2 Section readings – Co-evolution
Reading 6.1: Systems thinking: the second step
It was noted in Reading 4.1 that the processes of analysis and synthesis in conventional thinking are based on the concept of an object. There the concept of systems was introduced as a set of interrelated objects. In this concept of a system it becomes important in the process of analysis to include both objects and their relationships. Other readings, in Parts 4 and 5, analyse natural systems, and investigate the understanding of behaviour resulting from synthesising objects and their relationships into a system. But this first step into systems thinking, looking at objects and their relationships, is limited in its usefulness.
During the development of systems thinking in the latter half of the last century, a number of people noticed that this object-based systems thinking did not work well when applied to human-activity systems. As a result the focus of the definition of system was changed to consider a system as an input–process–output structure, what I have called the second stage of systems thinking. This does not mean that the first definition is wrong but that it was incomplete. A system in the new definition is not characterised by attributes but by a purpose, that is if you take a snapshot at a point in time you can then contend that the system intends to produce more output from its input by a later time. The purpose encapsulates the relationship the system has with its environment, which was missing in the first definition.
The reasons for the limitations of object-based systems thinking were twofold. Firstly, in general, human-activity systems are far too complex to be able to define object components well. What objects would you include in a ‘railway system’ or ‘education system’? This meant that, without realising it, different people were thinking of different systems and so could not come to agreement on their analysis. The problem was overcome by first defining the purpose of the system being considered. What purposes could you assign to the ‘railway system’ and ‘education system’?
Secondly, in object-based systems thinking there is no reference to the environment of the system: remember the example of the lion in Reading 1.2. A lion in the zoo is not the same system as the lion in its natural savanna habitat, because the relationship between the lion and its environment is very different in the zoo to that in the natural environment. So whilst you might consider them exactly similar as objects, you cannot consider them exactly similar as systems.
Now, with the new system definition, when analysing the interior of the process boundary it is much more useful to do this in terms of subsystems, which themselves are systems, and therefore, also defined by their purpose. Here it is important not to confuse the system boundary with the process boundary. The system structure that we are now considering input–process–output does not have a clear boundary, and it cannot be differentiated from its environment in the clear manner that applies to an object. But what you can do is identify the process boundary, and this in the case of the lion is the same as the object boundary, since all lion processes take place inside the lion’s skin, which is what you think of as the object boundary in this case.
What you gain by this change is that now you can analyse any system into its subsystems, and furthermore analyse subsystems into sub-subsystems, etc., and in reverse synthesise a system from its subsystems, etc. The processes of analysis and synthesis work with systems in the same sort of way that they worked with objects in conventional thinking.