What are Complex Systems?
Systems are how, at a given time, we interpret some part of reality as a set of component parts that are being affected by the interactions that connect them. This rational way of analysing things, and assigning casual links, is inspired by the Newtonian vision of the universe as a mechanical system. It allows the system to be "run forward" in time, and to provide predictions - subject to the assumption that the system structure itself remains valid over that time period. However, since Darwin we have known that everything in the living world evolves and changes over time as it runs forward. This means that any predictions concerning the future of a particular system will only be correct if an evolutionary change does NOT occur. And this is NOT PREDICTABLE from within the system itself. Complex systems thinking brings together the ideas of evolution and of system as it concerns the manner in which particular system structures emerge and are transformed over time. It tells us that a system can restructure and transform either as a response to its changing environment, or as a challenge to it. Either way, we will see a co-evolution of the system and its environment over time.
But, at the lower level within the system, not only do the "components" of a given system structure change their connections and interactions, but also the components themselves can change their inner nature, as their internal "systems" may also evolve their internal structure, modifying their behaviour and their responses to local circumstances.
Figure (1). Evolution invades/creates new dimensions and attributes while forgetting others. Mechanical systems just run.
Structural change, either inside "system components" or in systems arises when some initially unimportant and minor factor encounters a positive feedback of interaction that amplifies it, until it becomes a new aspect or component of the system, thus transforming the original system structure. So, evolution occurs when an instability occurs, and some initially small perturbation grows from insignificance to play a role in a new system. These feedback loops are only possible when there are non-linearities present in the interactions and behaviours of system components. There always are. The only question is what kind of perturbations may be seeking to probe the stability of any current structure, and how frequently do they strike? Also, the capacity to evolve or transform will depend on the strength of the "defence" mechanisms of the system that may be ready to spring into action and quell any incipient disturbance. Essentially then, complex systems are systems that are capable of self-transformation over time, as the result of two simultaneous dialogues: between the emergent system description at a given time, and all the non-average, the richness and confusion that is probing its stability over time, and secondly, the dialogue between the system and its environment, which either responds to change in the system, or provides threats or opportunities to it.
This fundamentally changes the basis of reflection concerning organisations and social systems. Instead of viewing them as functional machines whose performance management aims to optimise, we see them as creative, adaptive entities that explore, experiment and learn over time, changing their goals and strategies, and transforming themselves and their environment over time. The Complex Systems Research Centre is therefore exploring the importance and implications of these new ideas for successful management and policy development in social and organisational systems.
Impacts
The impacts of complex systems thinking are fundamental. Instead of basing our strategies and actions on PREDICTION, with the development and implementation of a plan designed to take us from "here and now" to "there and then", we have to adopt a more creative approach. This implies more frequent monitoring and reassessments, with an awareness and capacity to change the targets and goals, and to make use of what is working, and reduce what does not. This is an approach that recognises the constant need to learn about what is happening, and to try to make sense of it as fast as possible. In essence it carries the idea of a "knowledge" based activity even further, by considering not only the knowledge involved in an activity, but also how that activity fits in the world that is being created and discovered.
Another impact clearly is an acceptance of change, and a general shift in favour of what is small and flexible rather then large and fixed. Obviously, there are some decisions like building a new generation of airline, or sinking oil wells that are very high capital costs and long term decisions, with the attendant dangers of changing overtaking them. But in general, in many other domains of business, commercial and urban development solutions that allow for adaptation and change are clearly favoured by these ideas. Similarly, it is important to be able to put a sufficient value on flexibility, and indeed to improve tools of our risk analysis. Most risk analysis is concerned with variations that can occur within a particular, fixed system structure. Clearly, complex systems thinking introduces the idea that a more radical uncertainty may exist which is related to the possibility of structural change occurring, and of entirely new factors coming into play.
Other impacts concern the need to explore and experiment, in order to learn in a changing word. This implies the need to permit and accept the failure of experiments as well as their success. Obviously, success tends merely to confirm our previous understanding of things, while failure allows us to learn as it points out the inadequacy of that knowledge. New knowledge is really only created by the testing of new conjectures and this automatically implies a necessary frequency of failures. Managers that never experience any failures in what they try are not really trying anything new. The importance of moving away from a culture of "blame" becomes evident. If learning is risky, and any failure leads to blame, then people will simply not try anything risky, and the organisation will become a fossil. The system will "lock in" to its current structure, and eventually death will ensue.
On the other hand, clearly, an organisation cannot become totally engaged in exploration and experiments, since it must also earn a living while learning. Therefore the skills of leaders need to become focused around what degree of "exploration" to engage in and what fraction of functional, productive and efficient behaviour to maintain.