Dear Jonas,
Thanks for your note. The formalisms are
descriptions. They don't resolve. They model.
It's what we learn through models and formalisms
that give us clues for ways to resolve some kinds
of wicked problems rather than treating them as
intractably wicked.
Complex adaptive systems include human actors and
human observers -- they may or may not involve
wicked problems. When people resolve
disagreements by negotiation among differences or
by postulated agreements, complexity is not
wicked.
It's difficult to say that I'm an optimist. One
can just as well say that I am a pessimist, but
that I see this kind of work as necessary no
matter how difficult. Finding out what doesn't
work is always important. Every time Edison
failed with the filament of his incandescent
light, he'd say, "We're one step closer. Now we
know 784 [fill in the number] things that don't
work." He went through more than 1,000 trials
before succeeding, and his career as an active
inventor involved tens of thousands of failed
trials for his several hundred successes.
It's not quite up to the level of King Lear, but
struggling with challenges more like to fail than
succeed is one of the tragic necessities of
research.
"How shall I live and work?
My life will be too short,
And every measure fail me."
-- King Lear, Act IV, Scene 7
Yours,
Ken
Wolfgang Jonas wrote:
--snip--
I agree with most of what you say. You are
probably right that Rittel spoke more about
disagreement than about complexity.
But, as soon as CAS includes human observers,
then we have to deal with disagreement (in
"politics, ethics, taste or other areas"), which
cannot be resolved by formalisms.
And then we do not have to refer to Gödel (who
demonstrates that even formal systems cannot be
described completely and uncontradictory from
inside), but we may turn to 2nd order
cybernetics, which I find more appropriate.
I think our main disagreement lies in the degree
of optimism regarding the degree of possible
asymptotic approximation to a kind of consensual
view of designing?
--snip--
|