Well 'co-stupidity' is certainly an interesting word. It seems somewhat
similar to a word or phrase that I often use, 'error-covariance', but I
prefer the latter because it carries a remedy along with it.
> "Co-stupidity" describes the collective inability of groups, communities,
> organizations and societies to see what's happening in and around them, and
> to deal effectively with what they find. ...
We are not dealing with a universal truth here -- there are a few examples
of groups, communities, organizations, and (perhaps) even societies that
have functioned well, seeing problems and dealing with them effectively.
But yes, it's mostly true, collective intelligence is much less common
that collective stupidity.
I can also agree with these statements:
> The know-how exists with which to
> dramatically improve our collective intelligence.
>
> We could build the capacity to be wise together instead of co-stupid.
Yes, of course the know-how exists. I am one of those people who clain
to have this know-how, so I have to agree it exists. Here you go:
To improve our collective intelligence we need only use
combinatorial optimization methods to match people so as to
minimize overall error covariance.
For some reason that escapes me I have an awful time getting people
to swallow that last sentence. So whenever I find someone talking
about improving our collective intelligence I look eagerly to see if
they have found some new way of explaining things.
But what I usually find is people who claim that we can improve our
collective intelligence by just trying hard enough!
Mr. Atlee starts out well:
> It seems to me that the most powerful thing we can do
> to deal with the social issues and opportunities we each care about
> is ...
(insert drum roll here)
> ...to join together to increase our society's ability
> to fruitfully deal with ALL the social issues and opportunities we face.
Sorry, you've lost me. Isn't joining together precisely the problem
to be solved? Is the difficult problem of how to work together
to be solved by working together? Hmmm.
> If we do that, then our issues will likely be handled better.
> If we don't do that, then our issues will likely continue to be mishandled.
Well, then, it looks like the issues will continue to be mishandled.
Can we only figure out how to work together productively by working
together productively? Then we're doomed.
There's a slightly better attempt embedded in an earlier paragraph:
> ... Once we are in a group or society, our collective
> intelligence or stupidity has little to do with how clever or slow we are
> individually -- and everything to do with how well our system is designed,
> how good our process is, how wisely we handle information, and how well we
> all work together. ...
I'm not sure "our system" was ever designed, I think it just grew. Perhaps
we could design a better system, but who is to do the designing? We don't
work well together, as Mr. Atlee has pointed out, so how can we succeed
in designing a better system?
The same comment applies to "how good our process is". It isn't.
But what process have we for improving our process? Not one that
works, I suppose, or we'd notice the process improving. Hands up how
many people see things improving.
Of course collective intelligence depends on how wisely we handle
information, but how does the author propose to make us wiser?
Even though I'm obviously in a disagreeable mood, I won't disagree
with the statement that our collective intelligence depends on "how
well we all work together." I think collective intelligence IS just how
well we all work together, so that sentence sure sounds like a tautology
to me, and I have to agree -- it's true.
I'm sorry for being so unkind, and my sarcasm is a bit excessive, but
there is much too much of this stuff out on the net, and I'm sick of it.
We can't solve society's problems with pep-talks about the importance
of working together or exhortations to try harder. We need some of
that "know-how" mentioned earlier. We need tools and techniques, or
in other words, social technology.
If part of the problem is
> ... the collective inability of groups, communities,
> organizations and societies to see what's happening in and around them, and
> to deal effectively with what they find. ...
then we need to address that problem directly. Here's an example, (maybe the only
example): You may have heard of the Delphi Method studied by the RAND
Corporation some decades go. That's the name of a technique, a process,
designed to help people reach good collective decisions and collectively make
various predictions and estimates about unknown things like future technology.
It was memorable (and perhaps unique) for actually addressing the
problem, but unfortunately the Delphi Method didn't actually work.
It was an interesting idea, but there was no mathematics behind it.
I have a little rule of thumb that says if there is no mathematics
behind something it can't be hi-tech and probably isn't worth much.
As it turns out, it really doesn't matter how many people you ask to
think about a problem, and getting people to submit ideas anonymously
(to encourage free discussion) doesn't seem to help much.
What does matter is who talks to whom, who works with whom, and
whether those people are prone to the same kinds of error. "Who talks
to whom" is the combinatorial part, optimization just means seeking
the best, and being "prone to the same kinds of error" is (roughly)
error-covariance.
If two people have similar error-tendencies then they shouldn't be
working together. We all have similar error-tendancies -- we are
much more similar in the ways we make mistakes than most people
realize. Nevertheless for each one of us there exists a few people
with very different tendencies -- all we need to do is find those
people and work with them.
On http://www.island.net/~dpwilson/moretech.html I put it this way:
There are people out there who share your interests but make very
different kinds of mistakes than you do. It can be enormously to
your advantage to associate or communicate with such people.
There are people out there who not only share your interests but
make almost exactly the same kinds of mistakes that you do. It can
be terribly to your disadvantage to associate or communicate with
such people.
The Internet and the World Wide Web can make it easy to find
people with the same interests as you, but currently it doesn't
provide much help in discriminating between the ones that can help
you avoid mistakes and the ones that will provoke you to make even
worse mistakes.
That's the problem in a nutshell, and the solution is actually measuring
error-covariance by getting people to answer a lot of factual questions
and cross-correlating their errors.
More about this is to be found near the bottom of my new home page,
http://www.island.net/~dpwilson/index.html which you should look at,
if only to read the rather exciting (to me) news at the top of
the page.
dpw
Douglas P. Wilson [EMAIL PROTECTED]
http://www.island.net/~dpwilson/index.html
http://www.SocialTechnology.org/index.html
