Bill Hobler points out that
1. It is necessary to discover and understand the micro level
features of systems and seek the resulting macro behavior.
2. Omitting the process of synthesizing macro behaviors from
knowledge of micro features is ignoring lessons learned centuries ago.
The temptation is to correct micro feature errors and not invest in
synthesis. This is short sighted.
Both very valid points. However his examples relate to purposive systems.
It is methodologically easy to analyse a purposive system using the
scientific method in an impersonal reductionist way because the analyst
remains outside the boundary of the system which he is observing. When it
comes to human activity (purposeful) systems reductionism often leads to
abysmal failures. In these situations the problem solving system is made
up of at least the analyst, the client, the methodology being used, and
the problem owners in the situation. Hence the analyst is part of the
system he is analysing! His mere presence effects the system in some way.
In order to examine a problematique of this nature using reductionist
approaches it is necessary to draw boundaries about relevant subsystems,
and to consider whatever lies outside a system's boundary as constituting
its environment i.e. the constraints that it takes as GIVEN and which it
does not try to change. If these boundaries are well chosen (see
Churchman) then the performance of the relevant systems can be optimised
(subject to their 'environmental' constraints) with a certain degree of
success. If the boundaries are badly drawn the result is suboptimization
at best, or worse still, the precipitation of a self- defeating system.
Soft systems engineers are therefore very wary of being reductionist, and
tend to adopt as holistic an approach as possible right from the very
start of an analysis. The methodologies that they use are designed to
permit such an approach, and are not similar to the method of science.
Keith Sandrock
KSAND@hertz.wits.mech.ac.za
Keith Sandrock Systems
FAX 27-11-339-7997
--"K SANDROCK" <KSAND@hertz.mech.wits.ac.za>
Learning-org -- An Internet Dialog on Learning Organizations For info: <rkarash@karash.com> -or- <http://world.std.com/~lo/>