RE: the spark of creativity

From: Don Klemencic (klemencc@sgi.net)
Date: Tue Feb 29 2000 - 20:45:44 MST


I wasn't really thinking of Dawkins and memes, but of course you're right,
those are obviously related to this premise. I had thought that Bateson's
Mind and Nature, which has been an important book for me, predated Dawkins'
The Selfish Gene, but I found, after your post, that Bateson published in
1979 and Dawkins in 1976. I haven' t read Dennett's book yet, but I'll make
a point of it. Thanks for the reference. Another author that comes strongly
to mind is David Deutsche and his book Fabric of Reality. His major thrust
is to find deep synthesizing principles underlying reality. I don't recall
his explicitly endorsing this one, but something that seeks an underlying
principle for all creative processes would be something he would be
interested in.

Obviously, the important and primary question is whether it's true.

I was very much thinking of something beginning in the unconscious, with
only the final phases emerging to consciousness. The amount of "underground
work" needed in each case would be commensurate with the difficulty or
complexity of the subject. It is a commonplace that a way to deal with a
problem is to struggle with it for a while, and then back off. Revisiting it
later, things seem to clarify that were not clear earlier. I've had this
experience many times and I've read many authors referring to it. I think
the most famous, almost theatrical case was the nineteenth century German
chemist Kekule who had worked hard to understand the structure of the
benzene molecule. He reported falling asleep and dreaming of particles
flying around. They formed themselves into writhing snakes. When one of the
snakes bit his own tale, Kekule awoke immediately with the correct idea that
the benzene molecule formed a ring. But I think the massive common
experience of ideas coming together after we've worked on them and then
backed off, is more important than famous anecdotes for establishing the
importance of the unconscious for thinking.

As to whether the mechanism operating in the unconscious to produce new
ideas has the form of natural selection, the proof would be if AI
researchers discovered that they needed to build structures that embodied a
natural selective process in order to see creativity operating in their
systems.

                                                        Don Klemencic
-----Original Message-----
From: owner-extropians@extropy.com [mailto:owner-extropians@extropy.com] On
Behalf Of Dan Fabulich
Sent: Saturday, February 26, 2000 3:41 AM
To: extropians@extropy.com
Subject: Re: the spark of creativity

> There is a proposition that all creative processes have natural selection
as
> an underlying mechanism: 1. Mutation or a random information source
external
> to the system. 2. A progressive winnowing of the information for
> compatibility or usefulness. 3. Amplification and combination of the
> survivors to compete at a higher level. (I use 'natural selection' rather
> than 'Darwinian process' because Darwin's view excluded, unnecessarily I
> think, feedback from phenotype to genotype as the Lamarckian view
allowed.)

Of course, this is just Dawkins's meme theory, taken to further
conclusions than he probably would have endorsed. I believe Dennett has a
very similar theory.

Here's a fact of the matter: in writing this two-kilobyte e-mail, I did
not start with two kilobytes of more or less random drivel in my mind,
alter it character by character, at random, until I happened upon an
e-mail that satisfied me, and then type it out. The process was far more,
dare I say it, *logical* than that. (When Steven Pinker came and gave a
talk at Yale, I asked him why his book, _How the Mind Works_, didn't refer
to memes at all, and that was more or less the response that he gave;
though he gave the example of Shakespeare's _Hamlet_, which also doesn't
appear to have been spelled out via natural selection.)

You'll notice, as well, that most processes in the human body don't work
as you describe. The human body has organs and organ systems which
perform specific functions, not by acting at random and then selecting the
organ behavior that works, but by performing more or less the same
function from birth.

Steven Pinker puts a lot of weight on the "mental machinery" idea, and I
think there's a lot of truth to it. However, isn't there *some* sense in
which a natural selection of random data takes place in the brain? Sure,
we don't SPELL that way, since we have developed brain machines to take
care of that purely mechanical task for us, but perhaps we wouldn't call
it "creative" unless there was SOME element of the Lamarckian in it?
Notice as well that my spelling, in this e-mail at least, isn't all that
"creative" anyway.

While this is the area where meme theory is at its strongest, I think we
find that even here we don't have very much evidence that would support
the theory. Why? Because natural selection/winnowing needs a large
sample base before it becomes a plausible mechanism. Darwinian theory
works because evolution takes place over a Really Long Time, thus allowing
many different random possbilities to select from. It's hard to guess how
many different concepts I had in mind for this e-mail, but I find it
highly unlikely that I had more than, say, two or three for this e-mail,
and of those two or three, I can't see any structural relationship
whatsoever which would lend me to think that any of them are mutations of
one another or a single idea. (I expect that the same could be said of
_Hamlet_, though now we have no way to tell.)

You might now object that the reason it seems to be this way to me is
because most of the natural selection work has been done subconsciously;
the two or three ideas which have emerged from this may not look related,
but deep down inside, they really are. If this is the only area where we
can safely assume that natural selection occurs, then this theory is
scientifically unverifiable, since *nobody* has access to my unconscious.

The only way you COULD verify a theory like this would be by attempting to
give some account of the brain chemistry/neuron firing patterns/what have
you under which some particular observable phenomenon could
uncontroversially be taken to represent a particular idea, and then
demonstrating that there are many such ideas, bearing subtle and purely
structural differences between them. I sincerely doubt that you'll find a
theory which meets these criterion, though, of course, you're more than
welcome to try. I'm sure you'll be published for it if you succeed. ;)

(With that being said, I want to qualify my claims by saying that I DO
think meme theory is good explanation for SOME things, in that it does
explain a lot about how certain ideas move about culturally. The
criticism I bring to bear is that human thought can be accounted for
*purely* in terms of meme theory, or even just human creativity. To be
more precise, I think that focusing on memes is a good way to think about
culture, but the wrong way to think about individual human intelligence.)

-Dan

      -unless you love someone-
    -nothing else makes any sense-
           e.e. cummings



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:27:06 MST