filling up their "universe" with lots of different
shapes (and note that such objects individually have only a finite set
of possible behaviors).

I raise these two possibilities AS POSSIBILITIES, not as facts. My point
is not that we are not finite state machines, but that it is hardly
obvious that we are finite state machines. Anyone who claims that we are
must deal with the kind of possible behaviors I have just outlined.

As for Tipler, I too have read his books. I liked his first one on 
Anthropic ideas of cosmology, and thought he and Barrow had valuable
things to say. I doubt, however, that we will find the Universe to be
finite (except for its past) and therefore do not find the notion that we
are finite state machines clear or obvious at all.

			Best and long long life to all,


				Thomas Donaldson

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=11226