(un)predictable future LO20438

AM de Lange (amdelange@gold.up.ac.za)
Wed, 20 Jan 1999 15:09:24 +0200

Replying to LO20328 --

Dear Organlearners,

Leo Minnigh <L.D.Minnigh@library.tudelft.nl> writes:

>This is a story about a Frenchman with a long name, Venus
>and predictable and unpredictable future.

Thank you for this delightful story. Poor old
Guillaume-Joseph-Hyacinthe-Jean-Baptiste le Gentil de la Galasiere
He was fully caught up in the drama between simple systems (like
moving planets) and complex systems (like the weather).

I would like to make a few notes on your thread.

A system is simple when one parameter can be known by knowing a few
other parameters. A system is complex when many parameters need to be
known to know the remaining unknown parameter. In a simple system the
relationship between its few parameters is fixed. In a complex system
the relationship between its many parameters changes.

Many thinkers on complex systems think that complex systems and simple
systems exclude one another. The planetary system and the weather
system seems to strengthen this idea. But is not really the case. A
complex system usually have simpler subsystems to it. For example, in
the complex weather system it is possible to measure at any particular
place two of the following three and determine the third by the gas
law: pressure, density and temperature. Another example. A business
may be very complex. But a simple subsytem of it is that its income
must exceed its expenditure to remain in business.

So what makes a system complex? Many parameters need to be known
precisely to describe the behaviour of the system. This entails that
when a change happens in one parameter, it induces changes in many
other parameters. This is known as multivariation. Furthermore, the
way in which the parameters influence one another also changes. It
means that a definite change in a certain parameter will seldom
produce the same results in another parameter on different occasions.
This is known as hysteresis. Finally a complex system has layers of
complexity because of past emergences and posible future emergences.
Some parameters of one layer of complexity may not figure in other
layers of complexity.

The problem with any complex system is that we still need to know
precisely the values of all parameters but one to describe the system
fully. This requires time and brain power which is not always
possible. When two or more parameters are unknown, it is impossible to
determine with certainty the value of each unknown parameter in terms
of the values of the known parameters. Hence certain predictions about
the precise development of the system becomes impossible.

But it does not mean that uncertain (probable) predictions are also
impossible. This is where fussiness creeps into the picture. Rather
than working with a unique (point) value for each parameter, we work
with a fuzzy (interval) value. In the case of the weather system as a
complex system, for example, we do the following. When the weather in
a certain region is in its summer phase, we may exclude predictions
concerning the winter phase. (The past week in Pretoria has been high
summer. Temperatures in the afternoon is on average 34C. We have been
sweating each day. I can predict a temperature between 30C and 36C for
tomorrow. I certainly do not predict snow for tomorrow.)

By introducing fussy values in the changing relationships of a complex
system, we begin to perceive new patterns characterestic of complex
systems such as irreversibility, bifurcations, attractors,
sustainability, essences, etc. These patterns have little, if any,
meaning in simple systems. Thus our predictions of the future takes a
new shape. For example, it is possible to predict when a bifurcation
will happen and what will become of it.

Best wishes

-- 

At de Lange <amdelange@gold.up.ac.za> Snailmail: A M de Lange Gold Fields Computer Centre Faculty of Science - University of Pretoria Pretoria 0001 - Rep of South Africa

Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>